(Topic ID: 153216)

RETROFIT Classic Bally/Stern DIY Plasma-to-LED Conversion Display Kits

By acebathound

3 years ago

Topic Stats

  • 759 posts
  • 60 Pinsiders participating
  • Latest reply 7 months ago by Wolffdp
  • Topic is favorited by 91 Pinsiders


Linked Games

Topic Gallery

There have been 217 images uploaded to this topic. (View topic image gallery).

IMG_3133[1] (resized).JPG
ballystern_led_display_single (resized).jpg
tester1 (resized).jpg
tester2 (resized).jpg
display3 (resized).jpg
display2 (resized).jpg
display1 (resized).jpg
IMG_4811 (resized).JPG
IMG_4809 (resized).JPG
20170530_141157 (resized).jpg
20170517_182622 (resized).jpg

Topic index (key posts)

22 key posts have been marked in this topic, showing the first 20

Post #1 Refurbish kit for Bally/Stern 6 and 7 digit displays. Posted by acebathound (3 years ago)

Post #362 PCB Depopulation image (Stern DA-100 Rev C) Posted by vid1900 (3 years ago)

Post #363 PCB population image (Stern DA-100 Rev C) Posted by vid1900 (3 years ago)

Post #364 PCB 5v population image (Stern DA-100 Rev C) Posted by vid1900 (3 years ago)

Post #395 SMD pads to enable comma Posted by acebathound (3 years ago)

Post #439 PCB depopulation image (Bally AS-2518-15) Posted by vid1900 (3 years ago)

Post #440 PCB population image (Bally AS-2518-15) Posted by vid1900 (3 years ago)

Post #444 PCB Depopulated (Stern DA-100) Posted by eh97ac (3 years ago)

Post #446 PCB depopulated (Stern DA-300) Posted by dothedoo (3 years ago)

Post #469 PCB depopulated (Bally AS-2518-21) Posted by Cheddar (3 years ago)

Post #505 Link to gallery of photos of sample installations Posted by acebathound (3 years ago)

Post #515 display and filter color options Posted by acebathound (3 years ago)

Post #530 install and review of steps with plenty of pictures Posted by RocketFromTombs (3 years ago)

Post #558 Promotional offer on orders Posted by acebathound (3 years ago)

Post #568 Link to all the board schematics Posted by acebathound (3 years ago)

Post #612 PCB Depopulation image (DA-300 Rev C) Posted by vid1900 (3 years ago)

Post #630 Parallel Resistor locations on BALLY AS-2518-21 6-DIGIT Displays Posted by acebathound (2 years ago)

Topic indices are generated from key posts and maintained by Pinside Editors. For more information, or to become an editor yourself read this post!

You're currently viewing posts by Pinsider acebathound.
Click here to go back to viewing the entire thread.

299 posts in this topic match your search for posts by acebathound. You are on page 1 of 2.
#1 3 years ago

These are DIY conversion kits that include a new led digit panel PCB board, led digits, resistors and a few other components that will allow you to upgrade an out-gassed 6-digit or 7-digit Classic Bally/Stern display to LED.

You'll need to supply your own donor boards (ie. old displays). So if you want 5x 6-digit displays, you'll need 5x original Bally or Stern non-working 6-digit displays to convert. Most people will find it's not *too hard* to accumulate non-working displays as project machines are bought/sold

GALLERY OF CUSTOMER IMAGES (completed displays)



  • Efficient for 30+ year old technology (similar or better than after-market led displays)
  • Aesthetically pleasing design
  • Lower cost option (for people with old out-gassed display boards)

Here's a picture of the converted displays with WHITE digits & colored filters being used (all except the credit display which is a "brand x" amber digit led display):


(test-fitted, board not modified & no led digits in it yet)


http://www.pinitech.com - "Pinball Inspired Technology"
Kits, upgrades and test equipment for pinball machines

#9 3 years ago
Quoted from CactusJack:

I guess the big question is: is there an available right angle header to make mounting easy? I wouldn't want to have to solder and then align a bunch of bare wire to mount it.
What was the pitch of the old plasma tube leads?
But it's an interesting project for us cheap bastards!

Yep, that is the issue actually. Noticed that earlier today & unfortunately not a header pitch I can find, so something creative will have to be done there. The pitch on the old plasmas I think is 3.18mm (0.125").

Here's the idea I had. After unsoldering the gas plasma display, the holes on the 7-digit Bally look big enough for regulator 3.96mm headers. So as long as other Bally/Stern 6 & 7 digit display boards are like that, my initial thought is.. for support & to make wiring up everything else easier, have at least 2x of the 3.96mm headers on the "led display" pcb. Maybe one on each end.. and solder those in first to get the display lined up & to have the support. They could even go in "dummy" connection through-holes that aren't used. That way you get the display all lined up correctly & at the right height (again some things to work out to make setting height easy)

One you have it mounted like that, if there's then small wire connections for the rest of the connections that drop down from the display into the plasma footprint's holes.. it'd be easy to identify where each wire goes as the holes would align.

There's definitely going to be some out-of-the-box thinking on this one.

#11 3 years ago
Quoted from CactusJack:

The holes are already pretty wide. I wonder if there is a metric pitch that might work in smaller chunks. With small pin diameter In smaller strips lengths, you might be able to make up for the corrections every so often.
Otherwise, your idea of alignment pins on either end might make the bare wire jumpers in between, tolerable.

I found some 0.1mm headers that I thought might work (at least for a good amount of the run), but as soon as I typed in "right angle" I came up with nothing. Might be something out there, I'll do a bit more searching. I just know that the more odd-ball stuff, it's like good luck finding it cheap

#12 3 years ago
Quoted from SealClubber:

You mad scientist you. I think interest in this would be pretty high once you figure it out. For me, it would have to look relatively clean and be cheap enough compared to the alternative to make me want to mess with a kit. Ie, a $5 difference wouldn't cut it.

Yeah, I'm with you on that.. I'd also want it to look clean and not have some huge mess of wires going all around the board. And there'd have to be incentive to going through the trouble of desoldering components off an old board, so price would be a factor. It may be best to leave it as a full kit and not even bother offering the assembled "led digit boards" since if you can desolder/repopulate the old display board, then soldering in 6-7x led digits and a few other things to a new board won't be an issue. Also, these would just have generic 0.8in digits with decimals, with the upside that they'd get the job done for less.

I'm fairly picky about spacing things. So digits will be spaced evenly, they'll line up with where the existing displays were. I'll try and have it so it's easy when assembling to get everything all lined up. If I can't make all that happen and get the price somewhere people would want it, then I won't do them.

So with that said.. what if the price was at or under $100 for a conversion kit for a set of 5x displays? Just throwing out some numbers.. but I do realize price will be a factor on these.

Update: After looking at some digit pricing, RED digits are the cheapest [seems typical of anything led related]. Other colors may increase costs a good amount. Also depending on mosfets, etc components.. costs may be a bit more than I originally expected. I'll attempt to hit something that's appealing given the nature of this type of "DIY" project, but if I can't I'll possibly figure something else out (as far as just offering boards with digits and people can source the remaining components themselves, etc).

#15 3 years ago
Quoted from SealClubber:

Sounds like a decent price. Stock replacements are like $30-40 nowadays. The bad news for you in my case is I have at least 5 more good working B/S 6 digit displays. So it might be a few games before I would need this.

Haha.. nah, not bad news at all. Can't beat having a reserve of displays in case one goes out!

#22 3 years ago

Some more progress this morning.. connected the rest of the digits. Guess all those years of fooling around with full-blown led display ideas that didn't really go anywhere paid off some lol.

Currently it's just using some transistors, but I'd like to get an idea of current draw with that setup first.. then possibly switch over to mosfets on the drivers. Part of what's making this possible too is having done load tests on Classic Bally/Stern machines recently, which allows me to gauge how the circuit's actually working & compare with what else is out there. Worst case, if it's like some of the led sets where it'd be better to have a 5A regulator on your SDB and you only have the 3A, then this switching regulator project could be an option for that.


Excuse the cruddy quality of the video.. just used my camera for this and apparently it's horrible resolution in video mode

#23 3 years ago

Always learning something Think I figured out why the "less than efficient" aftermarket displays I was testing was so power-hungry. Seems to do with transistor gain and saturation voltages. Found that out the hard way since it happened with the components I was using as well. Measured a rather large in-rush of current of several hundered mA on power-up, then it would settle down. Tracked that down and it's looking a lot better. Some more things to tweak I think, but getting closer to making this happen!

#25 3 years ago
Quoted from LyonsRonnie1:

With all due respect, and I hope you appreciate the joke

Yep how I feel enough times haha.

#27 3 years ago
Quoted from vid1900:

Many people have entire boxes of outgassed or bambooed Bally displays.
This is a great idea.

Yep, if they haven't already been tossed. I'm sure tons and tons of these boards in the last 5+ years have been thrown away because of the cost of the replacement plasma displays versus just buying a brand new led display.. and very small value the dead boards have other than for parts or for someone that stocked up on plasma displays when they were cheaper. Joe Homeowner with limited space that gets a machine to fix up, with 4x bad displays in it.. probably isn't going to hang onto too many dead displays before either giving them away or throwing them away.

It's really nice to see some other people like this idea And it's coming along.

I'm gonna try out some white led displays with the vinyl.. I was experimenting with that 4 years ago when I did this piece of art You can actually see a small piece of vinyl over the single digit in the breadboard in the 2nd photo up from the bottom. I think I was turning an amber led into orange. You're limited with color of the led segment itself and the color wheel... but going with white segments.. you can change the color to whatever the vinyl is.

#32 3 years ago

Created a project page (and waitlist!) for this over @ http://www.pinitech.com/lab.php

For those that expressed interest.. and for anyone interested that hasn't commented.. if you could get signed up on the waitlist, that'd give me a better idea for the size of the initial board order. I'll probably order 50x boards minimum, but those could go pretty quick if a few people want a full set [which I'd imagine at least a few people might]. I don't want to go too nuts with quantity because waitlist numbers still aren't anywhere near actual initial sales (and this is still relatively new as a way to gauge orders/materials) but also don't want to underestimate too much either.

Put in the number of boards you'd need per however many sets you think you'd order initially. So if you'd order 1x single display conversion kit just to try it.. that's fine, just leave the quantity as one. If you'd want a single set to have enough for a single machine (5x displays) then enter 5x.

Again, for pricing I'm targeting $20.00 per display or less (full conversion kit that includes the new "plasma replacement" digit board + components, and also components to repopulate the old display board). Will probably separate those so that the people that want to order the components to repopulate the old boards can do that. There would also be discounts for full sets (ie. 5x displays).. and possibly further discounts for multiple sets.

#33 3 years ago

Working on engineering on this thing a bit. So there's going to be 13 signal connections from the plasma footprint to the led digit pcb for a 7-digit display, 12 signal connections for a 6-digit (ie. you won't have to solder 30-40 pins for the "plasma" glass replacement). Then I want some support on the "led digit" board. As I mentioned previously, I'm thinking that could be done with 3.96mm right angle headers spaced evenly.. 2x on each side maybe. They could go in unused holes on the plasma footprint.

Another option would be using small right-angle brackets for support and using the existing screw holes on the display board. That'd be ideal as it may also allow for separation of the two boards instead of having them soldered together (depending how the other signals are brought over). Anyway, not sure where to look for brackets. Looking like I'll need one end of the right angle to be 15mm with the screw hole about 4mm from the end. The other end doesn't matter, could be any length.. just need to know where to put the hole in the PCB so I can set the height of the digits appropriately. I'd also want the bracket to support threading of a screw.. so I don't need a washer/nut on them. Anyone know where I can easily look for something like that?

#37 3 years ago
Quoted from LyonsRonnie1:

Hi Ace;
Do you have a ball park idea of how many components we'll need to replace to kit these? I'm just trying to decide if it's worth the time, etc. Are we talking replacing 18 pieces or more like 50 pieces?
Thanks for doing this!

The 7-digit Bally board I'm playing around with, there's 22 transistors and about 36 resistors that were replaced to move over to a 5v circuit with common low voltage NPN/PNP transistors. Even if it stayed at that, I'd have some technique suggestions on removing the components that would make it quite a bit easier. But I'd rather reduce the component count further. I'm sure it can be done [at least to some extent].

I'm actually wondering if I can get away with leaving the 2n5401 and MPSA42 transistors in the circuit. Those are higher voltage rated transistors, but they operate off a 5v base voltage. I'd just be replacing them with lower voltage rated transistors, so I'm not even sure they need to be replaced. Haven't tried leaving those in there yet, but it'll be part of a 2nd trial design. That would reduce the components to be replaced by over 1/3.

2N5401 - http://www.onsemi.com/pub_link/Collateral/2N5401-D.PDF
MPSA42 - http://www.onsemi.com/pub_link/Collateral/MPSA42-D.PDF

The ~36x resistors being replaced.. 7x are getting replaced with 0 ohm resistors and could just be jumpered over instead of removed. There's another 15x you could solder a resistor in-parallel to the current resistor and get to the value needed, instead of removing the original resistors.. it'd be another option that could save some time.

This has been kind of an iterative process. First step was to see if I could use the existing footprints and not have too many hacks to get the component board converted over to something that could drive an led board connected to the plasma footprint. Then there's reducing component count on what needs to be replaced and/or removed. Then there's the design/engineering for making the led digit panel (plasma glass replacement) as easy as possible to wire up & get at the correct height, as well as giving it support... and ultimately making the whole thing look nice.

Open to suggestions on any of this. I'll likely be trying with leaving the higher voltage transistors in-circuit on another board early this week, but if someone knows why that wouldn't be a good idea.. or wouldn't work, let me know.. it'd save me some time

#38 3 years ago

Soo.. don't want to get hopes up too much... eh who am I kidding

I left the original transistors in there, and aside from what I think will amount to tweaking transistor gain.. it's working! At first a few of the digits weren't as bright as others, but with a tweak to the base resistor on the 2N5401 digit drivers, they now look better. Still a very slight difference, so I'll keep at it!

There was no big in-rush current when I started this up in the game, stayed right around 1.6A that I found was typical load for the boards in Nitro Groundshaker. Looks like it might fall somewhere in about 70-85mA measurable current usage when in the digit tests. I'll have to verify that. Some other aftermarket displays were 36mA, while the worst case one was 228mA!!

And as for limiting the number of modifications -- I cut one zener diode off the board and unsoldered 7x resistors to replace those with higher value (couldn't get there by putting resistors in parallel since they were too low a value). There were then 7 jumper wires on the back of the board and 14 resistors to solder in parallel to other resistors on the back of the board. That's it! This greatly simplifies the conversion process. It's almost too easy


#39 3 years ago

Pictures posted!

I didn't think it'd be possible to convert one of these without a mess of wires or some kind hacks if using existing footprints on the component board. I was happy enough getting there. Now, with minimizing what needs to be unsoldered with through-holes cleared to 7x resistors, and then having the ability to just jumper and solder parallel resistors in on the rest of the stuff.. that's a game changer. It would allow just about anyone with soldering ability an easy way to do the modifications to the component board, with minimal tools (ie. don't necessarily *NEED* a Hakko then). The rest is connecting the led digit panel.. which I'll make as straight-forward as possible.

EPIC in my eyes.. I hope it stays this way! Like I said, don't want to give *too much hope* here, but so far.. so good!

#40 3 years ago

Might be splitting the difference some in what gets replaced on the component boards..

I've been playing around with this 2nd way of converting the boards, with leaving more of the original components & although it was working fine there were a few digits that weren't as bright. Attempts to correct this proved futile.. even tried narrowing down the issue by cutting a few of the transistor legs and piping everything before the digit driver over to a different digit driver, and as far as I can tell.. it was just the digit drivers themselves. So either variances in the transistors causing a slightly different gain or slightly fatigued (if that can even happen) or possibly just that using discrete transistors as current drivers with multiplexing isn't a perfect science as far as uniform current use among digits (whereas it may be more-so if using ICs like ULN, 74HCT244, etc). One other interesting thing was some of the digit drivers had F585 markings (Fairchild?) while others had M585 (Motorola?). Unfortunately the couple of digits that were noticeably dimmer happened on both markings, so couldn't blame it just on different manufacturers.

After replacing the 7x digit drivers.. current usage per digit are more in-line with one another and my eyes aren't picking up any differences either.

#41 3 years ago

How's this look to everyone for a 7-digit design?


I think this should allow for some versatility in mounting. First of all, there's holes for a right-angle bracket for support, the other end of the bracket would align with the pre-existing holes on an original Bally/Stern component board. It looks like it'd set the display depth to about exactly where the plasma glass sits (considering the thickness of led digits I'm using to gauge).

There's then a set of smaller headers at 2.54mm pitch that you could use 0.1" right-angle header pins on, and normal 0.1" Molex connectors (other end of the wires would get soldered directly to the component board's plasma footprint positions). This would allow removal of the led digit display panel from the component board. I doubt it's a huge concern for too many people, but it doesn't cost anything to throw an extra footprint on the board & allow for the option.

Then there's the 3.18mm pitch header that matches the plasma header on the component board. This would allow for easier wiring since the footprints would match up. I removed any unused/extra holes on the footprint. Small wires could be used to wire between the component board & display panel, or I'm pretty sure you could also use SINGLE 3.96mm right-angle headers (at least on Bally boards whose plasma through-holes are large enough to fit a 3.96mm pin) and still have the display sit at the correct depth. I may add some or all of the unused plasma holes back in in case people want the ability to use them for additional support. Anyway, this should all give plenty of options on mounting that suits your needs/budget. LMK what you think!

#42 3 years ago

And here's the 6-digit..


One other thing to note... on the header that matches with the plasma one.. I left duplicate connections on there (ie. each plasma display had 2x pins for the digits.. not sure if that was due to current use or what). Anyway, figured leaving them there will be good as an extra precaution if a solder pad is "iffy" and you still want to wire 1-to-1 (having the small wires connecting both boards just go straight out and down to the plasma footprint). Again, doesn't cost anything.

But.. there will be 15x connections to be made for a 7-digit and 13x connections for a 6-digit.. to connect the display panel between both boards. Any of the additional are just extra connection points for the digits enable lines.

#45 3 years ago
Quoted from CactusJack:

I think you should put the unused connector pads back in. If there is an alignment issue to the matting holes in the master display board, it is much easier to clip the un-needed pins then to have to pull them from headers or separate into single pins (of which I see a few). You can always put "X's" above the unused positions to indicate what to cut or not populate with a direct wire feed.
Lookin' Good!

Ah, but the only way I see any header being used is to create something custom..

One possible option is to use 1x2 (ie. single position, dual-row) right angle 2.54mm headers. They'd get soldered in vertically with 3.18mm pitch between them (picture the plasma footprint in the boards I posted above, with another of the same footprint below it spaced 2.54mm apart vertically). This would create the custom 3.18mm pitch needed to mate both boards. Having the 2.54mm headers initially as 2x vertical pins would eliminate the usual issues with trying to align a single pin at the correct angle. Once soldered in, you'd have to cut off one pin from each 2pos header (effectively creating a single row of pins needed for mating with the plasma footprint). So you'd have 13x or 15x of these 1x2 headers to solder in, but that seems like it'd go relatively quick.. and it'd then eliminate running separate wires/harness for signals. The boards could then just be mated & soldered together. The only thing is, the led digits wouldn't sit as far back (3.96mm right angle pins push them back further than 2.54mm pins) and the right-angle bracket for additional support probably couldn't be used. So as far as support, I'm not sure if 13-15x soldered 2.54mm connections is adequate enough.

Another option would be a narrow adapter board. The adapter board would again use 2pos 2.54mm headers at 3.18mm pitch between them for alignment, but this time they would be straight headers. Depending on 6-digit or 7-digit, 13-15x of those 2pos headers are soldered in, you again cut off one side to create as single row for mating. This small board gets plugged into the Bally/Stern plasma footprint. On the reverse side is a "normal" connector foorprint (or two). So then the digit panel plugs or unplugs with normal connectors on it. I'd think the bracket could still be used for support. That one would be even more of an exact science though in design. It'd also add a bit to costs with having another pcb as an adapter. That said, it could be an "optional" way of doing things.. and an optional cost.

This is all as a 2nd option instead of running individual wires. I guess I figured 13-15x individual wires per board isn't too bad. They'd be maybe 1" long if you were mating 1 with the plasma footprints.. or 3-4" if going to the alternate 2.54mm footprint that would allow for a disconnect.

#47 3 years ago
Quoted from CactusJack:

Okay, My mistake. When I read the word "Header" I just assumed you have located some strip connector that would work. As opposed to saying "row of pads".
In reality, if you can provide the right angle metal bracket to secure the LED board to the master, I see no problem simply running the necessary wires to do the interconnect, especially if any redundant connections to the old plasma are omitted. Once mechanically joined, I see it being a simple process to lay the display flat forming a Tee-Pee and then just feed bare wire off a roll and solder both points, clip and move to the next set of pads.
I don't feel a connector is necessary as it just introduces a new problem point for a bad connection down the road.
I like what I am seeing and glad you are doing all the Work!

Cool Yeah that's very true with the Tee-Pee setup as you described. Push the wires through and solder them in without flipping the board around any more. Should make it pretty painless.

#52 3 years ago
Quoted from gianfri:

You were thinking of 100$ for a 5 displays kit, I don't know if with orange or red led displays.
You can find new led displays with orange leds from 30€/piece already assembled and ready to use.
As someone said before "Why would I do all this work to save few $$" ???
The new ones have also brightness regulation and low power consumption.
I think therefore that in order to gather market interest, the price of your kit should be much lower and justify the work needed and the missing new features.
Please take this post as my honest opinion and suggestion to you and not as a try to criticize your work.
Good luck for all your work

I'll let others be the judge of the pricing and whether it's worth their time. Personally, I think $90 for a set of conversion kits for 5x displays (in RED) or around $95-100 (in WHITE, that gives the ability of using led gels) is pretty fair. Your Zaccaria display conversion kits that you linked are 29 EUR (about $31.75) and I see you're selling those for 25 EUR (about $27ea) on eBay. How is $20 USD (or $18-19ea if buying a set) not low enough to gain market interest? Much lower and there's not much profit at all to go through the trouble of doing this kind of thing.

As far as whether people want to spend the time.. there seems to be a good amount of interest in this thread. I dunno, I like the idea of not throwing out the old boards. This was a neat project and it's cool to see some other people liked the idea too.

On power consumption.. I've been testing a lot and also doing general load testing on machines, and this would "play nice" with the 5v regulator. Maybe not as much as some of the aftermarket displays out there, but certainly much better than one of the "less efficient" ones. Most machines probably use 1.6-1.8A. Even the 3A regulator seems to have enough headroom to support another 1-1.5A, especially with it being multiplexed.

#55 3 years ago

I think there's just some concern over competition with people doing similar things. And that's understandable, but I'm not sure the comments against a conversion display favor a full-blown kit (more on that below). I think there's probably much more of a market [in the US anyway] for Bally conversion displays than there would be for Zaccaria. And maybe price is a factor there too as it looks like the conversion Zaccaria display kits are about $27 USD at their cheapest, whereas a full-blown kit is $48. Perhaps if the conversion displays were $20, then compared to the other option of $48 full-blown displays they would get more attention.

Anyway, I'll address some of the other thoughts..

1. Cheaper Assembled Displays. If there's $30ea assembled displays somewhere as mentioned, I'd love to see a link. But this would make any of the current full-blown display kits out there not worth selling too.

2. Cost. From what I've seen, even at $20 per conversion display -- that's still cheaper than anything else out there. This makes up for lacking a few features I think. And it would very likely be less than $20 if a full set was purchased (ie. $90-95 for a set of 5x which puts the per display cost at about $18-19ea). I did have thoughts of offering these several different ways too.. if people wanted to source components themselves. I figure people wouldn't mind having options both ways though, just like I'm sure these kits won't be for *everyone* as some people would rather shiny new boards for the entire display or something already assembled & completely plug-and-play.

3. Build time. Obviously it depends on experience and tools, but I don't see labor/time involved in a conversion kit being any more than with a full-blown kit that would likely take 45-60min build time per display. I actually think a conversion kit could go pretty fast after someone did a few of them.. probably 20-30 minutes each. Maybe closer to 1hr if you wanted to completely remove components instead of soldering many of the resistors in parallel on the back of the board.

4. Aesthetics. The display would look as-nice as other kits out there when installed in a machine. Most people aren't looking at the component boards and the component board would still look nice enough as well. On the back, worst case there's a jumper wire and some resistors soldered in parallel. The drawback as with most DIY displays is no commas (or custom digits). Plenty of people seem fine with no commas though if the price is right.

5. No Dimming / PWM. Yeah, that's a feature this wouldn't have. But if someone wants to adjust the brightness they could swap out resistors on the segments easy enough.. or even use 1pos SIP sockets on those resistors for the first display and swap resistors around until they found what brightness worked. I'm pretty sure even on displays that have a dimmer, most people are just going to set the level initially when installed & it's never going to be touched again. It's a nice "feature" but to me kind of unnecessary after the initial install. These are DIY, so it's easy enough for someone to swap out resistor values if they want to achieve a different brightness.. and once you find what works you could use that as the model for other displays being built.

6. Power Consumption. I've seen mentioned *over and over* how led displays could tax the 3A 5v regulator. Instead of guessing what load was being added, I did both analysis on the machine load without led displays & then again with a few aftermarket led displays added. Only one of the aftermarket displays hit the regulator hard (about 220-230mA additional load per display added). I used these tests while working with this conversion display idea & it looks like they'll be about 80-95mA measurable load per display in the worst-case of all 8s held on the display. I need to verify that again, but that's considerably less than 220mA from the "less efficient" aftermarket display that has been sold & used in many games. None of this is an argument against adding power saving components (ie. PWM, regulator) on full led display kits.. but there are plenty without them and the machines can support it because of multiplexing & the head-room even on a 3A 5v regulator. At least from what I have seen with the load, there appears to be plenty of head-room.

7. Savings/Economics. People seem okay with saving $35-40 by buying a set of displays as a kit over assembled displays that have custom digits. I'm a bit surprised by that, but everyone likes something different & saving money is still saving money if you don't mind putting in the time. Saving near $100 by converting old displays (while also having the positive benefit of helping the environment) should then be all the more appealing.


I hope that addresses some of the things mentioned. It really isn't a big deal even if there wasn't interest in this any more than for my own use, but it's looking like there's enough interest to justify having some of these made up and not having to worry too much about getting "stuck" with materials.

For me, $150 for kits is too close to $185-195 for an assembled display that's plug-and-play and uses custom digits, not enough to motivate me to go the kit route. I'd rather save the time and buy assembled. A $90-95 conversion display set though.. if the labor isn't too bad and the end result is nice -- now that has my interest. And the reason why is that many of these older Bally/Stern games are sub-$1000 machines. It's easier spending $200 for a new display on the later System 11, WPC, etc games when the games are $1500-2500+. But putting an additional $200 into a $600-800 Bally/Stern, that's too high a percentage I think. That doesn't really make sense to me, unless it's a long-term keeper or you only have a few machines. But a way to upgrade to LED displays for $90-95 per machine.. now you've got my attention. Now the numbers work out (with respect to machine value versus display cost).

#56 3 years ago

I was just checking out the Zaccaria conversion display pictures a bit more. It's actually quite remarkable to have arrived at a similar design on the one conversion display that requires connection via a plasma footprint. I didn't even notice that design until now, and that's the honest truth. But it looks like there was the same idea -- where recreating the plasma footprint 3.18 pitch footprint & having it so you'd use jumper wires between the displays. Then having screw-holes to allow for brackets to be used between the boards, that would both align the plasma footprints & add support. It's just interesting now that I see it.. how similar the thought was there.

I have had things like that happen before -- like my Bally "bench led display" where I thought I was doing something totally different with having a jumpers & momentary buttons to switch between the different players & credit/match. A year later I tripped across an old project page from Leon where he used a big rotary switch for something like that. It's really weird when that happens

#58 3 years ago
Quoted from dothedoo:

Oooh I'm totally in for a set of white! I want to try some different colored gels.

Yeah I'm really leaning on just going with WHITE / RED digits on these, unless for some reason the led gel idea doesn't work out. RED because they're significantly cheaper than any of the other colors and could be reflected in the display set price. WHITE because it would be the most versatile.. and you could get led gels to arrive to an "AMBER" that was to your liking -- or create any other color. Since white is similar in cost to other colors, it makes more sense just to limit to the cheapest option & the most versatile option.

#61 3 years ago

Some additional thoughts on the conversions (and later, getting a bit technical for the people that want to know). As far as replacing transistors. My feeling right now is that some people may be able to leave all the original digit driver transistors in the board without an issue, some people may need to replace one or more of those if they notice that a few digits are dimmer than others. I think probably the way to go is to do the conversion without replacing the digit drivers, try the display out & then swap out any transistors that are noticeably more dim -- or swap all 7x out if necessary. Ultimately I have to see how the first few sets of these I convert go.. but I don't see it 100% necessary to replace those transistors, only if needed.

I've been working to reduce power consumption on the converted display as much as possible, with consideration of the limits in the existing footprints on the board. I'm pretty sure where things are at right now, there's maybe 100mA usage per display, if that. I've had it read anywhere from 75-95mA in the machine, with all 8s being held. I need to verify that again.. and then once again after converting more of original displays over.. but that's basically where I think things will end up. I don't think that's too bad, all things considered.. and when compared to the 220mA per display that the more power hungry aftermarket display is utilizing.

As far as reducing power consumption a bit further, there's about 24 mA being used by the segment NPN's (2-3mA each) between the 4543 and base of the NPN -- since there's no current limiting resistor there. A 10k resistor between the 4543 segment lines and NPN base would remove most of that additional load. So you'd then likely be at about 50mA-70mA measurable current per display. I'm not sure it's entirely needed.. I like the idea of a resistor there, but it would mean cutting the NPN base leg.. clearing its through-hole and soldering a resistor to the NPN base & dropping the other end of the resistor in the through-hole. It's an option though if someone wants to go through the trouble. But the original design just had the 4543 tied directly to the NPN base.. and seems like it was wasting some power on its own (I measured 13mA on a 6-digit and 17mA on a 7-digit) -- with signal lines "faked" at the bench (ie. 4543 segment lines active) on a plasma display.

#62 3 years ago
Quoted from gianfri:

As I told you, but I will not publish any link to commercial websites unless you want it on private message, you can find new Bally Led Orange fully assembled and with brightness regulation displays as of 27,9€/piece and a full 5 displays set for 135€ (already assembled).

I don't mind you posting a link to those? I'd rather know about cheaper assembled display options. Go right ahead and post a link -- if this type of project isn't worth doing because there's already very cheap assembled displays out there.. I'd rather know sooner than later.

Quoted from gianfri:

I cannot tell you more about unassembled kit but I'm sure you will agree that with the unassembled ones you'd get closer even if not as cheap as the modification kits.

I do cost analysis as I'm sure anyone designing things would. I see at least 30-40% if not more cost in materials with doing full display kits. I would think there is *a little* wiggle room in pricing there too (ie. a full display kit *might* still have some profit at $25 per) but at some point it becomes not worth the trouble. A conversion kit has more wiggle room since it drops the cost of the additional PCB & components. Still, the led digits are the more expensive part, as I'm sure you're aware. A large part of the reason I didn't do display kits earlier was because I figured for people to be interested the price would have to be $130 or less for a set. Apparently I misjudged the market there, as it seems people are perfectly happy saving just $30-40 over the cost of an assembled display by buying the displays as a kit.

Quoted from gianfri:

I'm not going to context your estimation in time, but for doing the Zaccaria ones, if you add resistors, jumpers, led to solder on PCB etc other small things you need between 45min and 1h. If you add on top of that that on Bally you also need to solder all the wirings from the PCB to the main board (I haven't understood if you also need to replace transistors) I don't think you're going for 20 mins.

I see 20-30 minutes best-case for people with experience. I assemble a lot of boards, I'm pretty sure I could hit that easily. You'd have to have the right tools & technique down.. but I'd imagine most people that bought one of these would find they become more efficient after the first few and get closer to a 30min assembly. Again though, it's going to be dependent on tools & experience. I'll have a better idea of timeframe when I get boards in and do some conversions. If I find issues or if there's no way even after doing a handful of displays to hit a 20-30 minute mark, I'll update that thought. I really don't see 1hr though. Someone without proper tools or limited experience, sure.. I could see 1hr. But there's people that built some of my other kits (ie. 64 switch tester) in 2hrs.. and others that built it in 45-60 minutes. It just depends on experience and technique. I'll be making recommendations for technique to make assembly quicker/easier when possible as I usually do with kits.

Quoted from gianfri:

For the power, when I tested it on my prototype, I had a bit less than 100mA per display, which is way more lower than 200 or 300mA. I'll test again with the new Release Canddate and let you know. Here you have a challenge.

I'm at around 100mA. I think you could probably reduce the power consumption a bit further, especially with a regulator & PWM. 100mA actually seems pretty high for having PWM and a regulator on the board if that's measurable current usage while in a machine & not with one digit locked on (with signals faked) at the bench -- ie. actual current without multiplexing. These conversion displays are nowhere near 200-300mA measurable current while in a machine -- closer to 100mA if that. The reason the "less-than-efficient" aftermarket display I keep bringing up was 220mA current usage was because of improper transistor gain, near as I can tell. When I first played around with components on these conversion displays, I was hitting near 200mA as well, until I fixed the transistor gain issue. Now it's measuring 100mA or less with all 8s held on the displays.. while in the machine, with full 5v getting to all the boards. I think that's pretty decent, considering there's no on-board PWM or regulator & I don't see 500mA additional measurable load (due to multiplexing) for 5x displays or 1A "calculated load" hitting the regulator being that big of a deal, especially considering there have been worst offenders on the market for YEARS and I've seen very few people mention having problems even with those.

Quoted from gianfri:

Btw, go haed, I don't want to stop you from doing this, I just want to warn you about some weak points I have learned from my direct experience.

I don't mind the feedback/opinions. It's all good Sometimes more than one person works on something similar. I've wanted to do some kind of full display replacements for a while. And this was a challenge finding both a sleek way to easily convert an old display board, as well as having it look nice, as well as being conscientious of power usage. It was a fun project, and as I've said.. even if for some reason it worked out that it was only going to get used in my own machines, I'm happy enough.

#63 3 years ago

I've obviously been a lot more vocal about all the technical aspects of what I'm doing here. It's just how I am, because to me.. these are neat little projects, challenges many times.. and it's cool pulling off something that works out slick. Plus it's nice educating people a bit as I learn things too. I'm not the smartest guy doing this stuff and I don't claim to be. But if I want to pull something off and have it look nice and function well, I'll find a way. Many people wouldn't have started a thread to judge interest and then started writing about the technical aspects of solving a solution like this.. but I like hearing what other people think, being able to gauge interest some on an idea & most importantly.. possibly having someone think of something I didn't to ultimately create a better design.

In regard to this link -- http://home.scarlet.be/~cv006274/ledbal/eled.htm I actually just found that again late last week *after* I had figured out I could get away without removing so many components. Would have saved me time if I found it earlier lol, instead of working through the schematic myself. 3-5 years ago I'd probably have followed the build process in those designs verbatim & thought nothing of it, but doing a conversion as outlined there adds a tremendous amount of current on the displays (and in-rush current too when powering on a machine) because of not handling transistor gain properly on the digit drivers. I think near 200mA per display measurable in the machine, and likely MUCH HIGHER actual "calculated" load if you were to just follow those steps. With some small tweaks to the components to set transistor gain, you can significantly reduce current usage. That's been part of what I've worked on.

Anyway, I don't mind any opinions, no harm done. We're just two people solving a similar solution (which is ultimately attempting to offer a lower cost way of upgrading to led displays in these machines). Really the only time opinions bother me is when someone is attacking a product or has blatantly copied someone else's work. That's not the case here. But, I do want to say.. if there was any thought that there was any copying done with this display digit panel design, then I'm sorry to give that impression. I never saw those conversion displays with the plasma footprint that you linked. It really is quite amazing how two people arrived at a very similar solution to solve the type of displays that just have a 3.18mm pitch plasma footprint for a connection. I'm not a fan of the way the board is "cut out" in order to use 2.54mm headers on it, I think wires running to the plasma footprint would be much easier and look nicer. Anyway, the idea was similar enough.. so it goes to show that logically it made sense.

#64 3 years ago
Quoted from gianfri:

As I told you, but I will not publish any link to commercial websites unless you want it on private message, you can find new Bally Led Orange fully assembled and with brightness regulation displays as of 27,9€/piece and a full 5 displays set for 135€ (already assembled).

Ok I found possibly an example of what you're referring to, not quite *as cheap* .. yet these don't have brightness regulation & are a bit power-hungry (I'll leave it at that).

On this page:

PinLED 5 x Bally 6 digit Displays - 139 EUR (about $152 USD)
PinLED 4 x Bally 6 digit, 1 x 7 digit - 145 EUR (about $160 USD)

The thing is, if they aren't available from a US distributor, then there's tremendous shipping costs from EUR to import them. Probably near $30, which puts it at about the same place as what we can get inexpensive aftermarket led display sets for here. So unless there's a US distributor selling a full assembled display set for $150 USD, then there aren't any lower cost options. Only if you ignore shipping costs.

Maybe you can provide a link to the ones with brightness regulation. Or PM me with it and I'll post it even, I really don't mind someone posting a link like that & getting the facts out there. I haven't ordered any materials yet and if there's some "deal-breaker" for doing this project, as I said.. I'd rather know about it sooner than later.

#65 3 years ago

Ah here we go..

139 EUR / $153 USD (4x 7-digit, 1x 6-digit)

129 EUR / $142 USD (5x 6-digit)

29.9 EUR / $33 USD (7-digit)

27.9 EUR / $31 USD (6-digit)

See, that's all that was needed. If someone's going to throw a price out there of much cheaper assembled boards being available from somewhere, then might as well post links and get all the facts out in the open.

For *some reason* the shopping cart EUR price isn't matching the item page price. I chose the 7-digit set and it's showing as 116 EUR instead of 139 EUR (so less $$) and it looks like other products are doing that too. Not sure if there's some kind of accidental discount or what.. but anyway, at the very top of the page the "shopping bag" amount is correct. Even based on a 116 EUR price (which might not happen as checkout is completed), it looks like it'd be 39 EUR shipping to the United States. So 155.81 EUR total ($171.52 at today's rate). Considering the shopping cart is probably showing an error, it should probably actually be 178 EUR (139 EUR + 39 EUR shipping) or $195 USD at today's rate.

Those do look nice with custom digits & having brightness control, etc. Looks like someone may have contracted a Chinese company to add the extra part for a comma, as the displays otherwise look like some of the common LED digits with thicker segments that can be imported. Either that or someone had better luck than I ever had in finding those smaller digits with commas. I've seen larger digits with the commas, but not the size needed for pinball displays.

If those were available in the US assembled for that price I'm sure they'd draw some interest. Yet, considering the cost of Rottendog being maybe $30 more, it's probably not worth anyone importing them because any cost savings per display would likely be erased in shipping expenses and import charges... even for the wholesalers. They'd save a few bucks per display, but that would easily get eaten up by the costs of shipping & import duties. It's actually a bit surprising someone is selling those for that price in the EU even, unless they worked out a higher wholesale cost.

I'm still not seeing any reason why conversion display kits wouldn't look attractive to people. Even in a scenario of $150 assembled displays being available at some point in the US, saving $60 would still be worth the time for some. If there's one thing I've learned, it's not to try and base what other people might think solely on your own opinion of what is and isn't worth it. Time and time again I've been proven wrong in what people might feel is or isn't worth their time. It's actually a bit humbling sometimes, as things I've created that I think people should be thrilled with the price & time saved to diagnose issues.. sometimes have very little interest

#67 3 years ago

It's surprising Zaccaria changed as many things as they did through the generations of machines (even just judging from connectors and displays). The board on the Zaccaria looked like it was snipped along the through-holes for the plasma display. I didn't realize that's how it was from factory. That seems odd!

I'm not sure if the display market is saturated? I think there's probably people that have been waiting for some way to achieve displays for under $100. There's also a market for complete kits. And also a market for already assembled plug-and-play (for *many many* people that don't have soldering experience or can't be bothered).

I don't expect to sell hundreds or thousands of kits -- it really doesn't matter much to me as long as I'm not sitting on things that never sell. I have plenty of things that don't sell fast I'm just happy enough if I buy say 100x boards initially and make the investment of digits for those boards... and enough of them sell to recoup most of the investment. 20x sets shouldn't be too hard to find a home for, and I'll be using some sets myself. Beyond that, it really doesn't matter. If they sell out, I'll order some more pcbs / digits and keep it going. Worst case, I can always just sell any bare boards I have left off or use them myself. Just having some fun here too

#69 3 years ago
Quoted from gianfri:

I have just tested and the new led display is between 40 and 70mA, depending if full brightness or not, with all led on and at 8 position which I think is very good.

Yep, that sounds much better!

Not sure how to do the email notifications..

#70 3 years ago

I'm going to finalize the led digit pcbs & get them sent off early this coming week. I'll probably just order 50x or so boards each of the 6-digit and 7-digit. And place some led digit & component orders as well. If all works out, I may have some boards in about 3 weeks. Here's hoping it works out the first time

#72 3 years ago
Quoted from gianfri:

You're quick!
One question, it looks ike you're using 0.8" 20mm width led digits, right?
Have you tried the 18.6mm width? They should better fit together and respect the 7 digits aspect. (6 digits do not have this problem because you have enough space in between)

Yes the current design has silkscreen outline of the 0.8" 20mm, but I was looking at both widths and considering going with the narrower 18.6mm. I'm going to mock something up and see what each option would look like (printed out, cardboard stand-up inside a machine probably haha. The narrower width probably would be a better choice. Plus I already think I have some other digits in that pinout.. and could use some of those up on my own displays.

I really want to put led displays in my shuffle alley and if I don't get this order in soon.. I'm back to working on my house and other non-pinball projects -- and design time is very limited. So that's the motivation

#74 3 years ago
Quoted from gianfri:

I fully understand you and I know how much time and money I spent in testing led digits samples before finding the good compromise price/aspect wise.

Yep, I actually got some quotes from a US led manufacturer back in 2011.. to see what custom digits would cost. I used Photoshop to mock up the digit.. and scanned in (via copier) a bunch of old plasma displays, to get the segment width and angle right. I had a pretty nice looking digit created Then found out tooling charges would be $4,000.. and would have to buy a minimum of 1000x digits (but without additional tooling charges after that). Anyway, custom digits are nice.. but expensive. I'm really surprised after all of this time there aren't 0.8 digits with commas available. I tried getting a quote on that [today actually] and told them I was fine with everything except the decimal.. just wanted that to be a comma. Was told $800 tooling... then a minimum order of 60,000 digits. Not gonna happen lol.

#75 3 years ago

Hah.. here's what I had sent the led manufacturer in 2011 (just a small sample since it's a bit off-topic).


I think it'd have matched better to the originals than a lot of displays out there at the time. After scanning a bunch of the plasma displays in, I chose the version of the plasma font that I liked the most at the time (different manufacturers had different looking digits).

Generic digits are fine though, considering what it would cost to have custom digits made up and small volume of sales. Just surprising there aren't already some sources for 0.8 digits with commas after all these years.

#77 3 years ago
Quoted from ForceFlow:

Is there a way to find out if a company already has the tooling on file to produce digit displays with a comma? Obviously, it's been done by some of them before.

Not that I know, needle in a haystack kind of thing. Maybe you contact some overseas manufacturer that already has it, but they might tell you it's custom anyway and try to get you for the tooling charges. Most of the companies have big enough customers that someone trying to buy a few thousand digits is a blip on their radar. They'd rather something like that 60,000 unit order.

I've contacted multiple sellers over the years for obscure things.. you can burn a lot of time that way. One of the other things I drove myself nuts with was a source for TIN 2.54mm round machine-pin male headers. You can get GOLD no problem.. and you can even get the mating female socket strips in GOLD or TIN no problem. But good luck tracking down reasonably priced round TIN male headers. Usually if you do find someone selling them, they're *very* expensive in comparison to the GOLD. A few times I thought I finally found some and then saw a $5-10.00 price for a single strip of 30 pins. I can't imagine who would buy them at that price.

#79 3 years ago

Ugh.. a reasonably priced supplier for the white led digits pointed me to some datasheet info on the WHITE displays and they read 470nm (Super Blue) according to the color chart here - http://www.oksolar.com/led/led_color_chart.htm So I questioned it and he was basically like "good catch" and mentioned there's a "coating film" on the segment surface and they display white when the blue light passes through that. What the heck? lol. If that's true I wonder how the led gels might be affected by something like that. It just seems odd they'd even go through the trouble of using blue leds and converting the light. I'm wondering if it's actually the case or just a way to explain an error on the datasheet. Anyway, he assured me segments light up white.

Now I'm trying to work out how they'd do that by learning about additive/subtractive color mixing:

Seems like if there's ultimately white light coming out, that's what any additional filters would work off of. But I have yet to find another white led digit say it's something like 470nm.. most are somewhere in the expected WHITE wavelength (something like 5500K nm).

#82 3 years ago
Quoted from vid1900:

White LEDs can be filtered to any color.
So you could have red, orange, green, yellow - whatever fits the theme of the game.

Yep including matching original plasma color.

#83 3 years ago

Been playing around with the design some.. here's some updates to it so far.


-Moved plasma footprint closer to the bottom of the board so if using bare wire (as CactusJack mentioned) there's less of a chance for the wires to be able to move into each other. I like that approach and think it'll be the easier method (versus using stranded insulated wire -- though that could probably also be used). Using bare wire or solid core insulated that is stripped back an inch or so each time you can solder in place, bend at 90 degree angle.. leave 1/4" from edge of board and snip. Onto the next. 12-13x wires per display. Then the led pcb can be mounted to the component board (wires fed through the plasma footprint & brackets attached). Then solder the wires to the component board once mounted.

-Moved 2.54mm headers up, as I see these easiest to plug/unplug from the back of the board. Added opening in middle for wires to pass-through to the back of the board. Headers are labeled on the back. These would only be used if you wanted a way to detach the led digit board from the component board.

-Changed to 18.6mm width digits. This helped get the positioning for the 7-digit display a bit tighter.

-Spent some time approximating where the digits would be & getting a silkscreen mock-up of the 7-segment digit added. Used a digital caliper on the 6-digit display to make sure the digits were spaced about where the plasma displays put them.

-Rounded the corners on the top of the display so they aren't so sharp.

#86 3 years ago
Quoted from CactusJack:

Do you think the round decimal point could be squared off into a bar with a small mask decal? If so, you should include connection if possible for those that want the feature.

It's possible to fake something that resembles a comma. What I did here was use small pieces of electrical tape & just kept roughly the same angle as the vertical segment and blocked off a tiny portion of the round arc on the decimal.


Other thoughts I've had over the years were having a round "dot" (think hole-punching electrical tape) and then overlapping it just slightly on the decimal on the top left and seeing if the resulting moon shape would look nice-enough.

There's also probably some type of "V" shape you could do as well.

I've yet to think of something that would be super easy/consistent. Having some kind of decal made up that aligned in the corner of the digit would make it easy.. but the resulting decal may be too delicate/thin for that approach. Matching the angle with electrical tape doesn't seem like it'd be too hard of a task. Then you'd just need to get the exposed areas roughly the same thickness (for 2x digits per display.. so 8 digits on most games).

For the 7-digit board, the comma/decimal connections are being brought over to the led digit pcb so the connection would be there if someone wanted to fake the decimal into a comma. I don't think the 6-digit boards had comma support originally, so no commas on that one.

#87 3 years ago

On the topic of using bare wire for plasma connections, does 18awg sound better than 22awg? I'd think the larger gauge would be a bit more rigid. I can leave the plasma footprint holes just slightly larger than 1.02mm diameter so it'd support that.

ebay.com link » 18 Awg Gauge Tinned Copper Wire Buss Wire 100 Length 0 0403 Silver

#89 3 years ago

Updated the 6-digit to reflect the newest design changes. Also made the bottom opening a bit bigger on both boards to allow more room for wire to fit under there if building with a "quick disconnect" (wire harness option).


#90 3 years ago
Quoted from ForceFlow:

Maybe cut a small piece of black vinyl on a vinyl cutter that can be placed on each digit display in the corner where the period is?

Yeah, I know there's some guys that are doing custom decals for things. I'm sure quite a few could be fit on a single sheet. I can look into some kind of option there and ask some questions once I get the actual digits in that will be used. I only have the 20mm width digits right now.

I placed the order for some WHITE digits. So hopefully they work out & once I get those I'll have a better idea of what may be possible with the decimal. Worst case it shouldn't be *too hard* for people to fake it with electrical tape (since many people have that around). Best case would be possibly a decal that could align with the corner of the digit and easily convert the shape into something more comma-like. You're limited though because of the smaller size of the decimal (in relation to length of a comma) so can't block off too much of it. I think just straightening the arc on two sides and giving the comma the most length & width is probably the best way to go.

I'm just not sure how delicate the vinyl would be in that size & if there'd be different thicknesses that would make it less-so. But for the amount that would probably fit on a single sheet.. if that option materialized, could just give 2x what was actually needed in case a few ripped during application.

#92 3 years ago
Quoted from ForceFlow:

Honestly, I wouldn't bother doing it with electrical tape. Too much work.
A decal or vinyl cut-out would be pretty easy to apply.

Yeah I'd imagine enough people are in that boat. I probably wouldn't bother myself with electrical tape, it doesn't matter *that* much to me to lose the comma. Years ago it did more.. for some reason.. but now I'd just like to get some displays in machines & having the ability to put colored led displays matters more than losing the comma on the 7-digit. Certainly a nice-to-have though if there's a way, so the comma circuit will be available for connecting on the 7-digit led board & could be enabled if/when there's a decal or alternative solution -- or for those that want to mess with it beforehand.

This design has been challenging to say the least. Working with old boards, some tight clearances, exactness of measurements.. several different revisions of the boards. Then attempting to make some type of DIY kit that can go together with relative ease. Not to mention having it look nice, not involve too many hacks, be fairly efficient, make wiring relatively easy considering the obscure 3.18mm pitch footprint of the plasmas. All-in-all, much easier to create custom boards.. but that said, I'm pretty happy with where things have evolved. The true test will be getting everything in and completing a full set for in-machine testing. Hopefully all the mocking up, calculations and testing that's been done so far have flushed out most of the issues.

#93 3 years ago

And this is what happens when you're waiting on some questions to get answered by the pcb fabricator..

Extra credit sprucing up with some images and retro fonts


#96 3 years ago
Quoted from Wolffdp:

Neat idea. It's interesting to see how other people attack the same problem.
I'll give you a piece of advice from my experience making the "Wolffpac" displays and originally thinking I could sell them in the ~$25.00 range: The additional expenses added by eBay fees, shipping and Paypal fees adds almost $8.00(!) to my cost to sell each display. In my spare time I've been slowly looking into some other options to eBay, but since this is not my day job, I haven't gotten very far.

Thanks! Yep, many ways to attack a problem. I doubt there's much cause for concern from anyone else selling kits or assembled displays as it'd be a niche market for people that have old boards to convert. If it even works out without a hitch.

In regard to eBay fees, Paypal fees.. I generally try to price that into the products. Especially with eBay, where I typically tack on an additional 10% of the website price. There's too much money lost @ 10% selling fee there to price things the same as direct purchases IMO. If I was you I'd just consider adding on the 10% fee over at eBay to encourage people to buy direct from your website & use eBay more for the exposure.

FWIW I have no issue with where assembled boards are priced at or full kits even. I think the prices of the stuff are where they need to be, I wouldn't be able to offer assembled or full kits for much less than what's being offered. This is just another way to go about things.. bringing it's own wide assortment of design issues along with it.. including a new one I just saw with the 7-digit display since its driver transistors sit practically flat against the plasma display holder -- where the new PCB would also sit. With the plasma footprint at the very bottom of the board, that's going to cause clearance issues (or worst yet, a chance for something to short against a transistor leg).. so looks like it's a good thing I didn't put the PCB order in yet.

#100 3 years ago

Here's maybe a dumb thought (got lots of them hehe)

Considering I just found an issue with having the plasma footprint low on the board (at least for the 7-digit) -- I started moving that up in the design, then thought about adding some SMD pads as another possible option and what that might gain for wiring.

Then.. I'm looking at a plasma board.. with display on it and an "AH-HAH" moment of just cutting the display off (for boards with the old glass still on them) and leaving the legs of the plasma cut high. Don't have to unsolder the plasma display & you have all the wiring at the correct pitch to solder to the new board.

Picture's worth a thousand words..

Does that seem like a decent option? So you could cut the leads reasonably high, then mount the led digit board on its brackets, trim the plasma leads & solder them onto the smd pads.

#102 3 years ago
Quoted from vid1900:

Certainly looks doable.

I tried cutting a glass out this way, I get about 7.5mm of lead if snipped at the top, with micro-shear flush cutters. The top of the SMD pads on the latest mock-up are 4.4mm. To give an idea, the top of the text of the pin identifications on the latest mock-up is about 7mm. So seems like there's plenty of length & I'm sure a quick method could be worked out to bend all the plasma leads back some so they sit flush against the PCB, ready for soldering! Saves having to remove one more thing, makes wiring easy for people that have displays with bad glass & can save the plasma leads.

I like it

So as far as kits go, I'd include 22awg stranded insulated wire pre-stripped and cut to length to run between the higher up plasma footprint and component board. Also an optional add-on for "quick disconnect" wire harness materials for the people that wanted to do that. Anyone that had display(s) with the glass still on them would then have the option of just cutting the plasma leads high and soldering them to the new PCB. Could also use bare wire to the SMD pads if the display glass was already removed. Either way, lots of options.. and SMD pads are low enough that there shouldn't be a chance of wiring crossing connections even if using bare wire there.

#104 3 years ago
Quoted from CactusJack:

Are you planning the pads on the front or the back of the LED board?
I know it might make soldering a bit more difficult but I am thinking it might look a bit sloppy to see the flat Plasma leads soldered to the front of the LED board as opposed to the back. Maybe its just me.......

SMD pads would have to be on the front, as the LED board will sit in back of the footprint on the component board. Can't bring the board forward as the digits would hang out *much further* than the original displays at that point. I'm trying to adhere as close as possible to original display dimensions/depth to be on the safe side (as far as clearance between backglass and the display). Luckily I *think* everything should be pretty close to original, as it just worked out that way with the brackets found + pcb/digit depth. The digit faces should sit about where the front face of the plasma holder does, maybe just ever slightly more. Clearance issues & designing around an existing board have made this one a challenge. Still, it should look pretty slick I think regardless of what method(s) are used.

There may be some aesthetically pleasing way of hiding soldered plasma display leads, but I don't see it as a big issue since it'll all be close to the bottom of the board. Will be doing these PCBs in *BLACK* as I think that looks best. If it bothered anyone enough, some black trim (black foam tape, vinyl or electrical tape) would probably work. I'll definitely be taking some pictures of the various ways to build these once I get some boards/materials in.

#107 3 years ago

Paper print-out approved!

Lots of paper has been harmed in the making of these


Leads seem plenty long enough (they'll bend back some as the pcb will sit back a bit further than what's shown here). Once I get some boards in I'll get one soldered this way. Hard to picture it until you can actually solder one together like that.

#110 3 years ago

I'll not be touching Gottlieb Sys1 lol. There's a number of reasons .. custom digits, probably not much to salvage from the original boards.. not to mention lower demand for Gottlieb in general & many more original Gottlieb displays having survived long-term due to the lower voltage used on the displays or the type of display. The market is actually pretty soft for new glass for those because it seems not many people need it. And I'd imagine there's a good amount of used displays up on the market often-enough with the machines getting parted out since they're not super popular.

Bally displays had been on my agenda a long time, so I was going to do something for Ballys either way. I actually had a full replacement set ready to prototype before I started on this project. The only reason this conversion thing happened is I was trying to cross the idea off my list.. deeming it impossible, too hackish, too power hungry, etc. But it didn't happen lol.

#113 3 years ago

CactusJack had me thinking -- since he'd like the ability to have the plasma leads push through the pads at the bottom. I was against having regular through-holes there to minimize any clearance issues on the back of the board. But, if I can discourage people from using regular wire (ie. bare wire, stranded wire) by shrinking the hole size to flat pin plasma size.. then that might be a good compromise.

[edit] Actually, after considering this a bit more.. the reason I wanted to do SMD pads front-side only was to move any conductive pads from the bottom of the back of the board to avoid any possible way for a connection to inadvertently be made there between a solder point & nearby resistor/transistor legs. In addition, it's going to be a bit too low to be soldering on the back with the board mounted on its brackets & tight against other components. So looks like I'll be sticking with the SMD pad thought.

#114 3 years ago

I'm going to look into decal options for turning the decimal into something that resembles a comma a bit better (for 7-digit displays that normally had a comma). Not saying it's definitely going to happen -- or that if it does it would be in-time with when I would begin offering kits for sale (providing everything else works out). But just just to get an idea of what's possible there.

Here's my thought with it..

So it'd be easy to align using the corner of the led -- of course dimensions need to be exact like everything else with this project lol.

#117 3 years ago
Quoted from gianfri:

I forgot to thank for the answer and I have the next question, is white or gel led display the same?

Two different things. The first is led digits that light up WHITE in color. Then there's what's essentially a transparent colored overlay (led color gel). You could also use colored vinyl, etc. But the led color gels are made for photography purposes -- specifically to convert white wavelength into various colors. Vid linked this website earlier in the thread: http://www.leefilters.com/lighting/led-02.html

So with white led digits.. you then put this thin overlay (led gel / filter) on top of the digits. This allows only the color of the filter to come through. You could also stack filters on top of one another to produce other colors/shades.. but if doing that you're limited in what colors can be produced from the already filtered color. It'd be like trying to get "blue" digits when they are lighting up red. The white digits give the most versatility since any color can be filtered from white light.

There's some neat articles if you Google for "color theory" or "additive color mixing" or "subtractive color mixing":

#118 3 years ago
Quoted from gianfri:

well, you can always suggest how to do and let brave people do it themselves because if you have to sxplode the budget for details then you really risk that the interest will be killed by the reality of the final price

I think the decals will be cheap.. there are a few Pinsiders providing a service to create custom decals. A full sheet would probably fit at least 20x rows of 10x decals per row. Probably more than that even.. anywhere from 300-500x tiny decals per sheet. I don't know what the limitations are of size or number per sheet, but it's worth exploring. If it can happen, I'd probably include a few extra per display just in case they were torn during installation (since it seems if it's possible, they'd be somewhat fragile).

1 week later
#121 3 years ago

Some Updates..

  • I'm still going to be waiting on digits for a few weeks most likely. I've ordered ULTRA RED and WHITE.
  • PCB order is going in today (or I'll eat my hat!). I've tweaked a few more things, but no *major* changes from the last picture with the SMD pads on the front of the board.
  • I've ordered some light filters, turns out they aren't adhesive backed. I'm actually thinking that's not a bad thing though, as I have some ideas for easy application & it will make them fully reusable whereas once vinyl is stuck on and removed, the adhesive goes to crap.
  • I'll have enough materials to build out roughly 15-20x sets [I think] but it may depend if everyone wants WHITE digits -- or how many 7-digit sets sell over 6-digit, etc.

I'll continue to think out-of-the-box on this one, but will probably be keeping a bit quiet on any further developments until I have pcbs and materials in -- and these are ready for sale. My current time estimate would be early-to-mid April as long as all the stars align. Should have pictures I can post of the displays in-use with various filters as well.

I've spent *a long time* getting the digit height and placement right.. especially on the 6-digit. I think people will be very happy with these boards if everything goes as I've designed them. There are still some unknowns right now (mainly not knowing if there could be variances in dimensions of physical parts), but as materials come in I'll have a much better idea of exactly what these will take. And whatever I come up with for instructions, it's going to be straightforward and as easy as possible to convert.. most of the mods should be able to be done from the back of the board & fully reversible. I know everyone likes simple. I'll do my best making these a very attractive alternative to what's out there.

#123 3 years ago
Quoted from Cheddar:

I'm down for 4x 7 and 1x 6 all in white with blue filters (if you sell filters too). Should look awesome in Xenon

Yeah I probably should have distinguished 6-digit or 7-digit on the waitlist. I'll be ordering around a 50/50 split of each, possibly favoring the 6-digit since one of those is also needed for a full 7-digit set.

Ideally I could have done a small prototype run on these, but this design came after a board order I'd already made.. and then some other test equipment I thought I might have flushed out the design for wasn't happening too quick either. So I just decided to order *enough* based on what was on the waitlist & what I'm comfortable eating if it doesn't work out for some reason -- and basically go for broke Here's hoping some kind of luck is on my side.

I'll have some filters -- was even considering doing sample packs for the first round of orders. People pick a color they want and get those filters.. but then also small sample swatches of the other colors so they can play around and know what they might like. Eventually that could turn into an add-on "sample pack" but it'd be a nice way to give back to the people that initially supported this idea.

#124 3 years ago

PCB order is in! I may hold off on remaining material orders until the digits ship, as funds have been a bit tight and the material costs on these are quite high. Digits are not cheap!

Could be time for a sale on something soon! lol.

#126 3 years ago
Quoted from Cheddar:

I personally would love a sale on the bally bench display. It can be difficult to use the switch tester in the cab when you can't see the displays!

I think I could make that happen, I know a guy that knows a guy

I've had some wacky thoughts of one-day event sales too -- like a "name your own price on a single item in your cart" sale hehe. I'm thinking that one would need some other stipulations if I ever did it. Just a hunch . Could do an Easter Egg hunt with having to search for a deeply discounted sale item in an odd place.. wrong page or part of an image.. Midnight Madness sale that only occurs for a few minutes when the clock strikes 12!

#130 3 years ago
Quoted from Cheddar:

How about a rockin deal for all the kits in a category?

Category type of bundles are likely to happen at times through the year I was doing a "MEGA BUNDLE" for assembled Bally diagnostic tools last December. I'll be offering that kind of thing again & will likely also include KIT bundles this time around as well for anyone that's interested in all the kits in a category -- at least for the PINITECH products anyway.

Quoted from gianfri:

I know a guy who knows a guy who would be interested in buying some memory testing tools... the problem is that it is not part of your own products therefore you cannot apply discount on those Pity!!!

Yeah, unfortunately wholesale costs prevent doing anything crazy with the 3rd party products.

1 week later
#131 3 years ago

Just got notice that boards are done and shipping! Looking good so far, from a picture anyway


Now here's hoping the led digits can ship soon. Apparently Chinese Spring Festival pushed back orders quite a bit so factories have been running behind and my digits weren't available for immediate shipment. I'm told they should have them this weekend.

Looks like it's about time to put in those other material orders..

#132 3 years ago

Boards came in for these and I also got the brackets today, looks like a great fit and makes it nice and stable. Note, this is an unmodified board in the pictures. Just test-fitting the display with the brackets.



#133 3 years ago

Test fit with the "dummy display" in Space Invaders Credit/Match.


#135 3 years ago

Thank you sir.. now for the other 99 things that need to go right to actually go right hehe.

#137 3 years ago

7-digit display test fit..


Also just to show what I was talking about in terms of limited clearance by the 7-digit driving transistors that are moved closer to the front of the board -- yeah I'd say that's close Adding any solder points back there would have caused some issues.


#140 3 years ago

Good news, got word the LED digits were in.

Bad news.. they made the wrong model & stamped the part number I had ordered on the model. Different pin-out, larger width.. can't use them. They can remake but I've been in contact with some other suppliers as well, and may try a different company. Still probably 20-30 day lead-time unfortunately. Glad they didn't ship them to me that way.

I'm going to request some samples from another manufacturer and see if I can get those in quicker. I'll probably wire up some other digits I have around so I can test the boards further and work on documenting the upgrade steps.

#141 3 years ago

Okay, think I found a new LED digit supplier after a few exhausting days of searching and contacting suppliers.

Game On! I'd expect digits to come in sometime around early May. Hoping that I'll be able to start offering the kits for sale shortly after that.

#143 3 years ago
Quoted from Grizlyrig:

The run around sucks but we are still paying attention,don't give up!

Thanks! Great to know there's the support!

There's some other small developments, but I'll hold off until I get digits in and can start offering these for sale before announcing those. Nothing monumental, but just some little niceties I think, if they work out. And possibly a few more as I get building a few of these myself and get the process stream-lined.

I'm really happy with how these look though, even just as a bare board test-mounted. Worth the time spent on driving myself nuts with the design!

1 week later
#146 3 years ago

I placed an order with a different led manufacturer. Paid for the order & got confirmation they got the payment last week. Was told they would be scheduling it and notifying me of an exact date, but haven't heard anything yet. I'm going to give it a couple more days and will then be contacting them for an update. Normal lead time is apparently 20-30 days on this stuff.. I guess including 7-10 days of not really much happening once the order is taken

In other news, I dropped the "red digit" idea. I have more WHITE digits coming than I had initially tried ordering before. And I also ordered BLUE. So if for some reason the WHITE + filter doesn't pop enough for some people that want BLUE displays, there would be another option.

#149 3 years ago
Quoted from jj44114:

Original Bally 6 and 7 digit displays in pretty good shape sell on ebay for around $40. All the LED based knockoffs don't look close to the appearance of the original plasma tubes.

Similar debate of led versus incandescent in the lamp sockets.... or vinyl records versus cds. Ultimately comes down to preference. To each their own. Some people I'm sure will settle for nothing less than the original plasmas because they'll feel no led display sets out there match the appearance. And that's okay.

I'm for led displays.. even ones using generic digits have grown on me. I've gotten over *needing* them to match in appearance to the plasmas as long as the full set in a machine match I'm good with that. And being able to use different colors in different machines is very appealing to me.

#154 3 years ago
Quoted from shimoda:

I still wish there was a more cost effective solution for Sys1. $250 for new display sets - well, that's half of what many are worth.

Yeah really putting any money into a System 1 game is pissing into the wind as it won't increase the value of the game unless when it's sold, the person buying it appreciates all the upgrades & values those additions just as much. So you just have to love the game(s) enough to warrant throwing additional money into them for upgrades or new boards. For the displays it makes more sense to just use the originals unless you can justify going belly up on the game.. or if you can sell the old set to help pay for the new set.

#159 3 years ago
Quoted from shimoda:

It's a shame Sys1s don't get more love. Frankly I find them as fun as most other games of the era at the end of the day. Not always do I feel this way. Sometimes I just wonder why they weren't as cool as some Bally's but then I get tired of playing those a bit and play the Sys 1's and enjoy them just as much. Nonetheless, they really don't sell all that well.

Yeah they aren't bad games. Like anything "slower" you just have to enjoy it for what it is/was and put yourself in that time period.

Movies, video games.. everything is really the same way. Special effects were crappier 10 years ago.. crappier yet 20-30 years ago. Video games are noticeably outdated pretty quick as graphics improve. So people that think anything old is lame and need the next thing to top the current will never appreciate the old stuff. But they're missing out on a lot of fun and entertainment in old video games, old pinball machines, old movies, etc.

And heck, if it seems too "slow" or lame... then add a drinking game element to it Usually does the trick..

#160 3 years ago
Quoted from Cheddar:

So while we're waiting for these I think it would be cool to have a technical discussion about preparing old displays for upgrade when the glass doesn't work.
Will it be safe to troubleshoot a board with the leds attached or can the decoder be tested on the bench?
Any ideas?

The only thing that could possibly happen if diagnosing with leds attached would be what normally kills/damages leds (ie. no current limiting resistors being used, causing the max current spec's to be far exceeded even for a short amount of time. That may be kind of hard to achieve though -- there's no GND connection going to the led panel. There'd also be current-limiting resistors on the connection the segment drivers are completing. So I would view this as -- if a digit driver locks on, okay you have VCC going to a digit's common pins.. but that won't do any harm. If a segment driver locked on, you have a path to GND through a current limiting resistor, again no harm. If a 4543 output is bad.. multiple segment drivers could be locked on.. but providing all the current limiting resistors are there.. still no problem.

Mainly it would just be a concern of an inadvertent connection between a segment *after* the current limiting resistor & GND. Which is something that is possible on any display, but there doesn't really seem like some failure scenario of an un-modded display that would cause that kind of thing except if *both* resistors on the circuit the segment drivers is completing were to be shorted and the segment driver completes the circuit directly to GND. Or otherwise if someone were to accidentally manually short the segment signal after the current limiting resistor to GND.

Oh and........ if Pin #1 on the header of the component board wasn't removed and/or trace cut for high voltage & this were to be powered on in a game. That would cause some issues for sure

#162 3 years ago

Aftermarket led displays at $250 are actually reasonably priced I think all things considered. I wouldn't have thought that 5 years ago, but material cost is fairly high, there's $8 in Paypal fees on $250. Uncle Sam gets a big share of the remaining net profit for doing nothing. More of the profit gets eaten up by assembly costs. Or if you're hand-soldering it's 5hrs+ of solid assembly time/labor.. and works out at about $20-25/hr if you're quick. Sure you can do SMT but you're still placing components and soldering through-hole parts. Guess what I'm saying is.. $250 is reasonable once you look at all that's involved. I think anyway. Didn't use to, but I do now.

#165 3 years ago

Project Update
I got a date of May 5th on the digits, they'll then need time to ship. So might have them by mid-May at best. I don't expect to have kits for sale for a bit as I'd still need to get 1-2x sets of my own assembled & tested, document various ways of wiring things up, pictures, instructions, etc.

I suppose it'll be good to do some level of beta testing on these.. and I know for sure that I'll have some people interested. Who wouldn't want quicker access to see one of these in action? So I'll be figuring out some of how I want to do beta testing as well.

One other update is.. I'm going to be able to reduce current consumption on these some more just by not driving them near 30mA per segment. The newly ordered digits I'm getting max out at 20mA constant current (instead of 30mA) so I'd already be driving them at a lower current even at that. But these displays should be plenty bright even at 15mA or less. So good things all-around in reducing total current consumption.

#170 3 years ago

Not sure how many Zaccaria were imported into the US over the years, but for sure the market is *much* smaller there on a conversion display or even regular displays for those machines just because the US wasn't their main market. There will still be US collectors that try and acquire all Zac's and then want to upgrade to LED at some point and will order half a dozen or more displays.. things like that happening, but in terms of volume of potential people that would be interested the larger market of course is going to be in whatever the largest market was when the machines were being produced since that's likely where more of the older machines still remain.

#171 3 years ago

New week, new teaser photo(s)

Here's what the display looks like when reusing the old plasma display pins. I used a flat-head screwdriver to push each individual pin back into the SMT pad so it would sit flat against for soldering. There may be a better technique to do more pins at once, but it really didn't take too long to get them all pushed back and then soldered. I then clipped the leg at the top of the pad. May be good to push them back, cut them a bit lower than the top of the pad and then solder. I also soldered more legs than were really needed (6x redundant digit line signals).. anything in parenthesis doesn't *need* to be soldered. I thought they looked pretty clean overall. This would be one of maybe 3-4x different ways these connections can be soldered.. and probably the fastest way if you're working with a display that still has its plasma glass installed.



#173 3 years ago
Quoted from CactusJack:

That turned out really well.
One caveat I see is that the possible strain on the SMT pad of the old lead trying to pull back on it (unless the leads were pre-bent and cut to the proper length before soldering) thereby making the solder pad becoming detached from the PCB. This could especially be true if a DIYer, that allows the solder pencil to pass too much heat to the pad when soldering. A possible remedy would be to convert that SMT pad into a custom pad with a small feed through that would produce a mechanical "anchor" through the PCB.

Yeah I had pre-bent them so they were flat against the board when soldered, seemed like the best way to go about it for both the tension concern as you mention & also just making it easier to solder them all at once. Adding a blob of solder and pushing a pin into it seemed like it could get kind of frustrating (leave marks in the blob of solder, have the pin want to pull out if you release it before the solder cools, etc). Not only that but somewhat of a chance of solder flinging then too. Pre-bend and get all the leads in place.. then solder.. then no tension to worry about & no solder flinging back either.

I made the traces pretty thick between the pad & the corresponding through-hole to allow for some heat to be taken away from the SMT pad. I think it'll be okay, but will definitely update if I find out otherwise after assembling a few like this.

3 weeks later
#175 3 years ago
Quoted from Andyball:

It has been awhile.... Any update for the anxious


Digits are in and I'm currently assembling some of the various revisions of display boards (both Bally & Stern) I had laying around for my own testing to see if there are any issues with anything. I'll try and get some photos/video posted so people can see where things are at.

I'm also trying to get a handle on all the various revisions of the displays, gathering schematics, etc as I'd like to cover as many of the revisions of possible in the instructions (as far as pictures) and also be relatively sure I have an idea of most of what's out there for display revisions for Bally/Stern. I'm thinking the way to do this is upload photos of the display revisions I have to a website & then have people help identify if I'm missing any and send me pictures of their boards. I'd like to be fairly sure I'm covering most of what's out there and it'll work with this type of conversion.. or be able to identify what revisions might not work. Should have a web page setup next week for that & will post something here when it's ready.

I'll post some more updates soon! But things are still moving along with these -- no show stoppers yet and they're looking pretty sweet IMO

#177 3 years ago

Ok took some pictures of display revisions I had & also a few from PINWIKI/Google that were different.

I don't have any DA-300 displays as donors currently.. I'll have to track one of those down or rely on a beta tester to get one assembled & send some pictures.

Here's the link with the pictures of various display revisions (warning, a bit photo heavy):

If some of the people with boxes of these things could take a quick look and see if they have anything that looks entirely different that'd be appreciated.

There's also a PINIWIKI section with display pictures, but I think I've covered what's over there.. except the weird 4-digit one (which I also have one of in my possession but the component board looks the same as a 6-digit board):

#179 3 years ago

I'm not seeing anything so far in any of these revisions that wouldn't make this conversion universal among the various displays or eliminate any of the revisions I've found so far.

#180 3 years ago

Teaser photo/video I just got a chance to test these out yesterday -- BLUE LED digits on Players 2, 4 & CREDIT. WHITE LED digits on Player 1 with an overlay just over a single digit (doubled up amber overlay that doesn't quite match the original plasma). Player 3 is the original plasma display for comparison.


Video (sorry for the poor quality, need to take video with better equipment):


The overlays will tone down the brightness of the digit some, not sure you can really get away from that. I'll do a comparison of the BLUE digit versus WHITE with blue overlay as I get some more WHITE digit displays built out. Will also try running the segments at 15mA instead of about 10mA as they are currently to see if that makes a noticeable difference.

For the upgrade on these I was just doing modifications to the back.. a few things to cut off the front, but then just adding the couple dozen resistors to the back of the board, jumper wire and pulling pin #1 on J1. I didn't have to change out any digit drivers, but there's one I'll probably change out because the digit is slightly dimmer. Doesn't seem that will be as much of an issue as I thought it could be.

The conversion probably took 30-40min per display. I'd think hitting a 30min time wouldn't be too hard for experienced solderers after they assemble one or two of these and get a feel for things. If removing components instead of doing modifications to the back of the board, it'd probably be closer to 1hr per display.. I'll be assembling some like that too and getting a feel for things.

It was exciting getting these in a real machine for testing.. I've only been testing single displays at the bench. To me it's night and day over the plasmas, I like the slightly larger digits & segments personally. And led displays in general definitely make the machines feel more modern. Happy with how these are turning out.. more to be done, but coming along!

#182 3 years ago
Quoted from Andyball:

Now I am really teased. I like the amber and toned down. Not too bright.

I've actually heard people ask if there's a way to tone down the led digits, not that they saw these displays in action (I didn't have the digits in) but just a general comment about aftermarket led displays. I told them with the WHITE with an overlay they'd be toned down some naturally. It basically amounts to it not BLARING in your eyes. So I don't see it as a problem really.. I think most people are going to find they like the look.

Quoted from Andyball:

Not replacing drivers is a real good thing. I am not sure what the part count replacement is for the Bally 6 digits, I would personally prefer to remove the old components and replace then with the correct values. Not a big fan of piggyback. This may not be the best for most of the people as they may not have the right desoldering equipment or skills. The added time for removal is about 5 seconds per hole and, at the end, about a few minutes to pull all parts out and clean. 45 to 60 mins per board seems fair.

Yep, that's what I'd recommend for anyone with the proper desoldering equipment. I just wanted to offer an alternative method. The downside is, you're reheating old solder if you're not removing it entirely. So far I haven't had issues doing that with the first set of displays I've done. Some of the solder seems a bit cruddy -- or was already reworked before, but all the displays have worked. I'll be completely removing components for the set of WHITE displays I do and I think you're right, probably looking at about 45-60min there to remove components & replace -- with the right experience, tools & technique.

Quoted from Andyball:

Are you planning to have all the versions done before release? There Seems to be a lot. I still have to check what is my version. I think I have the -21. However, most of my plasma displays are good and strong.

That was the thought. It's going to depend on what else surfaces as people go through the boxes of displays they have. STERN so far seems to have had the most major revisions. I'm interested in seeing if there's other BALLY revisions I currently don't know about.

Quoted from Andyball:

I plan to find other boards instead of using working ones.

That's the idea. Plenty of dead/dying displays out there for donors. No way I'd want people to take a perfectly working plasma display to do this conversion. You're better off in that case trading the working one for a pile of dead ones with someone, or just waiting until you get a machine in with a few more dead/dying displays so you can make a complete set of dead displays for upgrading one machine and throw the perfectly good plasmas in the other machine.

#184 3 years ago
Quoted from Cheddar:

DA300 if you like I can priority it to you

May take you up on that & possibly grab a few other displays from you too!

#186 3 years ago
Quoted from Cheddar:

Just leave me enough 7s to do my xenon

haha.. will do!

1 month later
#188 3 years ago

Sorry guys, been a crazy month for me working on a large house project that's disrupting a pretty heavy traffic part of the house until it's finished and has left me with little time for pinball activities. I've even had to throttle sales via the website some over this last month as to not get derailed too long. The good news is, that will be wrapping up soon (1-2 weeks) so I can get back to these displays. Things should speed right along at that point (testing of 7-digit models, load testing, documentation) & as long as all that goes well as it has so far.. beta testing will be right around the corner.

1 week later
#189 3 years ago

Interesting small update. When I last left off with these, I attempted some load testing & was getting a low 5v even without any led displays connected (around 4.4v) with the modified SDB I had used before with load testing. Had a bit of time today to look into things. Turns out the culprit was the FLUKE 115 meter I was using. For whatever reason it's not handling the full load even though it's fused up to 10A. Same setup, with a FLUKE 177 gives me 4.94v. Go figure. Just happy I didn't have to chase my tail too long on that one, thanks to having some other meters around.

#190 3 years ago

Stern DA-300 successfully tested thanks to the donor from Cheddar =]


#195 3 years ago
Quoted from Cheddar:

Did the board work as-is or did you have to fix it first?

That one worked as-is

#196 3 years ago
Quoted from barakandl:

You could do the PCB layout and offer them up as a complete kit. I think people would go for that. Nothing too expensive or complex to assemble on the display PCB.

Oh but that takes the fun out of converting an old display and turning it into something useful again! Kind of the gimmick or novelty of it all. There's people with boxes of these old boards that would otherwise just sit on the shelves forever. This gives those boards purpose and clears the space. Then plenty of machines that are bought every day with 3-4 failed displays.. instead of being bummed, it becomes a fun little weekend project to upgrade the *same displays* that came with the machine to something usable again. Less waste in the landfill, less space since you aren't storing "parts boards" that you might use *some day*.. the waste is waste (ie. out-gassed plasma display, etc) and gets tossed, the boards go back into the machine. And you'll save a couple more bucks over a full dispay kit.

Plus, it's got this cool retro vibe going for it with the sheer number of transistors/resistors on those old boards versus the scaled down number of components on new boards. Pfft, who needs ICs when you can do it all with resistors and transistors!

Conversion Displays - "The Greener, Not Leaner, Non-Cluttering Alternative You Can Feel Good About"

I'm working on my marketing angles hehe.

1 week later
#199 3 years ago

I'd like to judge interest in a *very* "rough around the edges" 1ST ROUND OF BETA TESTING on these conversion kits to basically get the project moving along & to gain some valuable feedback/suggestions based on other people's experiences.

Before getting *too excited*, let me explain briefly what that might look like..

There'd be limited instructions, possibly just a component list of things to change per display model and quick description of the conversion process (recommendations of required/optional modifications, some tips on parts of the build, etc). *Emphasis on limited instructions.* Should be enough for the more experienced individuals to know what needs to be done, but not for anyone that needs every step detailed with pictures when building out kits. Participants would need to have the right tools, be comfortable with troubleshooting Bally/Stern displays, be able to fill in the blanks some, & provide feedback/suggestions based on their experience. There may be additional small modifications to the display or slight differences in what's included with the kit(s) as the project moves forward, but at this time I can't imagine anything major.

Doing an initial beta test this way takes a lot of pressure off me honestly in not feeling I have to "get everything right" just to get some of these boards into the wild & get some feedback. The feedback would help finalize some decisions on how far to go with instructions, what to suggest for build steps & even the feasibility of this entire idea (cool/awesome or too cumbersome?). If some people that consider themselves experienced with modifying boards attempt the conversions and have issues, then I know the less experienced people will also have issues. So at least for this project, "relaxing" my own expectations of where things need to be at for an initial round of beta testing seems most appropriate to move things along.

So with that said -- who'd be up for that kind of thing? As long as there's enough interest, I can whip up some more details early next week on discount pricing for kits for the beta test, etc -- and may even be able to start sending some kits out later in the week [as long as I'm not finding I have to place any additional material orders].

#204 3 years ago
Quoted from vid1900:

I can make up a set of detailed assembly instructions + photos.

Definitely appreciate that offer & may take you up on it at some point!

Initially during the beta test I'd want to keep any detailed assembly info / photos private, as the "added value" part of these conversions will have been the additional modifications to make the displays more efficient than the steps the "old school" guides for creating conversion led displays (linked earlier on in this thread) had outlined. If/when the kits go official, it's not something I have much control over in preventing.. but right now I'll be asking beta testers not to post any detailed pictures/info of those modifications that might "reveal the magic" too early. That said, I don't mind discussing via PM with the participants possible changes that could improve efficiency more if there's something I've missed or ways to make certain parts of the builds easier that I never thought of.

I might relax on the whole "don't reveal the magic" perspective once beta testing is underway. I've not been bit *too bad* on being public with things, but am aware that happens & have found recently a few ideas/implementations of mine that others have copied -- and that's enough to know it's best not to let the cat out of the bag too early when you can help it

#205 3 years ago
Quoted from Drewscruis:

I'm game, where do I sign up.

I'll be posting some additional info early next week. For now, expressing interest here or via PM works!

For some reason I thought you might be!

#208 3 years ago
Quoted from Andyball:

If you don't mind shipping in Canada, count me in.

No problem shipping to Canada, just higher international shipping costs that are still hard for any of us to accept

#211 3 years ago
Quoted from RocketFromTombs:

I don't think my current project machine will be far enough along for any beta testing, but I'd be interested once these go into production.

Awesome, great to hear!

#212 3 years ago

Trying out a few of the other color filters I had bought.

These are all WHITE DIGIT displays with color filters being used, except for the CREDIT display which is a "brand X" led display using actual AMBER digits -- just for comparison.


Player 1 has a fire red that looks more of a dark orange -- with a 2nd piece doubled over on the 1s digit it produces more of a red. Unfortunately the "red" color isn't really showing up in the photo.

Player 2 has a blue overlay. Compared to a display with actual BLUE LED blocks, it's toned down a bit but would still look nice. Refer to previous pictures here where the BLUE LED block displays are shown. https://pinside.com/pinball/forum/topic/interest-check-diy-bally-led-displays-using-old-boards/page/4#post-3161815

Player 3 is green. I'm not sure what games would benefit from green displays off the top of my head, but had the color Again the 1s digit has a doubled up filter and it makes the color slightly darker.

Player 4 is amber. 1s digit has a doubled up filter. I *think* the color is similar to a single layer fire red, but I'll have to verify that.

And here's what it looks like with the machine off. I removed the blue filter for display #2 in the photo to show what the WHITE digits by themselves would look like with the machine off and no overlay.


Just wanted to show what was possible with the WHITE digits & color filters. Of course you can customize using whatever color filters you want whenever you want to change the color. Combine color filters to make entirely different colors, etc.

I'll post a few more samples of other colors I bought at random. By no means would it be the only options available out there for led color gels / filters. Since they're used for photography there's just about any shade/color you could imagine .. not just red, but light red, flame red, primary red, medium red, dark red, orange red, etc.

#214 3 years ago

Some other colors. Lavender in player 1. Light Blue in player 2. Yellow in player 3. Still the amber in Player 4. The ones digit is doubled up on all of them. Even though the yellow shows a bit darker in the picture.. in person I didn't see that change really.

I color adjusted the lavender as it looked too blue. I'm probably misrepresenting still, it's not blue.. but not super purple either. Anyway just some more samples of what other colors might look like. Take it with a grain of salt.. different displays/monitors may be showing the colors a bit differently than they actually appear.


And here's probably a better idea of what the doubled up fire red producing more of a red digit looks like. Anyway it's the only red filter I have so that's the best I can do


IMO the colors look better in person, brighter & more vivid. I'm sure a ton of that has to do with the equipment I'm taking the pictures with and lighting, but some might be in-part due to how the displays multiplex and what that does to camera/video equipment.

#217 3 years ago
Quoted from too-many-pins:

Great machine for showing these! Nitro is one of my all time favorites as far as art goes! Gameplay not so much!

Not thrilled play testing these on that one... Zzzzz. Usually do more attract mode testing.

Quoted from Grizlyrig:

The blue looks really nice,breaks up all the red on that particular backglass.

Yeah it seemed like a crazy combination at first but grew on me quick. I like it. It's like led bulbs, sometimes crazy combinations work out to be a nice contrast.

#218 3 years ago

I know, I know.. I need to be posting some info on the beta testing & pricing. Still flushing out a bunch of details & ideas this week, including better ways to test these before you put them in your machines! More details to come as the week progresses, including that much desired pricing info. Who asks people how many displays they're interested in for a beta test before releasing pricing? All questions will be answered soon enough. Except the top secret confidential stuff of course.

For now, I'll leave you with my personal achievement of the day -- finally getting around to sticking a set of the BLUE DIGIT (ie. no filter here, just pure blue digits) LED displays in my Stern Black Beauty Shuffle Alley. The entire reason I got back into doing an LED display project was to upgrade the look of this machine, as I thought it'd look cool with BLUE displays & also thought the standard plasmas made it seem a bit outdated. Not sure how many of these machines are in existence but aside from seeing one or two pop up on eBay, you rarely hear anything about them.


What's cool though is it uses 7x displays & LM323K (3A) regulator.. I was hoping it used the 3A so I wouldn't need to swap out the board. It'll be a good long-term test of a "worst case" scenario with more displays than usual on the less-beefy regulator. I don't expect issues though as the load on these displays was looking pretty darn good to me.

Of course the picture is taken in attract mode. I don't have time to play games! Maybe after I get some of these kits out!

#219 3 years ago

OK I have obviously needed to get some kind of idea of ballpark pricing here as I had a few people ask in PM's when I was inquiring on number of boards they were interested in for the beta test. Have thought about it some over the last few days and I think I've come up with some numbers that should make people *very happy*. But you tell me

First I just want to say, there's a lot of thought that went into this pricing and the numbers I've come up with here are actually below what I expected to be able to offer full sets at. I'm pricing for volume, which is something I'm not accustomed to -- and these are also priced for what they are.. conversion kits, that require you to have spare bad displays as donor, technical skill to upgrade and diagnose them, and may even take some finesse or additional repairs to get them going -- but boy do they ever look oh-so-sweet once assembled! I for one, am *thrilled* at the idea of having WHITE digit displays, convertible to any color just by swapping led color gel filters out. I'm excited for people that buy the WHITE digit displays to weigh in on their own thoughts because of this.

So without further adew, here are the prices I'm targeting/estimating for the introductory price of the conversion kits.. when they become officially available. Hopefully these work out in volume & material costs stay fairly consistent.. and I can make these prices stick for the long-term!

BETA TESTERS will receive discounts on this pricing. More info on that to follow later this week.

Estimated "Official Introductory Price" for Conversion Kits ( officially unofficial )

  • 6-DIGIT BLUE DIGIT SET (5X 6-DIGIT) $89.95ea
  • 7-DIGIT BLUE DIGIT SET (4X 7-DIGIT, 1X 6-DIGIT) $99.95ea

**For WHITE DIGIT sets, add $1.50 per individual kit price (or $7.50 per full set price). This accounts for additional cost of digits + color filter of your choice (of what I have available). Filters may be cut-to-size or may require cutting (if buying a full set). Haven't figured that out yet.

Price would not include shipping.

Note: These are not available yet, just trying to give some idea to what official pricing might look like if beta testing goes well.

#220 3 years ago

Some additional info on the kits.

Kit Contents (still working on finalizing but this should be *fairly* complete):
-PCB (LED Digit Panel)
-LED Digits
-Bracket & Screws for Mounting
-A Few Dozen Resistors
-Jumper Wire for 5v Connection
-Foam Tape (optional to install around led digits to block light from backbox)
-Nylon Spacers (for 7-digit displays only. Bally 7-digit displays have a few traces running close to the mounting holes)
-Right-Angle Pin Header Strip (cut down as single pins, for use if your plasma display leads are removed or otherwise unusable)

*Note: The Foam Tape I got a good price on is 1/2" thick and will need to be cut vertically to fit better & also so I'm not sending out 14" or something crazy for each display when it's too wide anyway. Not something I originally planned on including, but I got a good price -- and figured why not to make it a bit more complete? Will see what the beta testers think and go from there, as there's perhaps other better options available.

Possible Bonuses that *may* also be included (at least initially):
-A few extra 2N5401 transistors for the digit-drivers (just in case there's a digit or two that are lighter due to a weak transistor). I'm waiting on these to come in (hopefully toward the end of this week). I have only had a few displays that I've had to swap transistors on because the digits were slightly dimmer, but it can happen.
-A few small samples of other color filters for the WHITE DIGIT sets (ie. enough to cover a single digit). That way you get to see how versatile the WHITE digits are.

#221 3 years ago

In addition to all of this, I'm working on some instructions for testing the displays at the bench before you put them in your machine. Simple test to light an "8" on one digit.. and then probe the other digit lines to light the "8" up. This will allow you to verify basic functionality of the display -- especially important if you're working with totally "unknown" boards from a box. Of course it will still be up to you to inspect the board & verify there aren't shorts or any major issues while you're modifying the displays. But you won't have to plug them into your machine and *pray* either.

Going forward if beta testing goes well on these I'll offer a small PCB as a DIY kit that will make testing of these easier for people that want to upgrade a large number of displays. That way you aren't messing with a bunch of your own alligator clips or wire harnesses so much.

Well, I think with all of that.. I'm going to call it a night. Tomorrow's another day & I've plenty to do to get these ready for the BETA TEST

#223 3 years ago

I was hoping to keep the positive vibes going on these this week with getting ready for beta testing.

Didn't last

I have a small set-back to figure out that I found yesterday when trying out the "comma enable" functionality (well, really a decimal) on the 7-digit Bally displays. The decimal didn't light. Immediately had a hunch what the problem was, but was hoping it was just a bad transistor on the comma circuitry. No dice.

The short of it is, for the way the segment driver circuits are changing to support the led display conversions -- resistors will be needed on the 4543 segment outputs.

The problem is, there's no great way to add those resistors in there. Still thinking through this one. I have a few ideas, including possibly just doing a small piggyback board at the 4543 (which would require removing the 4543 and replacing with a socket or soldering the piggyback in directly). The piggyback would allow resistors to be placed in-line with the segment outputs and segment driver transistors.

Beyond that, I'm not sure I've come up with any viable options that don't involve removal of components.. or something that is a bit of a hack (ie. soldering resistors in-line at the transistor pad, 4543 leg or cutting a trace to solder in the resistor). Working within the confines of the old board eliminates options pretty quick.

So with that said, is a piggyback at the 4543 an okay solution? Or does that turn people off to this idea? Just curious.

#228 3 years ago
Quoted from gianfri:

Is it possible to place the additional resistor on the LEDs PCB? even for a second release of the PCB... it's an idea...

I had that thought -- remove the segment drivers and reroute the signal from the 4543 to the collector pad so the 4543 segment output goes to the LED PCB instead. Then have the current limiting resistors on the LED panel, along with the segment driving circuitry.

Couple ways to attack it really, just a matter of figuring out what the simple solution is. I actually have the display working fine by going back to the original transistor configuration that *shouldn't* require resistors on the 4543.. decimal lights on the 7-digit and the display seems to work just fine. Just not sure yet if it's a solid solution and that's taking time to look into.

With that said, if someone with a ton of EE experience and understanding of using discrete transistors in an emitter-follower setup, send me a PM

#229 3 years ago

Ok another update..

Looks like replacing the 7x segment drivers with n-channel FETs will do the trick. Seems easier than doing a separate piggyback PCB for the 4543. I'm going to modify a full set of 6-digit displays for testing, along with a 7-digit. Should have beta testing back on-track if it works out, I'll know in a day or two. Can then make some additional material orders and as soon as those come in.. get some kits out for beta testing soon after.

Sorry for the delay/wait on these.

Feel like it's been 2 steps forward, 1 step back through much of the journey. But I don't give up easy

#230 3 years ago

Reconfigured the 7x 6-digit BLUE displays in BLACK BEAUTY (shuffle alley). Things are looking good

I now have a big pile of MPSA42's from replacing the segment drivers. I was going to snap a picture of lol. Despite being bummed having to replace those with FETs, the displays get more efficient this way. And everyone will for sure have plenty of spare MPSA42's. Gotta look at the upside

I also tested the 7-digit Bally display again with this setup, and the decimal lights like it should. Going to convert over some WHITE 6-digit displays tomorrow.

#231 3 years ago

Got some timing on fully desoldering & repopulating components on a 6-digit display using my Hakko 808. Up until now I was just soldering resistors in parallel on the back. I'm not super fast, but total time to convert a single display was 35-40 minutes. Keep in mind this was a display with plasma glass on it still, so I could re-use the old leads (soldering to the smd pads on the led panel pcb). Otherwise add another 5-10min for mounting with the single RA header pins. The display looks nice and clean since all the existing footprints were used. Just a single jumper wire on the back. I like it, think I'll be recommending that way in the instructions as the preferred "clean" way to convert a display. But soldering the resistors in parallel on the back is of course also an option.

In other positive news, yeesh these things are efficient! Surprising me even as I *expected* about 80-100mA per display. They're instead measuring 35-45mA additional load per display on the BLUE digit displays. Depends what value resistor is used for the segments (ie. "bright enough" or "BRIGHTER!!"). This matches efficiency of the better aftermarket displays.. even the ones with "adjustable brightness" that the user can be set anywhere from around 25mA (too dim) to 115mA or more (full brightness) for a single display. I'm happy WHITE will probably be a bit higher (45-60mA or so) since they'll need to be brighter with going through a color filter. Test is being done by locking a single digit with an "8" on it at the bench (faking signals, no MPU connected, watching that the 5v line stays solid). Also confirmed via machine load testing.

Additional materials orders going in today and tomorrow. Should hopefully be able to get some of these out for BETA TESTING soon!

http://www.pinitech.com - "Pinball Inspired Technology"
Kits, upgrades and test equipment for pinball machines

#234 3 years ago
Quoted from roar:

Is it too late to be added to the Beta Test list? I'm interested.

I'll be sending out a pre-order link this week to everyone that had expressed interest in the beta test & should have a better idea at that time how many sets are remaining.. but I think there will be room for a few additional people. Just want to give the people that had already expressed interest first crack at boards for the beta test.

#235 3 years ago

Finally snapped some pictures of the display I fully desoldered components off of to show how clean one of these can look if you have proper desoldering equipment and are fully replacing the components.




#237 3 years ago

Pre-order links have gone out to the initial group of people that had expressed interest in beta testing these conversion kits!

So it looks like they're happening!

At the end of this week I'll include a few additional people that have recently expressed interest (roar & Slim64). I do expect to still have a handful of the 7-digit sets left from rough estimates, but the 6-digit sets look like they'll be in short supply.. I used up a dozen of those boards myself!

http://www.pinitech.com - "Pinball Inspired Technology"
Kits, upgrades and test equipment for pinball machines

#239 3 years ago
Quoted from Andyball:

hunting is now on to find non working displays

Did you ever think in a million years you'd be looking for non working displays? lol.

#246 3 years ago
Quoted from Andyball:

That was quite a find! If I can find half a dozen, i will consider myself lucky. I had one display that was acting funny. I now have a second one but think it is a solder joint as it is ok when I move the display.

That just means you need to buy some more project machines!

#248 3 years ago
Quoted from Andyball:

Lol! True. Just the prices in my area are crazily, overly inflated. I plan to get a few more. I love more projects than playing them....

Hah, I have that problem with spending time more on projects than actually playing games! I keep telling myself to "play more!" Maybe this winter.. lol.

I've said this before but what I like in this idea is that when you do buy a project machine and it has bad displays, it reverses this psychological viewpoint of feeling hosed. Instead it's like "cool, I can upgrade that" -- even if you just toss the display aside and put a good spare plasma in the machine. When you get enough bad displays for a full set you can buy a kit & upgrade the bad ones.

There's something kind of zen in it all I think. Being able to reuse the old boards, saving a bit of money, not having to throw something in the landfill that is still usable, not having to take up space with bad display boards that are just there for "spare parts" (and really how many do you need for spare parts once you have a dozen or more). I feel like I'm helping to reduce a source of anger for people too. I know when I was fixing up a lot of Classic Bally/Stern games 4-5 years ago, it was RARE to acquire a game that didn't have at least one or two dead or dying displays -- burnt digits, fading, arcing, etc. I always thought "Ah crap!!! There goes any remaining good plasmas I have." Or you're just having to "live with it" and play the game like that.

It's *so distracting* to have bad displays in a game and just highlights any other imperfections the game has IMO, whereas a nice crisp set of led displays in various colors that compliment the game even on "beater machines" will make the games look A LOT better. It's kind of neat feeling like these conversion kits may help get some LED displays into machines that might otherwise have never seen them because spending a few hundred on displays can't be justified.

#249 3 years ago

Working on a product page for these, most of the beta testers have seen this but just thought I'd like for anyone else wanting to get up to speed.


#250 3 years ago

I hate to waiver on this, but I'm feeling like I should be recommending people without decent desoldering equipment & technique to solder the resistors being replaced in parallel on the back of the board to minimize chance of pulling up pads when doing the conversions. Not sure what the split was on how many people like that idea or how many people wanted to fully desolder.

I do like the clean look of a board that has had all the components replaced... just that enough of these boards are single sided & put up with less heat or stress at the solder pads because of that. And really what makes the matter worse is on half the boards I have, the component leads were bent 90 degrees flat to the pad making it harder to just desolder the through-holes with the Hakko. It becomes desolder to expose the lead, cut the lead at the 90 degree bend with micro-flush cutters, add a bit more solder to the pad and remove the small piece of the resistor leg. Then clear the through-hole with the Hakko. Of course doing this assembly-style on a handful of resistors makes it faster. Another option is just to cut the resistor off the top of the board and leaving some length on the leg.. so you can then heat the pad up on the back and push the leg through enough to cut that 90 degree bend off or just pull the leg out completely. I'm sure others may have better recommendations/techniques that could be discussed here.

If soldering in parallel, there's really just the segment driver transistors to deal with removing fully. Once you have one display of a certain model done with the parallel resistors, the others go quicker (just set the board upside down in TeePee position and cheat off that one when you do the next. Then of course anyone that has no fear and has great technique has the choice of fully removing the components.

#251 3 years ago

Can't claim that the colors are spot-on, but this should be a more accurate representation than other pictures I took. Used a better camera and messed around with ISO, etc until colors seemed closer to what I was seeing in person. So here's what a rainbow display looks like WHITE display with various 1" color filters on it. The left-most digit is without a filter.


#255 3 years ago
Quoted from Andyball:

For me it is full desolder. I understand that parallel soldering is easier. It may cause some other issues. You will also have to have 2 configs of parts. I use Pace equipment. With 90 degrees angled lead, it is straightened with the desoldering tip then sucked up. It is done in about 5-7 seconds per lead. Need a little practice but easy once you get the hang of it.

Actually what's nice is in this case the same resistors could be used either way. The values aren't too critical on what's being replaced, except for the current limiting resistors on the segments. When soldering in parallel it'll drop the resistance some and displays will be a bit brighter. Easy enough fix though since you can just snip the factory-soldered resistor from the top side of the board if you're mixing & matching displays that were done by soldering resistors in parallel & displays that had components fully replaced.

As for the display color, I am kind of debating the color to choose. I am a little reticent to change the original amber color as I feel that I am "cheating" the intended design. However, with the white display, it is very easy to experiment. So I feel good

I don't think you're alone there. Same thing seems like it happened with regular led bulbs *way back* and most people didn't want to change the game look too drastically from original. Now look at how much people customize their leds.. colored bulbs in GI lighting to accent or highlight different areas of the playfield, different colors in the feature lamps. Just a matter of time before more and more people are very accepting of complimenting and even contrasting colors in the displays. But you're exactly right.. WHITE gives you the chance to experiment or go back to a more original look. Best of both worlds

#260 3 years ago
Quoted from aobrien5:

Sucks that it requires so much component replacement on the boards, really. I'm going to have to work on my desoldering skills.

Yeah, again soldering resistors in parallel on the back of the board limits what you actually *HAVE* to fully remove to just 7x transistors. That said, I do get that some people would prefer to fully remove and replace the components.. especially if they have the right tools & experience for that.

What people should probably do is take a junkier display board and use it as a practice board to see how they do with desoldering components from the board. You don't even need to buy a conversion kit, just try messing around with a display board and see how you do desoldering or soldering in resistors in parallel with other resistors. Plan on wasting a board to get a feel for it & decide if a conversion kit is right for you or what method you would use if you did buy a conversion kit. No harm done if it doesn't work out, the display wasn't doing much good anyway & there's no shortage of bad display boards out there

I'll probably end up doing a full display kit at some point that can hopefully compliment these RETROFIT displays (ie. mix & match and they'll look the same). Then the people with only a few old boards to convert could save a bit of money but still make a full set without having to track down bad displays. Or people just wanting full kits have that option.

#261 3 years ago

Here you go. This is a 6-digit display (26 resistors). 7-digit will have 28-30 resistors depending if it's a Bally with comma circuitry or not. Ignore the jumper wire on this one, obviously that has ways of looking cleaner as in a previous picture I posted.


What I found works is using ammo pack style resistors (tape and reel). Keep them in the tape, snip one side of them leaving 1/8th to 3/16th of an inch lead length. Then they're all handy to point-solder one side of the resistor into place. Just add a bit of solder to pad of one side of each resistor you're doing, then hold the resistors (still on the tape) in one hand & solder the short lead of each resistor to the pads you had added solder to. After soldering each one, pull down on the tape and it'll release the resistor. What you'll end up with is one side of like 6x or 7x resistors of that value soldered on the board with a long lead hanging down. Then just press the long lead into the opposite pad (directly in-line with the pad that was soldered), snip the lead over-top of the pad & solder. You can do those steps for all 6-7x resistors of a value at once. Makes it quicker not dealing with individual resistors and finding a way to be more efficient getting them soldered in. You then just have the 7x segment driver transistors to fully desolder & replace.

Here's the topside of the board -- really no different from stock, but a few resistors and CR1/VR1 diode cut off. J1 PIN #1 pulled.

I'm not saying soldering in parallel is for everyone.. or even that the conversion kits are for everyone, but I think the end result either way doesn't look too bad. Just as I said, you probably want to practice desoldering on a junkier board if you *plan* to replace all the components, jut to see how it goes and/or perfect technique.

#262 3 years ago
Quoted from aobrien5:

Nothing but 6s, sorry. Maybe I can trade you a couple for a desoldering tutorial.

Speaking of which, I think it's a good idea to be linking some of those kind of tutorials here (and in a full-blown guide for the conversions).

I found this earlier today:

It's more on soldering technique than desoldering, but shows good and bad examples -- and what to do if you pull pads. I'm actually thinking it'd be good to include a link to that for any type of kit I sell

#263 3 years ago

For the Xenon fans..

7-digit tests with decimal enabled. BLUE on top, WHITE with blue filter on bottom. Had to disassemble one of my early prototypes and rebuild it just to get another 7-digit display lol.

#264 3 years ago

Is there a shortage of 7-digit display boards for donors just due to less games using those?

Here's a thought. If these display conversions go over well with everyone, I could do a special run of the 7-digit boards that would allow you to convert Bally 6-digit displays into a 7-digit display. The catch is a few modifications would also be needed. Bally wired J1 PIN 10,11,12 together on 6-digit displays. So BLANKING @ PIN #10 connects to PIN #12 where the 7-digit boards get the 7th digit signal. This is why when you plug a 6-digit display into a 7-digit machine, it makes all the other displays go haywire -- it screws with the blanking signal to all the displays.

I'd have to see what could be done to minimize modifications. I'm thinking cut the trace between J1 PIN #11 and PIN #12 on the displays. Then run a wire from PIN #12 to one or two unused plasma footprint pads. The special run of the 7-digit boards would then include some extra footprints for the 7th digit circuitry and pick up the 7th digit from the [previously unused] plasma pad. No other modifications needed.. the rest of what it'd take to run the 7th digit would be located on the display panel.

Just a thought on a way to get around the shortage of 7-digit donor boards.

Not sure I can do anything about the decimal (if that matters to anyone) since that's created off the 4543 outputs. Technically it's feasible, just more wires and some extra components, but not sure that one's really worthwhile.

#267 3 years ago
Quoted from CactusJack:

I know it defeats your original goal to reuse the failed display board but I think it would be much cleaner to just make a 7 digit driver board to go with your display boards.
As I stated to you off line, that is what I intend to do.
The issue is it has to work in both Bally and Stern 7 digit games (as I recall, the J1 Pinout is different). Of course, I would also look into making it universal thereby supporting both brands of 7 digit games as well as the 6 digit layout. Then, one driver gets married to the desired 6 or 7 digit display board.

Yeah that's likely the direction I'll head. Already seeing that enough people are coming up short on 7-digit boards, etc. It's going to make sense to offer a full kit option as well. If the designs can compliment each other some so you could mix-and-match conversion displays & full display kits, that could be interesting.

I like the conversion kits in being able to hit a lower price point than I could hit with a full kit. But they've been a *HECK* of a roller coaster to figure out. It would have been far easier to never have taken on that challenge -- but I know a lot of people including myself had wondered if there was some possible way to do a conversion on the old boards without it looking really hacked. The end result is pretty cool and stream-lined but getting to that point =

I'll be happy to offer both options. The full kit option will be a breeze in comparison lol.

#271 3 years ago
Quoted from gianfri:

Yes, I confirm that there are now tons of full led displays available
ebay.com link
And as I said previously the price difference is really tiny but I respect the green aspect of Wayne's project.
» YouTube video

I don't see $50-60 saved buying a conversion kit set over a full-blown kit set as really tiny.. that's like 35% less. I wish I felt like $50 wasn't much money lol.

#273 3 years ago
Quoted from ForceFlow:

As I said earlier, I really like the idea of being able to reuse the PCBs, especially since there are so many with bad class kicking around. I just wasn't sure if it would be worthwhile to pair it with a custom/new PCB since there already are drop-in display replacements available.
I do like the idea of those single-PCB displays, though. They look pretty slick.

Everyone's got different motivations. Mine go back to having wanted to create my own full displays since this project in 2011. Most people aren't going to be aware of the hundreds of hours spent "behind the scenes" by anyone creating things or what the driving motivations are. With these displays, it's not about competing with anyone or what's happened since 2011 as far as many other options being available. It's completing a goal that I've had for at least 5 years now, probably longer. Having something to show for literally hundreds of hours of research, development, time spent sourcing materials, contacting suppliers over the years. Putting something I feel is a solid product out there for others to enjoy.

And by golly, if I say something is efficient it's going to be efficient. I'll show people how to test the efficiency. You'll learn some things just by buying a kit. I'll have a table with mA usage for a single display and a mA usage for a full set.. and how that changes with different value resistors on the segments. There will be a section on bench testing the display.. what to connect to have it light up an "8" and then how to measure current draw if you have a meter, so you can see that it's measuring as expected and that it actually IS efficient. I dunno, I think that's kind of cool.. teaching people, demystifying things. The conversion kits aren't for everyone, but I'm sure some people will appreciate learning a little something in the process. For people that want a full kit or something totally plug-and-play, yeah I'll probably get around to that stuff too.

#279 3 years ago

Some kits are shipping out to beta testers today. If I wrote "digit drivers" on the anti-static bag with 35-40x mosfets in it, that's a mistake. I'm pretty sure I did this with a few orders. Those are segment drivers. Hopefully I don't miss anything in the kits, but if you notice something is missing when you get your kit let me know. Some people got extra mosfets with 6-digit kits. My bad

I am working on some instructions & included a link to those, but they're work-in-progress so if you get your kit in the next few days.. just know that the instructions are still being updated. I hadn't originally planned to have much for instructions, but I don't like leaving people completely in the dark.. so I attempted to put something together there.

Some more packages should be going out tomorrow.

#281 3 years ago
Quoted from shimoda:

I was traveling and so completely lost track of this thread until just now and am a bit fracked at myself for having forgot to get on that damn beta list as I have a Stern needing two replacements and the perfect test vehicle! If you need any more beta testers, let me know.

Out of materials right now. I was considering doing another PCB order but I may hold off until after the beta test & after I've had a chance to assess things myself. I've been a solid 3-4 weeks again with pulling the rest of the project together, packing up kits, working on some instructions, shipping out beta test kit orders this week. A bit exhausted to tell the truth and I'm planning on taking a bit of a break while the beta testing is going on. I'll be available to answer emails/PMs and will check this thread some, but otherwise will probably be laying low for a bit and not getting into any new projects.

If the beta test shows more demand for these, I have a lot of things to streamline to make it work. Material cost is quite high (40-50% of the price), packaging kits has been very time intensive. I didn't account for some materials like the mosfets that were later added, ziplock bags, etc. Shipping was underestimated for some packages because they all had to go Priority Mail due to weight & costs vary widely depending on what area of the US you ship to. Because of the sheer amount of materials per display set bulk orders of materials go very quick, but they're also bulky when they come in.

At this point I'm just happy-enough to get some kits out there for beta testing to see what people think & see if there's demand. From there, it's just going to depend on feedback, assessing all of the issues I've hit, looking at what could be streamlined more & then deciding whether there's some way to make it all work out in volume and still be profitable.

#284 3 years ago
Quoted from johninc:

First effort.


Solid first effort! Looks awesome already!

#288 3 years ago
Quoted from johninc:

First seven-digit done, now I just need three more donors. Edit: Digits are all the same brightness, just bad photography.

Yeah, Xenon's a tough machine to take a picture of because of the first layer glass having the smokey design on it. Seems kind of odd they didn't go cover the whole display(s) with that.

#290 3 years ago
Quoted from Cheddar:

Can you give extended instructions on enabling the commas. I'm not understanding the ones in the guide

Looks like you already have the "COM" signal wired between the LED digit panel & plasma footprint.

If you then bridge the smd pads on the back of the digit panel (see picture below) the decimal should then light.


#291 3 years ago
Quoted from Cheddar:

I need to work on getting the components to stay flat.

For the resistors you might find one of those cheap lead formers handy. I've gotten pretty good at sizing resistor bends on my fingertip.. and swiping down with the opposing hand's index finger & thumb to bend the leads. I usually find a good place on my finger to set the resistor and get a consistent bend.

For the mosfets, I first bend the middle lead back at a 90 degree angle to the bottom of the mosfet body. Then the two outer legs get bent forward at a 90 degree angle to the bottom of the mosfet. Then put your 90 degree bend on the front legs maybe 1/4" from the mosfet body. Leave the middle lead as it is.. and put the middle lead in the footprint hole first.. push it in most of the way, then fit the 2 front leads in their footprint hole. I'll try to document that better at some point. A bit hard to explain without a few pictures

#300 3 years ago
Quoted from Drewscruis:

I'm working on 6digit displays but it's only taking me about 10min a display to disassemble. That's using a hakko desoldering station. and another 10-15min to reassemble.

That's awesome to hear!

I think this was the scenario for one of the few displays I fully removed components on:
15min to desolder components with Hakko
10-15min to repopulate & solder jumper wire
3 minutes to solder digits to new PCB
5-10min to mount the digit panel

How have you all been dealing with those bent over resistor leads on some boards? Any tips there?

#303 3 years ago
Quoted from Cheddar:

Bench grinder?

I think that or dremel grinding would be too risky. Those pads aren't being held by much on single-sided boards & seems like it'd be too easy to grind a little too far.

Here's a forum post I found on the subject of leads bent over pads:

Seemed most people on that post found snipping components off the top and pulling the leads from the bottom was the best way to go about it. I'm definitely interested in finding out if there's faster/better techniques though.

#305 3 years ago
Quoted from Drewscruis:

I heat the lead with the hakko, bend the resistor straight with the hole in the tip of the iron pull the trigger, suck the solder done. Takes practice but been doin it this way for years.

What are you using to straighten the resistor lead? Looks like the Hakko tip? Would love to know how to do that and not be losing pads left-and-right

#307 3 years ago
Quoted from Drewscruis:

I'll try and take some pics/video the next time I'm tearing one down. But yes I use the tip for the hakko to bend the lead straight. I rarely loose a pad.

That would be most appreciated!

The time consuming part otherwise is having to be this surgeon extracting bits of leads, using multiple tools and not really having enough hands to make good time. Having 3-4 hands would help a ton!

I'd imagine with just using the desoldering tip, the trick is to minimize heat applied directly to the pad. Maybe add a bit of new solder as you're heating up the joint with the desoldering gun, sneak under the lead and pull up when the solder is fully melted. Make sure not to pry against the pad at all.. just pull up on the lead with the nozzle. Well that'd be my guess how to reduce pad damage anyway

#311 3 years ago
Quoted from Drewscruis:

Here are some pic's of the early style bally display with the pullup resistors added to the backside. also I used bare tinned solid core wire to attach the new display to the old base.

Looks really clean! What modifications?

I like the small jumper wire on that model display too.

#312 3 years ago

So for the people that have completed a set.. or almost completed, is this something you want to continue to see offered? Would you "do it again?" ie. Should I order some more PCBs and materials sooner than later?

#318 3 years ago
Quoted from Cheddar:

played around with setting the header pins. there is enough clearance to set them with the board mounted

Great! Yeah I found you could do that for a good amount of pins.. some exceed a 90 degree angle and may hit the hole, but require you to grab them with pliers and pull them up slightly as you're pushing into the hole. The trick with pre-inserting pins I documented, I know some pictures are needed.. but I actually got pretty quick at that too. Hey whatever works for people & if you find something that works that wasn't mentioned in the document or this thread.. it's great seeing some new ways of doing things.

The end result when all those right-angle header pins are soldered in looks pretty nice I think.

#322 3 years ago
Quoted from Cheddar:

made a test harness and it works great

it's hard to see but I crimped a lead into pin 12 to allow the seventh digit to be lit

Cool, that's pretty much what I did! Works really well

I plan to offer either a harness or small pcb board tester as an add-on at some point. Buy it once (or build it yourself) and it makes life a ton easier.

#325 3 years ago
Quoted from Andyball:

I haven't received the kit yet expect it soon. Is there any tracking number?

Sent you a PM with the tracking number. It's saying it was delivered yesterday?

#326 3 years ago
Quoted from CactusJack:

Works fine on a Bally Tester too!
One down, Four to go......

That's really cool to know

#334 3 years ago
Quoted from vid1900:

Tip #3:
Put all the boards in the dishwasher, and run on the "Pots and Pans" cycle.
The boards will come out clean, hot and dry; ready to be serviced.

I have to try this sometime when the wife isn't home to see it

#338 3 years ago
Quoted from Cheddar:

all done, they look great. Maybe should have gone a step down in brightness

There's a few things you can do. Swap resistors on the segments out for the higher value or use a neutral-density led gel / filter. The filter shouldn't change the color, it'll just knock down the brightness. But yeah if you're using those WHITE digits as white, probably want to go with the higher resistor value in the kit. Unless of course you put on a pair of sun-glasses when playing hehe.

#341 3 years ago
Quoted from vid1900:

Put a layer or two of blue gel over the displays and knock them down a little more till it's to your liking.

Those are actually WHITE digits being used as WHITE in Centaur. I think the camera just couldn't handle how bright they were and did this blue aura thing Ghostly!

#344 3 years ago
Quoted from Cheddar:

They look great in person

Oh good. So no sun-glasses needed

#352 3 years ago

Had a request to change the holes where the right-angle header pins fit into to make them slightly larger. I had actually thought about shrinking those slightly to see if I could have it so the pins would insert tighter, thinking that might make it easier to pre-insert the pins. I can see merit to making the holes bigger though -- that'd let the pin drop vertically easily even if you pre-inserted pins at the edge of a table.. then keeping the board tilted back some and positioning it over the plasma holes on the component board. I tried to document but apparently didn't make sense to many people Anyway, I can also see slightly larger holes helping if mounting the digit panel first & then inserting the right angle header pins as Cheddar had shown pictures of.<