Quoted from RandyW:So I add up all the rows and that is how many I need? I am sorry if I am missing the obvious.
For example, to build everything, do I need 48 of the vertical ethernet ports?
The numbers in the columns are the actual total that you require for the "bundle" listed at the top of the column. Since you have the "all" bundle, you only need the actual counts in the column - not the (horizontal) total of all the columns. Just order the counts in the right most column.
Quoted from Coyote:but the spread of the light.
I will try the 4 LED configuration and see how it goes. Frosted rather than clear dome helps to diffuse the beam. I take your point. My initial thoughts are to try to use the minimum number of different components, both for my sake and for those that build their own boards.
Quoted from Coyote:Instead of one, put three in each quadrant. (Or better yet, spread them around the whole face, instead of limiting to quadrants.)
This board is a one (main) board solution. This means that both the minute and hour slotted optos along with state detection logic are all on one board. This creates "hazards". These are components that are fixed position and cannot move. The state detection logic adds additional components (that can be placed anywhere) but these are components that do occupy additional board space. Around these, the support components (resistors and capacitors) need to be placed. Once those are placed, the lighting is finally placed. After all that, the traces have to be routed.
The other boards that are available (except the Ingo board which I think is still technically available) use the OEM two board solution. One board for the hour slotted optos and the main board with the minute slotted optos. Those boards also don't have any state detection logic as they simply put the slotted opto output on the switch matrix. See https://pinside.com/pinball/forum/topic/what-is-the-best-after-market-tz-clock-board#post-3928634 for more information. It appears these other boards also use the TCST1103 rather than any custom slotted opto. They still need to raise the slotted opto off the board to reach the minute hand so they can interrupt the slotted opto but the distance is not as far since the two board solution brings the minute board 5mm closer to the minute hand.
The hour slotted optos run diagonally across the board at 45 degrees and create the hazard. The minute slotted optos run around the edge of the board leaving a large amount of the board available for any lighting options and traces routed to them. I'm going to try the 4 LED solution currently implemented and see if there's a problem. If a problem exists then I will investigate looking into alternate lighting options.
Update: Looks like Cree LED CP41B-WGS-CK0P0154 might be a suitable LED. I'll try to get a small number of these and see how they perform.
Quoted from Coyote:However, you're killing yourself doing all these by hand still, and I'm afraid of even mentioning this to give you more of a headache.
Can't argue with that. I have put measures in place to reduce my workload so it doesn't become a job and stays "fun". It might sound strange but building boards is sometimes a good change of "scenery".
Victor, just to let you know Taxi and TAF #1 are still going strong on route and league play. Kinda hard to see but they’re there. 
29B2C04A-67D9-44A8-A970-5CC5E32DD4DF (resized).jpegE50EC74A-3A9A-4A22-9597-D4470E64D6CC (resized).jpegFB91A995-2443-455F-93C6-7A155C9B70FB (resized).jpegHere is my BOM list for the complete portable tester boards on Digikey. It should be everything, minus the relays and the 8x8 LED grid. Those had to be sourced thru alibaba and Arrow. There are a lot of extra LEDs in my list, because I needed them for other stuff and I generally added an extra of each component, just in case.
https://www.digikey.com/en/mylists/list/HLYGLC6YVJ It isn't cheap, but you can get most everything in one place. Comes to about $250.
Quoted from DumbAss:This board is a one (main) board solution.
Yeah, I realize that locations of some items would cause an issue. Heh. Potentially where SMD would com in handy. (There WERE SMD-mount optos in the past, tho I haven't looked in a looong time.) In either case, I agree with you, and look forward to your output! 
Quoted from DumbAss:It might sound strange but building boards is sometimes a good change of "scenery".
Totally get it. Second job is a physcially demanding job on a railroad - and I enjoy it. Lets me 'shut down and just do'. So yeah, I get it. 
I try to repair my System 11 MPU board : I can't find new 102M capacitors 50V (M is for +-20% tolerance). This capacitor est located in C22 on the board for example.
Alternatives found : 102K (+-10% tolerance), or 102M 250V.
Are these alternatives any good ?
Quoted from jejepinside:I try to repair my System 11 MPU board : I can't find new 102M capacitors 50V (M is for +-20% tolerance). This capacitor est located in C22 on the board for example.
Alternatives found : 102K (+-10% tolerance), or 102M 250V.
Are these alternatives any good ?
Smaller tolerance or higher voltage rating is always fine
Both can be used
Quoted from harig:Smaller tolerance or higher voltage rating is always fine
Both can be used
You don't think tolerance was desired by Williams? Example : to obtain 120 µF, a 100 µF capacitor is used with a tolerance of 20%.
Quoted from jejepinside:You don't think tolerance was desired by Williams? Example : to obtain 120 µF, a 100 µF capacitor is used with a tolerance of 20%.
No it is not done like that
The next part from the same patch could then only have as low as 80 uF and still meet the +-20% tolerance range
If they wanted a 120uF cap (uncommon value though) they would have specd accordingly
Quoted from harig:No it is not done like that
The next part from the same patch could then only have as low as 80 uF and still meet the +-20% tolerance range
If they wanted a 120uF cap (uncommon value though) they would have specd accordingly
Thank you !
Thought I would share a tool I recently picked up. I don't like dealing with analog boards. This is typically a sound board or a board that includes a sound section. The last sound board issue I had on a reproduction board required me to go visit a friend who has an oscilloscope. The last time I used an oscilloscope was first year university. It was a compulsory physics course as part of basic science.
I recently took in a WPC-89 pre-DCS Sound board for repair. No sound output. Rather than spending a lot of time at my buddy's place using his oscilloscope, or borrowing it (bringing it over to my place), I decided to purchase one. I did a little bit of reading on "desktop" oscilloscopes but stumbled on a "portable" (handheld) oscilloscope. It's a simple tool but also has a few options (different models). The model I purchased includes a DMM and a wave generator.
This afternoon, I used the tool to figure out what was wrong on the board. It turns out that I could've deduced the problem by lack of signals using a logic probe but I would not have been able to prove that the cause was the actual cause. Having the oscilloscope gave me the conclusive evidence to declare it was the problem before replacing the part and fixing the problem.
So what was the problem? There was no clock from the 8MHz oscillator. No data bus signals pulsing. Reset line was high so everything should be working. All the ICs were verified good in another board. I was able to see the absence of the E and Q signals. Those signals come out of a flip-flop and I was able to see the lack of input to those flip-flops. The input comes from the oscillator. You can't see the clock signals using a logic probe. It showed as a sine wave and the frequency is so high that a logic probe or DMM can't really visualize it.
It was quite satisfying knowing I proved the cause and fixed it with a single fix rather than shotgun multiple components. The nice part about this tool is that it includes a DMM. I haven't used the DMM but this might now be the tool of choice I bring with me when I go on site visits to fix things or to the NW Pinball and Arcade Show.
This is NOT an endorsement of or advertisement for the product. It is a description of my initial experience with it. I strongly recommend anyone interested in tools do their own reading and research and draw their own conclusions.
Someone just turned me on to your work as I was looking for some replacement PinBot stuff. Do you sell fully populated replacements for the main MPU and the auxiliary sound board on pinbot? I am not quite understanding the CPU and ROM chips need to me migrated?
Quoted from Aniraf:Do you sell fully populated replacements for the main MPU and the auxiliary sound board on pinbot? I am not quite understanding the CPU and ROM chips need to me migrated?
Please use PM. That allows this thread to stay closer to discussion of the boards and technical aspects of them. Thanks!
Quoted from DumbAss:Thought I would share a tool I recently picked up. I don't like dealing with analog boards. This is typically a sound board or a board that includes a sound section. The last sound board issue I had on a reproduction board required me to go visit a friend who has an oscilloscope. The last time I used an oscilloscope was first year university. It was a compulsory physics course as part of basic science.
I recently took in a WPC-89 pre-DCS Sound board for repair. No sound output. Rather than spending a lot of time at my buddy's place using his oscilloscope, or borrowing it (bringing it over to my place), I decided to purchase one. I did a little bit of reading on "desktop" oscilloscopes but stumbled on a "portable" (handheld) oscilloscope. It's a simple tool but also has a few options (different models). The model I purchased includes a DMM and a wave generator.
This afternoon, I used the tool to figure out what was wrong on the board. It turns out that I could've deduced the problem by lack of signals using a logic probe but I would not have been able to prove that the cause was the actual cause. Having the oscilloscope gave me the conclusive evidence to declare it was the problem before replacing the part and fixing the problem.
So what was the problem? There was no clock from the 8MHz oscillator. No data bus signals pulsing. Reset line was high so everything should be working. All the ICs were verified good in another board. I was able to see the absence of the E and Q signals. Those signals come out of a flip-flop and I was able to see the lack of input to those flip-flops. The input comes from the oscillator. You can't see the clock signals using a logic probe. It showed as a sine wave and the frequency is so high that a logic probe or DMM can't really visualize it.
It was quite satisfying knowing I proved the cause and fixed it with a single fix rather than shotgun multiple components. The nice part about this tool is that it includes a DMM. I haven't used the DMM but this might now be the tool of choice I bring with me when I go on site visits to fix things or to the NW Pinball and Arcade Show.
[quoted image]
This is NOT an endorsement of or advertisement for the product. It is a description of my initial experience with it. I strongly recommend anyone interested in tools do their own reading and research and draw their own conclusions.
Nice tool if you don't have a big bench O-scope.
I have some (more) planned time off that was originally unplanned. Some things in life pop up and you have to take those opportunities when they arise. Consequently, I shifted my calendar around. I had kept February open and filled March and April for board building. Having now completely consumed March as time off, I decided to shift February tasks to April and compress March and April board building into February. Basically, it's a squeeze of all commitments into 50% of the time. Work double time. That's all I've been doing (other than the occasional service call that had been setup a while ago).
Speaking of service calls, I went to service an IJ and when doing the final testing, I noticed the POA was awarding things inappropriately. To cut a long story short, I found that multiple switches in the POA were triggering but only when the shooter lane switch was closed. I thought it might have been a problem with the POA wiring and that I would have to pull the whole thing out to look at it. Inspecting the shooter lane switch showed I was wrong. When I initially looked at the switch wiring, it looked correct to me. I was only inspecting the WHT-GRY on lug through diode to banded end on lug to center lug to GRN-GRY. After assessing it was correctly wired, I still couldn't figure out what was wrong so I went to inspect it again. Then it was obvious. It is an example of "your brain sees what it wants to see". This applies to reading the written word but it also applies to visually inspecting things.
ij_shooter_lane_switch_wiring.jpg
I fixed the error but it was interesting to note that the game played perfectly fine until the multiple switch trigger in the POA. Unless you know how the POA is supposed to work, you'd think things were still normal. The POA multiple switch triggering only manifest when the shooter lane was closed so it only manifest when in 2-ball or 3-ball multi-ball. I guess ignorance can really be bliss. The pinball flips so who cares? 
What I've really been doing is grinding on board building to compress normally paced building into half the time. When I do that, I go faster and sometimes get ahead of myself. I don't shortcut verification. Verification of boards with digital logic or analog components must pass 100% before a board goes out. I normally don't test simple connection boards. I do visually inspect them though. With this time build time compression, I find I still make mistakes.
The switch level/edge test was not registering anything in row 8. I thought it might've been an IC issue so I swapped some ICs and tested them. Problem persisted. Took the board off the bench rig to visually inspect it. There's the problem. It's so obvious.
I had made some revisions to the System 11 sound (music) board to support Pinbot and F-14. The changes were minor and probably not important but I wanted the board to reflect the Williams schematic to eliminate the potential for incompatibilities. I have found that most of these System 11 sound boards do not have their CVSD (HC-55536) channel populated. It's clear that the board generates mostly synthesized music through the YM2151/YM3012 combination. The DAC is there for things like the "bong" or other supporting sounds. Since System 11C uses this board exclusively for sound, all boards for System 11C are fully populated.
What I discovered while verifying the revisions to the board is that the OEM Pinbot board does not have the CVSD channel populated and Pinbot sound/music passes. The OEM F-14 board does have the CVSD channel populated and actually requires it. Sound/music test 15 uses the CVSD to generate the effect. I installed a HC3-55536 (plastic) and it didn't produce any sound effect. I tried the ROMs in my first board that I installed a HC1-55536 (ceramic) in and it worked. Weird. I thought all HC-55536 ICs are the same. Clearly, this is not the case. Every HC3 I put it failed to produce anything for sound 15. I put in another HC1 and it worked. I will chalk this up to the "mystery of pinball" and leave it at that. It works with HC1 and all F-14 boards will get the HC1 instead of the HC3.
A week of WPC-89 Power Driver boards and then I'll be away for the month of March. These boards will finish the original March and April commitments.
Quoted from DumbAss:It is an example of "your brain sees what it wants to see".
Also an example of when a wire comes off. Stick it back on anywhere.
I could see an inexperienced person thinking no way that loose white wire with a gray stripe goes back on the other white wire with a gray stripe. I'll stick it on the lug without a wire.
LTG : )
I'm starting in earnest with the DMD boards I bought last year DumbAss .
Never done SMD work before, but I think I'm getting the hang of it.
20230220_152534 (resized).jpg20230220_152538 (resized).jpg20230220_152720 (resized).jpg20230220_152822 (resized).jpg20230220_152956 (resized).jpg20230220_153008 (resized).jpg20230220_173234 (resized).jpgQuoted from pb456:I'm starting in earnest with the DMD boards I bought last year DumbAss .
Never done SMD work before, but I think I'm getting the hang of it.[quoted image][quoted image][quoted image][quoted image][quoted image][quoted image][quoted image]
Instead of paste you can use regular solder. Just put a bit on one pad then warm it as you set the part down with tweezers. If it isn't aligned perfectly just warm the solder and move. Once placed solder the other end. I use this method all the time and it works well.
Quoted from Robotworkshop:Instead of paste you can use regular solder. Just put a bit on one pad then warm it as you set the part down with tweezers. If it isn't aligned perfectly just warm the solder and move. Once placed solder the other end. I use this method all the time and it works well.
Yeah before I got the hot air station, I tried with a JBC C210 clone station, and I got pretty good results like what you mention.
Where I was goofing up when I first started was using too much solder paste. Watched a few YT vids and took some bad advice.
The MGC Chem solder paste has a very fine tip with a curve that is just a pain in the $!@# to push, but does a better job dispensing the proper amount.
That and 340C temp/~50 cfm does a good job, and I can get a good quantity of parts affixed in short order once I get set up.
Hi all!
Stumbled accross this post when a buddy of mine mentioned it the other day... Loving all the work Victor and everyone is doing on these!
Just wondering - Is there a Central "BOM" location at all where they are stored so i can take a look at costs accross this side of the pond??
Looking at putting together WPC and Sys11 Bench/Portable testers (all being well) and am more concerned currently about the availability side of things on parts needed to assemble... Seems that either part costs have trebled, or some are just downright not available (at least not without importing from multiple places around the globe!). Neither an issue - just wanting know what i'm letting myself in for!
Thanks!
Paul
TIP102 and 36C continue to be in short supply (from reputable sources). I'm still waiting for my 36C's that I ordered from mouser.com back in july of last year, but did finally receive the 102's back in early january.
Quoted from pb456:Never done SMD work before, but I think I'm getting the hang of it.
Quoted from Robotworkshop:Instead of paste you can use regular solder. Just put a bit on one pad then warm it as you set the part down with tweezers. If it isn't aligned perfectly just warm the solder and move. Once placed solder the other end. I use this method all the time and it works well.
Exactly what I do. Use the finest gauge solder you can get. You really only need a tiny amount. To make the joint look like a factory did it then use flux afterward to "smooth" it out. Use water soluble solder and water soluble flux so that it cleans off easily.
I use:
There are many other manufacturers (products) that you can use.
Final (big) update for a while. I'm heading off for some more time off. I have to take the opportunities that present when they present. The number of these will only get fewer as time goes by. I worked double time to get everything I could get out before leaving. I had to leave some time leftover for my own tasks as well as travel preparations.
I revised these double ROM boards. Two for System 11 (identical function) and two for WPC (identical function).
There is a switch that enables the LED if you want visual feedback of the state. The other switch selects between the different ROMs with an LED to indicate state. I built ROMs for Earthshaker and Whirlwind as a test of both systems. The display strobes which is why it is impossible to get an accurate image of the game name but it should be obvious.
boards_272 (resized).jpgboards_273 (resized).jpgboards_274 (resized).jpgboards_275 (resized).jpg
Same deal for WPC-89. I used Twilight Zone and Indiana Jones as a test of this system.
boards_276 (resized).jpgboards_277 (resized).jpg
I had a request for the 8-Driver Auxiliary board and I had been sitting on the completed design for some time but never bothered to fabricate it.
I have been fighting this board for some time. First revision had physically incorrect alignment. Second revision had an adjustment for the alignment but I went too far in the adjustment. It also had a modification for the opto placement. This modification was to the mechanism used to make the 3D aspect of the board work. This is, without a doubt, the weirdest 3D puzzle I have had to solve in the board fabrication world. Third revision I halved the adjustment. Perfect. Free rotation.
boards_279 (resized).jpgboards_280 (resized).jpg
I do take feedback and changed the lighting to allow a "wide viewing angle" or "superflux" or "Pirahna" LED. It is also possible to install a regular "T-1 3/4" or 5mm LED if you so desire. There is a potentiometer to allow brightness adjust.
boards_281 (resized).jpgboards_282 (resized).jpg
And finally, everything came together because I used this test to verify the 8-Driver Auxiliary board as well as the Twilight Zone Clock board. I had to build the wiring and the power cabling to get this to work. Took a little bit of planning but everything works. It even draws from the GI (AC) and verifies the DC rectification on the Clock board. Yes. You can run the clock test on the bench.
EDIT: Re-gained access to the YouTube account so I posted the video.
While away, I will still be connected but the same caveat applies as previously. Expect slower response times. As I track everything, feel free to continue with inquiries. I will keep my work requests up to date. As a service announcement, the build queue is currently months out. If anyone does have an interest for a big board, expect a wait. Small boards or bare boards can be dealt with in short order. I have reserved April and May to play catch up on things neglected for months to years while I have been doing pretty much nothing but build boards. I might be able to use the down time during these months to figure out why the System 3-7 board doesn't work and try to get it working as well.
Victor: Thanks for all your hard work and help for the pinball community! Have a great and safe trip.
Quoted from DumbAss:Final (big) update for a while. I'm heading off for some more time off. I have to take the opportunities that present when they present. The number of these will only get fewer as time goes by. I worked double time to get everything I could get out before leaving. I had to leave some time leftover for my own tasks as well as travel preparations.
I revised these double ROM boards. Two for System 11 (identical function) and two for WPC (identical function).One System 11 using 1x 27256 and 1x 27512.
The other System 11 using 1x 27C1001.
Both these boards require a 6264 for the SRAM (or NVRAM) to keep the settings separate.
One WPC-89 using 1x 27C801 for OEM boards.
One WPC-89 using 1x 27C801 for my reproduction boards.
Both of these boards require a 62256 for the SRAM (or NVRAM) to keep the settings separate.
Great work! I am just curious - what is the difference between the ROM boards for OEM vs. your WPC boards?
The spacing on the rom vs. nvram socket most likely. I know he also indicated the system 11 one is not going to work on the rottendog board.
I'm looking forward to getting one for my pinbot to switch between pembot and pinbot without losing all the settings each time.
I had a super hacked up interconnect board in my Police Force. An email to Victor, and a short time later I received an amazingly nice replacement board with the correct resistors and everything.
I can't say enough good things about Victor and his boards. They are fantastic quality, have ALWAYS been correct out of the box for me, and his communication and customer service are at a level I have never seen before (in a good way). I have NO idea how he can get done what he gets done, but when he says he needs to take time for himself, he has certainly earned it.
Victor - Enjoy your time for you, get refreshed and make your days memorable. We'll all be here when you return 
Chris
IMG_7997 (resized).jpgIMG_7998 (resized).jpgIMG_7999 (resized).jpgQuoted from snakesnsparklers:Great work! I am just curious - what is the difference between the ROM boards for OEM vs. your WPC boards?
Quoted from slochar:The spacing on the rom vs. nvram socket most likely.
This is correct. The OEM WPC-89 CPU board has different spacing from my board. The OEM board is tighter (closer). I do try to keep as close to original spacing but I also like to space things out evenly in the given space. Even spacing produces a better cosmetic finish as well as not requiring tight soldering.
Quoted from slochar:I know he also indicated the system 11 one is not going to work on the rottendog board.
I should qualify that it may not work on the MPU9211, MPU011C or MPU004. I don't have those boards to measure and the OEM board is far more common out in the wild.
I can measure the Rd board if that helps. What do you need the distance between the RAM and the two rom sockets?
Quoted from slochar:I can measure the Rd board if that helps. What do you need the distance between the RAM and the two rom sockets?
Measurements are pin to pin for center of IC to center of IC and one side of IC to other side of adjacent IC. The measurements are identical for all three ICs on the OEM board.
You can convert from metric to SAE using 1" = 25.4mm
I use center of IC to center of IC as it is consistent between all components.
Are these boards basically selectable pass-throughs or something different? I ask because I have a Rush 2049 that uses 32- and 28-pin chips for the boot ROM and security chips that I could essentially use these for to switch between the different versions of the game(Standard, Tournament Edition, and Special Edition, the last two use the same chipset). The only real difference would be that the chips are far apart from each other and could not live on the same board. I'd have to use two boards, essentially, one for the boot ROM and one for the security chip.
Feel free to explain it to me like I'm 8 years old, too. I'm not an idiot, I just haven't been exposed too much to the world of custom board work. It truly fascinates me.
Quoted from DumbAss:Final (big) update for a while. I'm heading off for some more time off. I have to take the opportunities that present when they present. The number of these will only get fewer as time goes by. I worked double time to get everything I could get out before leaving. I had to leave some time leftover for my own tasks as well as travel preparations.
I revised these double ROM boards. Two for System 11 (identical function) and two for WPC (identical function).One System 11 using 1x 27256 and 1x 27512.
The other System 11 using 1x 27C1001.
Both these boards require a 6264 for the SRAM (or NVRAM) to keep the settings separate.
One WPC-89 using 1x 27C801 for OEM boards.
One WPC-89 using 1x 27C801 for my reproduction boards.
Both of these boards require a 62256 for the SRAM (or NVRAM) to keep the settings separate.[quoted image]
There is a switch that enables the LED if you want visual feedback of the state. The other switch selects between the different ROMs with an LED to indicate state. I built ROMs for Earthshaker and Whirlwind as a test of both systems. The display strobes which is why it is impossible to get an accurate image of the game name but it should be obvious.
[quoted image][quoted image][quoted image][quoted image]
Same deal for WPC-89. I used Twilight Zone and Indiana Jones as a test of this system.
[quoted image][quoted image]
I had a request for the 8-Driver Auxiliary board and I had been sitting on the completed design for some time but never bothered to fabricate it.
[quoted image]
I have been fighting this board for some time. First revision had physically incorrect alignment. Second revision had an adjustment for the alignment but I went too far in the adjustment. It also had a modification for the opto placement. This modification was to the mechanism used to make the 3D aspect of the board work. This is, without a doubt, the weirdest 3D puzzle I have had to solve in the board fabrication world. Third revision I halved the adjustment. Perfect. Free rotation.
[quoted image][quoted image]
I do take feedback and changed the lighting to allow a "wide viewing angle" or "superflux" or "Pirahna" LED. It is also possible to install a regular "T-1 3/4" or 5mm LED if you so desire. There is a potentiometer to allow brightness adjust.
[quoted image][quoted image]
And finally, everything came together because I used this test to verify the 8-Driver Auxiliary board as well as the Twilight Zone Clock board. I had to build the wiring and the power cabling to get this to work. Took a little bit of planning but everything works. It even draws from the GI (AC) and verifies the DC rectification on the Clock board. Yes. You can run the clock test on the bench.EDIT: Re-gained access to the YouTube account so I posted the video.
[quoted image]
While away, I will still be connected but the same caveat applies as previously. Expect slower response times. As I track everything, feel free to continue with inquiries. I will keep my work requests up to date. As a service announcement, the build queue is currently months out. If anyone does have an interest for a big board, expect a wait. Small boards or bare boards can be dealt with in short order. I have reserved April and May to play catch up on things neglected for months to years while I have been doing pretty much nothing but build boards. I might be able to use the down time during these months to figure out why the System 3-7 board doesn't work and try to get it working as well.
DumbAss I would be interested in purchasing at least two of the WPC-89 Dual Software / Single ROM boards (for Williams Demolition Man)...however I currently have Pinsound RocketCPU replacement MPU boards (I would think they would be fine as they are based on OEM board for the most part. I have OEM WPC89 boards as well though). I wouldn't even mine bare board and/or bare board only (with component part number Bill of Material) or with loose components and I will solder myself.
Quoted from Miguel351:Are these boards basically selectable pass-throughs or something different? I ask because I have a Rush 2049 that uses 32- and 28-pin chips for the boot ROM and security chips that I could essentially use these for to switch between the different versions of the game(Standard, Tournament Edition, and Special Edition, the last two use the same chipset).
The dual ROM board has one or two ROMs that store two separate ROM images that can be switched. The main (CPU) board consuming the ROM only sees the selected ROM. You could call it "selectable pass-through" but it really just maps the select switch state to the highest address bit of the ROM (or RAM).
I don't know what the requirements are for Rush 2049. I have ZERO experience with vids. I am sure they are very similar in that these are simple 80s/90s era "computers" typically with an 8-bit data bus and 16-bit address bus. They aren't complicated but to do things you need to understand how the system is put together. I just don't have any experience with them.
Quoted from Sensei:however I currently have Pinsound RocketCPU replacement MPU boards (I would think they would be fine as they are based on OEM board for the most part. I have OEM WPC89 boards as well though).
Whether the dual ROM boards are compatible will depend on the IC placement and spacing. Since I don't have the board in question and have never seen that physical board, I can't say if there is a compatibility issue or not. The spacing measurements I have for the OEM board are:
I haven't tested the OEM board in an OEM CPU board. I have a White Water CPU board that I pulled the SRAM off but haven't put the IC socket in yet. Once I get the socket in I can test the spacing measurements. A difference of 0.635mm in either direction probably won't make much of a difference between two ICs. However, that kind of error will compound over multiple ICs. It doesn't apply here.
I wanted to post a public thank you to j_m_ for the excellent work breaking the board images down into individual components. Assembling the system 11 board last night and had my laptop open with the images. Such great work. If you any of you haven't downloaded them yet, search this thread, he did an incredible job.
Quoted from newovad:If using the jumper, does it matter if it goes in the middle?
Nope. The outline allows for the 27 Ohm 5W resistor installed using the outer pads/holes and the 0 Ohm jumper installed using the inner pads/holes. You can install the 0 Ohm jumper using the outer pads/holes (as you have) if desired.
They are electrically connected. You can see this in the image for the front. The connection if unseen will be on the back.
Quoted from DumbAss:The dual ROM board has one or two ROMs that store two separate ROM images that can be switched. The main (CPU) board consuming the ROM only sees the selected ROM. You could call it "selectable pass-through" but it really just maps the select switch state to the highest address bit of the ROM (or RAM).
I don't know what the requirements are for Rush 2049. I have ZERO experience with vids. I am sure they are very similar in that these are simple 80s/90s era "computers" typically with an 8-bit data bus and 16-bit address bus. They aren't complicated but to do things you need to understand how the system is put together. I just don't have any experience with them.
When I say pass-through, I just mean that the traces are straight into their corresponding pins placements and there are no crossovers, intermediate translation chips, or power being taken from one pin location and applied to a different one for a specific chip. So essentially, these sockets and traces are dumb is what I'm getting. I guess the thing for me to check next is chip size/width to see if they're the same. I guess I can measure them against my TZ's MPU board's chips.
I think I'm just getting excited because your boards are kind of a proof of concept for my thought of something I've wanted to make for my Rush 2049 for a very long time: I'd like to make a front mounted power switch paired with a 3-position rotary switch where the user can switch between the three versions of the game; standard, Tournament Edition, and Special Edition. The rotary switch would control which ROM chip, Security Chip, and Hard Drive(or CF converted drive) would be booted from when the game powers back up.
I know it's possible, and I can think of ways to get it done, but I don't know if they're truly electronically feasible. You making these kinds of boards proves my idea is possible in a very small way and that's exciting to me. Even if the chips/drive selection method had to change to get it to work properly and safely, I'd be ok with that. Just knowing that it might eventually come to life someday, is really cool to think about.
Quoted from Miguel351:When I say pass-through, I just mean that the traces are straight into their corresponding pins placements and there are no crossovers, intermediate translation chips, or power being taken from one pin location and applied to a different one for a specific chip. So essentially, these sockets and traces are dumb is what I'm getting.
If this is what you're asking about then most of the signals are passed straight through. Some of the signals may be used for other logic decisions but the important address and data bus bits cannot be modified if you want the system to work properly.
There are many ways to solve this problem. You can use the two solutions that I have implemented. The first example is for System 11 where U26 and U27 are required. The two U26 images are combined into a ROM that is double the size. Same deal for U27. Then, there is a signal that selects which image. The second example combines both those dual U26 and U27 ROMs into a single ROM that is quadruple the size. Another solution is to have four ROMs and just the correct pairs. More ROMs = more board space required. There is right or wrong.
I guess the thing for me to check next is chip size/width to see if they're the same. I guess I can measure them against my TZ's MPU board's chips.
All (WMS) pinball ROMs are "wide". That is they are 15.24mm wide (not 7.62mm wide which is narrow). Their available storage size will dictate how many pins they have. Some ROMs also have 16 data bits but all (WMS) pinball ROMs are 8 data bits. I assume it depends on the processor being used.
Quoted from Miguel351:The rotary switch would control which ROM chip, Security Chip, and Hard Drive(or CF converted drive) would be booted from when the game powers back up.
Ah. Vids use HDDs. I assume they are IDE from those days. That's going to be an interesting thing (for you) to think about.
The issue with implementing this kind of feature is that you need to establish what the requirements are and how to work that into a selectable solution.
Quoted from Kawydud:I wanted to post a public thank you to j_m_ for the excellent work breaking the board images down into individual components. Assembling the system 11 board last night and had my laptop open with the images. Such great work. If you any of you haven't downloaded them yet, search this thread, he did an incredible job.
thanks for the kind words. I'm glad to be able to help out the community in my own little way. victor's done a fantastic job on these boards and it's so much nicer to work with brand new materials and components.
Quoted from DumbAss:Ah. Vids use HDDs. I assume they are IDE from those days. That's going to be an interesting thing (for you) to think about.
The issue with implementing this kind of feature is that you need to establish what the requirements are and how to work that into a selectable solution.
Well, it depends on the era the game was made. Almost all vids up until the mid 90's were all chips on custom boards. Then, a number of games went to a custom board set with some normal computer parts. Atari/Midway did this for a number of games where they had a custom board set, but it had a contemporary standard issue graphics card and hard drive. Then, after that, a majority of games were powered by just straight PC's with security dongles, special cards, and other things to make them more proprietary and hard to crack.
I think the hard drive selection device is the easy part of it all. There are selectors already made these days, they're just in a different configuration than what I'd like to see. The hard part for me to figure out is the remote ROM selection.
Quoted from DumbAss:[quoted image]
Measurements are pin to pin for center of IC to center of IC and one side of IC to other side of adjacent IC. The measurements are identical for all three ICs on the OEM board.
You can convert from metric to SAE using 1" = 25.4mm
I use center of IC to center of IC as it is consistent between all components.
So unfortunately the rottendog boards have the RAM below instead of above so the rom board won't work.... I thought maybe it was completely flipped but all the notches go to the right on both wms and Rd so no go.
Quoted from DumbAss:Exactly what I do. Use the finest gauge solder you can get. You really only need a tiny amount. To make the joint look like a factory did it then use flux afterward to "smooth" it out. Use water soluble solder and water soluble flux so that it cleans off easily.
I use:Kester 24-6337-6422
Kester 83-1097-2331There are many other manufacturers (products) that you can use.
As much as I've tried, I can never get it to be as accurate, non-blobby, or as quick as I can with hot air and solder paste. I thought my hands were pretty steady. Until I tried with the 1206 packaged SMDs!
I use a set of ESD safe tweezers to push/pull the ICs/components onto the pads as I heat them up with the hot air station. The trick, for me, is getting just the right amount of solder paste on each pad.
I use some MCG (?) chem solder paste that has a very fine needle like applicator on the end of the tube. Hard as heck to squeeze but if patient, can deliver just the right amount of solder paste onto each pad and then 'wipe' it off a the end of the pad. Has taken quite a few tries to get it down.
Denatured alcohol seems to to a good job at cleaning the board, and I check it out with 3.5x magnifiers afterward for errant crud or solder balls.
Waiting on a set of
Multilayer Ceramic Capacitors MLCC - SMD/SMT 1206 50VDC 0.1uF 10% SOFT X7R
To finish it up - I got the wrong sized ones when I did my mouser order, and can't get a solder bridge to fill the gap with the smaller parts.
Adventures in pinball
Quoted from pb456:As much as I've tried, I can never get it to be as accurate, non-blobby, or as quick as I can with hot air and solder paste. I thought my hands were pretty steady. Until I tried with the 1206 packaged SMDs!
The nice part about hot air is that the entire area is heated and the components can be moved. If you have enough flux, the component will just fall into place when the solder is molten.
Quoted from pb456:Waiting on a set of
Multilayer Ceramic Capacitors MLCC - SMD/SMT 1206 50VDC 0.1uF 10% SOFT X7R
To finish it up - I got the wrong sized ones when I did my mouser order, and can't get a solder bridge to fill the gap with the smaller parts.
If you think 1206 is small, that's actually one of the larger sizes available. It's the smallest I will go because I can't do anything smaller than that without it being a fight. I have some smaller resistor and capacitor networks that I have used for prototype boards. I have also used 0805 for zero Ohm jumpers. I think that's close enough you can solder blob a jumper but I also purchased some 0 Ohm jumpers.
The Twilight Zone Clock board (above) uses 1206 capacitors and resistors.
I try to stay busy. I have a bunch of things that I can deal with while away and am using the down time to finish some of these tasks. I thought the first of them would take me longer than it did so I am happy. It's the first step to get to product. I decided to revise one of the components used so I will need to make another prototype board to make sure everything works before proceeding to the final product. I hope this will provide another option for those looking for the solution but not wanting to have to purchase from outside the US or wait for shipping from outside the US.
Step 1 - understand what you are dealing with.
I have never used a PIC before. In fact, I didn't know what a PIC was until I encountered it with WPC-S. I always thought PIC meant "programming interrupt controller" from the early days of desktop PCs. Apparently, it now stands for "programmable intelligent computer". I don't think the PIC12F508 would qualify as "intelligent computer" given it has a maximum instruction space of 512 instructions and 25 bytes of RAM.
I spent a fair amount of time reading the documentation provided by the manufacturer and some time with their (free) development environment. It's nice that a development environment is provided. A C compiler makes writing code for this processor much easier. The instruction set architecture (ISA) is very unintuitive and different from other ISAs I have dealt with in the past.
Step 2 - build a board that allows exercising the GPIO pins and a board that allows testing potential software that performs the desired task.
The simple program to set GPIOs state or respond to GPIO state works. I had a little issue with GPIO2. I couldn't get it to work and I read the documentation and couldn't figure it out. A little searching showed a (different) forum post from someone that solved the problem. I was a little disappointed in the post because it didn't explain why it worked and had the tone of "it's so obvious why you need to do this to get it to work". I still don't fully understand why it's required even after reading the document again and again and again. Some things, I just learn to accept. Maybe someday in the future I might understand it.
I already built the verification boards to test the "dumb" version of the board. They can be used to test any potential "smart" version of the board.
I bought an IC from Flipprojets to use as a control. You apply power and the software calibrates the DS1803 (digital potentiometer) and the board senses either the narrow or wide depending on how it was calibrated and the capacitor configuration. Works like a charm! Great. So I know my prototype board works. Now I just need to write my own software.
After much reading (previously), the previous understanding of the GPIOs and how to program them, I got things working. It probably took about 30 or so iterations. Some frustration in the middle but the software works. I had to write my own I2C controller. The same software works for both narrow and wide configuration.
After completing this, I decided to use a different digital potentiometer to try to recover some board space. The DS1803-010 is dual pot (10k) and 16 pins. There is a Microchip product that is a single pot (10k) and 8 pins. It uses SPI rather than I2C so I now have to write a SPI controller. SPI also requires more pins than I2C. Good thing there's an unassigned pin on the PIC12F508 that I can use for the ~CS signal. This board is now paused waiting for fabrication and component acquisition. It will probably resume in April.
The final revelation is that I can now use this for some more "smart" functionality in other areas that could use intelligence and improvement. I just need to find the time to do these things.
Add another to the list of people that will buy this
I'd also like to use this for my Demolition Man WPC-89 to switch between the latest v7 ROM and James Cardona's Demolition Man 1.01 ROM. I have a ColorDMD LCD screen and a pinsound if that matters.
Great work happeneing here
File this under, "things that make me happy!" started the Sytem 9 build in Feb. Still waiting on some good RAM/NV, but used a beater chip tonight to see if we get the coveted ZERO! Yup we did! I'll admit I was pretty nervous flipping that switch. Kudos to Victor for tolerating my questions and building a board that will keep this Shuttle going for a long time!
dasys9 (resized).jpgThis post can be considered <TL;DR> by some. Therefore, if the content is of no interest to you then please stop reading.
Another small update that will probably make a few people happy. It definitely makes me happy because REV-00 of this board was fabricated in January 2022 (over one year ago from March 2023). I found a fundamental problem with the RAM addressing and a few other small things that got corrected in REV-01. That board was fabricated in March 2022. After some initial testing, I still couldn't get the coveted "0" to disappear and the BLANKING LED to illuminate.
Fast forward to January 2023 (early this year) and I finished writing a 6800 disassembler and a special purpose 6800 assembler. I spent some time looking at the post-on-self-test (POST) code as well as the diagnostic (NMI) code. See post https://pinside.com/pinball/forum/topic/dumbass-test-and-reproduction-pcbs/page/23#post-7327660 and post https://pinside.com/pinball/forum/topic/dumbass-test-and-reproduction-pcbs/page/23#post-7357889 for more details to refresh your memory. Those tests provided the baseline to know the address space is good. Valid RAM, CMOS and ROM addressing. Valid PIA addressing. The only thing missing is the IRQ (interrupt request).
That was one of the goals this time around. Write some code to verify that IRQs are being generated and responded to correctly. I only have a few EPROMs with me and I had used them verifying the board. Turns out the power supply to my board was flakey but I had "burned" the blank EPROMs available to me. I was looking for a local eraser and secured the loan of one (pending pickup). The best solutions that I have ever come across usually happen when I am lying in bed staring at the ceiling in the darkness. It almost never happens on demand. The creativity is 100% unpredictable.
I did stumble across a solution to my problem. When I generate the image to write (for 6800), I usually pad the binary with $DD which is not strictly an instruction but is affectionately known as "HCF". If you're interested, search for "HCF instruction". Doing this though causes the write to be a little longer since it has to write the 0 bits. So I replaced all the $DD bytes with $FF bytes. Turns out, this means I can use other areas of the EPROM if I carefully place the code where it needs to be.
I put the code at $FF00. I put the IRQ vector at $FFF6 and the reset vector at $FF00. The code at $FFF6 is an infinite loop (BRA *-2). If I put my new code at $FE00 and the ISR at $FE80, I can change the reset vector from $FF00 to $FE00 and this is a single 1 -> 0 conversion. A valid write without erasing. I can change the IRQ vector from $FFF6 to $FEF0 (could have left it at $FEF6) and put a BRA instruction to $FE80 where the ISR is located. This is a valid write that only changes 1 -> 0. Away we go. I can continue to test without erasing!
So here is the code that I wrote to test the IRQ. The initialization and loop simply sets up the system and does a slow blink of LED1 at a rate of about 1 second ON and 1 second OFF. The SEI and CLI surrounding the PIA port operation is required to prevent the ISR interrupting that code (re-entrancy) and altering the state of the PIA port while it is being manipulated by the loop code. Don't ask me why I know this happens (no, it's not because it happened with this test - I had been bitten by re-entrancy before so I have learned to spot it and avoid it).
This is the ISR. The first instruction uses a 256 execution count to delay the modification of the LED2 state. This effectively makes the duty cycle of LED2 256 milliseconds. The LED is turned on and off by just taking the low bit. The code blindly increments the state byte that causes the bit to change from 0 to 1 and back to 0.
Timeline:
Board with correction, running Black Knight software. Power on shows "0" and then goes blank. Correct behavior. BLANKING LED is ON (software executing). Oscilloscope probe connected to 2800.PA2. This signal is fed into the 555 timer as a "heartbeat". The oscilloscope shows the square wave of execution.
I guess I'm lucky. I have VERY high confidence the board is now correct from a basic code and logic execution standpoint. I need to revise the board (REV-02) to include the fix. For that board, I will either rob the components from the REV-00 or install some new ones and test out the rest of the board features. I think I crossed the major hurdle for this board. The software side is executing properly. Guess I don't need to loan the eraser any more!
Wanna join the discussion? Please sign in to reply to this topic.
Great to see you're enjoying Pinside! Did you know Pinside is able to run without any 3rd-party banners or ads, thanks to the support from our visitors? Please consider a donation to Pinside and get anext to your username to show for it! Or better yet, subscribe to Pinside+!
This page was printed from https://pinside.com/pinball/forum/topic/dumbass-test-and-reproduction-pcbs/page/24 and we tried optimising it for printing. Some page elements may have been deliberately hidden.
Scan the QR code on the left to jump to the URL this document was printed from.
Maryville, TN
Wiener Neustadt