...I believe mine do!

Finally getting around to letting koji know that bending the switch stack out a bit (so the first connection is broken more at the end of the plunger's action) seems to have worked. I do still get a good amount of hum, which is annoying, but it works! Thanks, @koji!

Some new questions:

1. the game has now started awarding score just for hitting the flipper. not sure what that's about! how to diagnose? totally new problem...

2. this might have to do with the previous issue in q1, but sometimes the game will just keep racking up points, for no apparent reason. I can't find a switch stuck on, and it only happens after playing a ball for a while. it'll stop once you drain. I think I might just be hitting something in the course of play that gets stuck on, and the machine is smart enough to know to STOP scoring it once the ball drains...but it doesn't just keep racking up points on the next ball right away. I think the next step is to go through and hit switch after switch to see if it starts the problem up, but does anyone have any experience/ideas on what might be happening there?

3. as mentioned above, I think the flippers feel weak. maybe it's just because I'm not used to this era of game, but I feel like there should be a bit more oomph. I've read Vid's guide, and I think I've adjusted things properly, but just to double-check, is there anything else I could use a meter on to see if maybe it's not getting enough voltage? not sure what to check, or what I'm looking for, or how...

4. does anyone know where to get SW-1A-133, which is the switch for the star rollovers? I can't find them anywhere...

5. because of the "constantly racking up score" issue, the score has rolled over on a game I "played"...however, the "highest score" is still a legit 800k score I've had when the game was playing more or less normally. does anyone know, does Flash not retain rolled scores? I've got a Rottendog MPU in there, as well as Pinscore LCD displays. I just find it strange that there's no indication that the score has rolled...

Any and all help appreciated, as always, thanks y'all!

how many points will help you narrow down what switch is adjusted too closely. Did you happen to just change any rubbers for new?

same thing, how many points will give you a hint as to what switch is repeating

Meter on DC voltage, red lead to the coil lugs, black lead to the ground strap in the head, you should get between 28-40 volts. If the flippers haven't been rebuilt might be time. Cabinet switches might need filing as well. Some people upgrade these to WPC style mechs if you do that you have to mod the upper right one's base to fit the lamps there.

Properly rebuilt ones aren't as strong as new games but they are plenty strong for the game itself. Usually there's a lot of slop in the mech and sometimes the wood holes in the PF are loose as well. Fill with hardwood dowel and glue, let dry, drill new pilot hole.

Order generic blades from either marco or pinball resource, and some contacts and build your own. No one sells "original" switches anymore.

Nope, they don't record that the HSTD rolled over, which led to an interesting thing back in the day called "nine-ing out" - where you tilt as you get as close to 1 million as possible (beating the old hstd) so that you get the 3 credits. System 7 has code to specifically handle this, and Bally games of the era can be set to award you the HSTD credits as well. Not so williams. 6 digits is all you get. There are rom mods to divide the score by 10, and also to put real 7 digit displays in there, but neither of those work with the Rottendog boards, which I understand is the original software, but something has been modified in it to have it work on their board. Details are sketchy on that.

I have had my Flash for years. Always felt the flippers felt weaker than my other games. This weekend I finally did a flipper rebuild.

My flippers still feel weaker than my other games. Bummer.

What was included in the flipper rebuild? If you are still using the original full metal bracket flippers, I suggest changing over to the newer style without the metal bracket or_ check our some of the posts abotu cross drilling these. You can easily shave off ~30-40% of the weight, and this GREATLY improves the speed and responsiveness.

In addition, make sure the EOS is right at the VERY end of stroke for max power.

I suppose the tell in Flash is that you can really crank the spinner. If not, then adjust. The Sys4-6 flippers are notoriously weak due to low voltage, but when tuned properly, it should not be particularly noticeable.

Thanks for the suggestion. Is there a pic or Vid post that shows what “right” should look like?

Typically, I think they say to gap at 1/4". So under the PF, manually move the flipper all the way up, and see how wide the switches get pushed away. Make sure that they only separate when the flipper is nearly fully engaged. You may experiment with a slightly lesser gap as well, just know what you are doing, as this is how coils are killed as well.

Made the adjustment. Such world of difference. Thanks!

Was trying to tune in my right sling and found this cap disconnected on the top switch .Anybody offhand know a good replacement. Ive got some small ceramic caps I used.on my silverball mania rated at .047 uf. Didn't know if those would work.
Thx for any help.

20200510_153512-1 (resized).jpg

22 uF cap and 100 ohm resistor are the components on the slings/pops.

Thx for quick response!!

I’m kind of new to this repair stuff.

Finishing up my flipper rebuild. When I got to the third flipper, bottom right flipper, the switch was a completely different switch than the other two, and of course different than the one provided. I couldn’t find a part number in the manual. Is this more complex switch the correct switch or was this a previous owner’s elaborate hack?

0C4BF78E-3C81-499A-B983-5DAE46FDB89F (resized).jpeg

This switch has the EOS for the lower right and powers the upper right when engaged.

This Marco part looks close to the original and cross references.

https://www.marcospecialties.com/control/keywordsearch?SEARCH_STRING=2010a

Factory switch:

Flash Flip (resized).JPG

Thank you for the quick feedback and part reference.

Strange Switch Behavior - Got my Flash up and running, but seem to be experiencing "incorrect " behavior with switches in a single row of the switch matrix.
2020-06-08_19-30-38 (resized).png
The switches do in fact register, but give the following behaviors:

6 - Left coin switch: Not used; game is on free play
14 - "3" Rollover: Also registers and lights the #4 Rollover
22 - Center Right Star Rollover: Scores double (either 200 or 2000)
30 - 3 Bank Left Drop Target: Scores the value of knocking all 3 targets down and resets the drop target bank
38 - 5 Bank Standup: Seems OK
46 - Right Special: Tilts the ball in play

At first I thought it may be a bad diode, but I tested each of the diodes on the switches and did not see any reading out of the ordinary (All were between .5 and .7). Has anyone experienced this type of behavior before, and if so what was the resolution?

Update:

After posting the previous thread, I noticed that the row 6 switches were exhibiting the behavior of the corresponding row 7 switch below the row 6 switch. Just checked the continuity of the pins for those rows in the driver board connector pins, and DID get continuity. Based on that, I'm assuming that is the reason for the strange behavior. Will remove this evening and check the backside of those connector pins and reflow if necessary.

Update #2 (Problem Solved):

It appears that Pins 3 (Switch Row #7) and 4 (Switch Row #6) of the 2J3 9 pin connector on the driver board must have been 'connected' via a solder blob somehow. They didn't appear to be 'connected' when viewing from the underside of the board, so all I can assume is that it was underneath the connector topside. Regardless, I replaced the 9 pin connector with a new one and the 'bad' switch behavior disappeared.

Wow. That is an esoteric problem. I hope that is it

Struggling with reliable scoring on Flash. Cleaned and polished all horseshoes and PCBs and find the targets don't score reliably when they drop - sometimes never scoring. They always reset correctly when all DTs are dropped. Tried replacing the horseshoes and PCBs from PinballLife and also replaced the target guides to make sure the PCB screws are tight. The scoring is still erratic and unreliable.

As a test, I added 0.047 uF capacitors across the middle contacts (two outer solder pads on the PCBs) and the drop targets score 100% of the time, but now they also score sometimes when the drop target bank resets.

In addition the 1-2-3-4 roll-overs and star rollovers don't score sometimes. Have disassembled the switch stacks, fixed the incorrect assembly so the correct contact faces are facing each other and polished and cleaned them and they still don't score sometimes. I see unreliable scoring with nearly every switch in the game.

Also looked at all 8 green wire strobes in the switch matrix and they all strobe from 5V to 0V with 40 uS pulses every 2mS.

Have restored dozens of games and never seen anything like this WRT unreliable scoring. MPU and Driver boards have new connectors, NVRAM, ROMs and were rebuilt by Allan D, the best guy in the Boston area for board repair. The boards look impeccable along with the wiring and connectors, including a new 40-pin inter board connector.

Has anyone seen unreliable scoring on their Flash and if so any ideas what may cause it?

Not sure what the fault may be with having all the scoring problems in different areas, sounds more like a driver board switch problem, but If your problem with the drop targets in particular is related to the horseshoe wiper contacts, the best way I have found to ensure reliability is the double up each horseshoe wiper. Basically have another blank horseshoe (with the riveted contacts removed) behind it. This doubles the tension to make a good contact.

In fact this is how they were meant to come out of the factory according to the drawings which show two contacts present per drop target. Also the two horesehoes wipers have different part numbers as one is blank & the other a regular horseshoe.

I also have tried capacitors - they don't work on these Williams games, only Bally's. They mess thing up & add extra scoring as you said.

My Flash I am currently restoring is pretty much all original & had two wipers per drop installed. I also bought a Flash drop target bank a few years ago & that too had 2 wipers per drop.

Can be a bit tricky to remove the contacts.

These have obviously not been cleaned up & adjusted, but this is the idea.

P1020938 (resized).JPGP1020939 (resized).JPG

I wonder if you could do a driver board mod/add on where you're driving the switch matrix with higher voltage (12 instead of 5) like WPC does (converting back to 5) to help the issues with the drops. I'm going through the drop thing as well with my time warp right now, it as flawless 5-6 years ago after adjusting the horseshoes and using contact enhancer on them (Deoxit) - every drop, every time, registered. Now after a couple years of not playing it (too many projects in front of the back row) it's just not working at all again (or maybe the deoxit has dried up).

The series reset is solid still. I thought maybe something in the intervening years has changed beyond the new boards that are made but I think that's it.... wasn't there a project going where an added switch did a one shot pulse to the board instead of depending on the horseshoes?

Thanks Joydivision that is excellent input. The Flash I am working on has both horseshoes installed. I should try tensioning them further, but am concerned they won't drop quickly (or reliably) if I go too far. I also used a thin film of PBR grease (Steve Young recommends it for EM style PCB/rivet contacts like this). Thanks for your feedback on the capacitors, it is really too bad they cause extra scoring as they solved the problem 100%.

slochar Interesting idea - Raising the voltage to 12V would help to eliminate false triggers like those you see in early Bally SS games when you press the flippers. Gottlieb, Bally and Williams all used 5V scanning in this era of games and every one I have worked on (other than this Flash) has scored reliably with the exception of Bally games with bad/missing capacitors.

The fundamental issue with this is that the targets are "EM Style" which have a target down park position (this is the series wired center solder points which triggers a bank reset) and a temporary "target falling" contact set which triggers scoring (these are the two outer solder points which are uniquely scanned for each target). If the target falls too fast for the scanning matrix to catch it, or if there is "contact bounce" where the contact is unreliable while the target is falling, the target won't score. On this Flash if I manually drop the targets with the PF vertical (pressing them with my finger to drop them) they score every time, but when the playfield is horizontal and the target is hit by a ball, they are falling faster and they won't score.

While researching this I saw a project in a different thread where folks were experimenting with two reed switches and gluing a magnet to the bottom of the target. Both reed switches would be closed when the target was in the down position and they tested this and found it worked. I don't own this Flash game and I don't feel comfortable with such a radical modification on a game I don't own - however I am tempted to layout a PCB if others want to have it fabricated and experiment on their games. I figure this had to work at one time from the factory - and replaced the horseshoes, guides and PCBs in frustration without success. Attaching a picture so you can see the Pinball Life PCBs and solder / wiring

The other aspect to this is the 1-2-3-4 are not 100% reliable either. If the ball is flown through the switch from the lower pop bumper then it often won't score. These switch stacks have all been cleaned, polished and I even reversed the incorrectly assembled contacts too. Which has me questing if this game has some problem in the MPU switch matrix like the 14xxx receiver.

Would enjoy discussing other ideas on this forum, this seems to be the biggest weak point (other than properly fusing BRs and solenoids and connector issues) on these games so whatever we can learn will help us all!

IMG_20200607_153110163_HDR (resized).jpg

I have read about people upgrading to Gottlieb drops, and trust me, this would be an upgrade for sure, as the sys4-6 drop system is horrible, and the reason (as I have been informed), that firepower uses standups.

Anyway, not sure where you are located, I have basically a box of Flash parts from a PF strip. Not sure if you wanted to just replace the whole assembly, if that is even where part of the issue is, or if it is handy to have some drop boards to swap.. but I am in Surrey, Canada (In the offchance you are local to here).

Sorry, have not gone back through this whole thread, but I presume for testing purposes, you can wiggle some of them slightly and make contact, it is more of a reliability issue in actual gameplay? FWIW, I did not use any grease at all on my banks, and only slightly torqued the circuit boards for each drop so that they are secure, but only just so.

Thanks for the generous offer koji I am located in the Boston area. At this point I have an extra set of guides, PCBs and Horseshoes after purchasing a new set to address this issue and finding the new parts perform identically to the original parts!

I have restored a dozen Gottlieb DT games and 100% agree with you their DT mechs are fantastic and by far my favorite. A Gottlieb EM set of DTs would work in this game (SS won't work because of the separate series wired "all drops dropped" switches). I have read of at least one Flash owner who installed Gottlieb DTs out of frustration and went back to the Williams DT because it changed the play so much for him and he missed the Williams feel (even though it is a pathetic design). Have read the same thing on FP - Steve Richey would have used DTs but they were so unreliable during testing that they used standups instead which was probably a good decision giving the parts Steve had available at the time.

I'll disassemble this again and add even more tension to these horseshoe pairs and see if that improves things and report back. My gut tells me there is something mysterious going in within the solenoid driver board (PIA, 7406 driver or 14xxx CMOS receivers) but I am trying to avoid touching the boards and causing more potential issues. Thanks again everyone, I appreciate the guidance and discussion!

That might have been me if you read it on RGP. If not, there's at least two people that tried this and hated it.

The ball just reacts.... wrong.

So when you have it in, powered on glass off, can you just try to wiggle the target? or even temp jump the connection to test that state?

I will say, once you have these drops dialed in, they are quite reliable in my experience. It's just getting them dialed in can be a pain :/.

Anybody that wants to try this, be very careful as those screw bosses on the plastic glides are fragile.

This is true, it's just the intervening couple years on my part that has made them unreliable again. Who knew, an EM style part in the game acts like EM parts over time and becomes less reliable with no use.

It sounds like you have everything just right to function correctly with the horshoe's adjustment, new PCB's & a bit of lube. That's how mine were when I did my Firepower retrofit with drop targets, I have never needed to adjust them once - that was done a few years ago. As you have done, the tension is just at the point where any more will slow the drop when dropping.

Ideally if you could swap over a known driver board to try so you can immediately eliminate that as you problem.

Looking at the schematic, IC 12 & 19 (7406) IC 15 & 16 (4049) & the switch matrix, the problem switches involved appear to be linked to row inputs through those IC's & the PIA IC 11. Just another possible fault with one of those IC's I guess.

When the glass is off the targets will score if I drop them with my fingers every time, but if I hit them with a ball, then they don't reliably score. My guess is that it drops slower with my finger than the ball due to the difference in friction between the target face and the ball or my finger. It is good to hear these can be reliable which was one of the reasons I reached out to the Flash group - have never worked on a Williams game like this and was wondering if it was even possible to have reliable scoring .

One of the differences between EM and this SS design is that the EMs used 25 (or higher) VAC which tends to better clean the contacts on a switch assembly like these DTs. You can see it is an EM design because of the spacing between the PCB contact traces which is way higher than a 5V system would need. The PBL replacements have much narrower spacing between the PCB swipe contacts. Also a SS game would not need a "temporary score trigger during target fall" contact pair like an EM game needs. Thanks!

I am going to try one more round of increased tension to the point where the tension is on the edge of slowing the target, they all drop super fast now. Also good point on swapping the driver board as a test to eliminate the drivers, receivers and PIA. I don't have one (and don't own this game) but I'll reach out to other local friends to see if I can borrow a spare, that may take me a few weeks. I did scope out the 7406 driver waveforms and they all were correct so I don't think the driver is the issue.

Also, not sure if you wrote the post which showed the FP conversion but if so, I really enjoyed reading about it!. Thanks everyone, I'll report back once I get back into this game this weekend.

So they score with touch, and they also reset when they are all down? They have a secondary connection on fully down as well.

I messed with the spring tension on mine as well.. never had any issue I could say related to speed etc. Do all the drops behave the same? that would be an odd symptom as well, but perhaps telling.

So I was looking at the way WPC uses +12v instead of +5v for the matrix to see if something could be grafted onto the game to boost the voltage to hopefully overcome limitations on the horseshoe contacts - but I'm not sure that's worth pursuing for a couple reasons. I'd assume that the gold plated contacts can handle +12 because they are essentially the same switches used up into the WPC era (although they are microswitches in most cases, there are still occasional uses of leaf switches).

The main hesitation would be if the increased voltage would even help. I suppose I can't know that for sure without trying it, I'd have to build a level shifter to slap onto the upper right connectors from the driver board but I don't see why this wouldn't work (unless the timing ends up being way too slow....) to use higher voltage (of course, I'll also have to source +12v somewhere on the machine....)

The only thing that has helped me in the past when adjusting these was the use of Deoxit, and changing the angle the actual contact meets the circuit board, to overcome flat spots on the contact and trails in the board. Like I said they stay solid for years, but in my case I think they needed to get played more often.

I know last time I had Time Warp out for a show it had sat a year and was flawless when set up at the show, but now we're talking years later.... now that I think about it, it might even be over 10 years so maybe it's time to clean that stuff up and replace. Funny how the all drops down series signal is still perfect on both sets of drops though.

Both the 5 bank and 3 bank reset every time when all targets are down. All 8 targets will score correctly (100% of the time) if I trigger them with my thumb, but when they are hit with a ball, some targets will score at times. They are all unreliable, there is not one target that scores 100% of the time with a ball, but you bring up a good point in that one target may be more reliable than the other. You are right, there are two connections one which is static (the inner pair of contacts on the PCB which are series wired to detect all DTs are down) and the other which is temporary / very fast (the outer pair of contacts on the PCB which have dedicated diodes / scanning connections to trigger scoring). The MPU is missing these very fast switch assertions during the drop action and it correctly detects the static condition when all drops are completed. What is really frustrating is there is absolutely zero reason to have two sets of contacts on a SS game since the firmware in the MPU could easily have scored and detected all DTs down with a single contact pair (like Bally, Stern and Gottlieb) but they choose to separate these functions. An EM of course requires a separate trigger and all targets down circuit, so they coded it to mimic an EM game!

Funny, i thought about a 12V WPC style conversion too - specifically to reduce the false pop bumper triggers you see on some Bally SS games like EBD. The higher voltage helps in improving the signal to noise ratio and prevent false triggers, but unfortunately I have the opposite problem with this Flash in that it won't trigger when it should . The scanning pulses are only 40 uS wide on this game so the higher voltage would actually make the scanning detection worse since the drivers would have to traverse a higher voltage span in the same timeframe. Thanks for the thought.

I had not considered the angle of the flat side of the horseshoe contacts - I had bent them outward (along with the rear spring only contact) but that does change the angle slightly. I am going to try that next when I re-tension these for the 3rd time!

Note that other targets don't score 100% reliably either, 1-2-3-4 roll-overs are ignored sometimes, especially when hit by a Pop Bumper and the ball flies over them quickly, the stand ups only score maybe 50% of the time, the star roll-overs score maybe 40% of the time, the outlanes score maybe 70% of the time. I have cleaned all contacts and even reversed some (to fix the factory manufacturing error) too.

This is such a fun title, I really want to have it 100% perfect. Thanks everyone, this is a great owners group!

I finally have the 5 bank DT scoring 97% reliably which seems to be as good as it is going to be. The key is to not only increase the contact tension even further (to the point where the DTs are barely dropping) but also to bend them so the flat rivet face is parallel to the PCB. The increased tension must be slowing down the drop action sufficiently for the switch matrix scanning to catch the temporary drop contact closure and making the contact as flat as possible to the PCB likely increases the contact time slightly too.

Thanks to everyone for your help! Have learnt a ton working on this game (my first classic Williams) and really appreciate everyone stepping up with advice to help out. Cheers!!!

I spoke too soon, after playing about a dozen games last night, Flash is back to scoring maybe 30% of the time on the drops. It is crazy, it was scoring nearly 100% with the glass off and flinging the ball with my hand! This thread has excellent recommendations so it is relevant to your game, but this Flash likely has a board set issue as some here have speculated. I'll be comparing this board set to a known working set soon and hope to find the root cause (interrupt trigger differences for switch scanning, etc). If I find a board issue, I'll report back here so others can learn from it. Stay tuned, it may take a few weeks. Good luck with your DTs!

Wow.. sorry to hear that. The drop mech is bad, but not that bad. The fact that it seems to be all of them, and not one or two in particular seems somewhat telling.

I recall reading that you had repinned everything.

Only other thing that comes to mind is I know sometimes with these boards, and particularly with a repin, maybe something is shorting a bit on the metal back plate?

Yes every board has been re pinned (not by myself - the work looks to be top notch). The game fails to score roll-overs probably 30% of the time, stand up targets 50% of the time - in general every scoring aspect of the game is not 100%. At this point I am planning to pull the boardset and bring it to a friend who has a known working set so we can compare the two (interrupts, PIA, switch matrix scanning, etc). Hope to schedule this in the next few weeks and I'll post back whatever we learn.

Do you see reliable scoring on your Flash game - every time you hit a standup, every time you roll over a outlane or 1-2-3-4 or do your games score 100% reliably or does your games miss scoring sometimes too? Thanks!

Great please let us all know if your new boards are scoring 100%. Good luck!

Hi there! Not sure if this is the right place to post this, so please let me know if I need to move it!

I picked up a Flash about a month ago that was mostly in decent condition. I cleaned and waxed it, put on new rubbers, new ball, cleaned the drop contacts and got them working correctly, reset the shooter rail angle to send the ball the right way, and last but not least replaced a busted rollover insert.

Anyway, after all that, I just don’t feel like the game is setup right? Perhaps it’s me just not being good at it? I understand the rule set and I can make the shots I need to make but it just feels like the game is designed to send the ball screaming down the drain or the ball is magnetized to the outlanes so most games last a handful of minutes. It’s super frustrating and I kinda regret buying it most nights I’m playing it.

I guess what I’m getting at is, what exactly is a “properly setup” Flash supposed to be? I see people throw that phrase around and never explain what it means. I have the shooter rail set according to the manual specs (half pull sends the ball up the figure eight path), table angle is at 3.5° like the manual says, the outlane posts are in the “conservative” position, and the flippers don’t feel particularly weak. Am I missing something?

Set your playfield angle from 6.5 to 7 degrees.
Let her rip . I like my flash steep and it plays fast
and fun with all New rubber rebuilt pops and flippers.
Good luck it's my favorite game right now in my collection

You suck.

Hmm.. 3.5" def won't play very fun for the modern day. Def recommend moving that up to at least 5 or 5.5>.

Flash is definitely not known for particularly long ball times, and the outlanes are hungry, however these are generally punishment for a missed shot etc, as it blasts off the slingshot.

I have to imagine the steeper pitch will allow you to better trap the ball with the flippers to make proper shots.

FWIW, here is a video I took of mine a while back. I was pretty pleased with being able to dial in a repeatable loop shot, among other little things.

Ha. Yeah, I’m sure that’s 98% of my problem.

I’ll try adjusting the angle some. Thanks y’all!

RE: Long ball times and hungry outlanes. That’s good to know and totally what I’m experiencing. I’d say 1 in 50 games I “get it” and can post a score above 200k but the rest of the time it’s a bear so maybe I’m doing it right-er. Thanks for the detailed reply!