New! Dark mode!

Browsing Pinside at night? Getting tired of all the white? Switch to dark mode using the button in the top right (or CTRL-B)!

(Topic ID: 278480)

All displays flicker (flash gordon) - not sure what to try next


By EvanDickson

64 days ago



Topic Stats

  • 45 posts
  • 8 Pinsiders participating
  • Latest reply 43 days ago by EvanDickson
  • No one calls this topic a favorite

You

Linked Games

Topic Gallery

There have been 7 images uploaded to this topic. (View topic image gallery).

DisplayHighVoltageRipple.jpg
DisplayBlankingSignal.jpg
DisplayInterruptGenerator.jpg
capture 2 (resized).PNG
Capture (resized).PNG
pasted_image2 (resized).png
pasted_image (resized).png

#1 64 days ago

Editting this for the future persons that find the thread because they have the same problem.

The issue: All displays are flickering on my Bally Flash Gordon (MPU -35). A video is at the bottom of the post.

Possible causes to be examined:
* Check that voltage to the displays is good (spec is 190V DC, manual says voltage can be dropped to 150-170 after displays are burned in).
* Cold solders / bad connections at displays (in this case all displays began flickering at once, but this is a likelihood for a single display flickering).
* Ground/short in machine (disconnect coin door connector, remove GI fuse, solenoid fuse under PF, unplug lamp, aux lamp boards).
* flipperwinkel.nl says to reflow solder at "Pin 36 on the display glass, R21, R22, R29.Q17.U1 pin 13. Unclear yet which board.
* The HV Capacitor on the power/solenoid board can be bad. Check for AC ripple.
* Check the blanking signal to displays - pin 10 on display connector, it should pulse.
* With oscilloscope, look for 430 Hz at pin 3 of U1 on the MPU.

#2 64 days ago

Check to make sure you don't have any slight grounding of any GI . I had a similar issue. It turned out one of the coin door lamp wires had worn insulation, and would occasionally rub on the coin door causing some crazy score display issues, as well as a bunch of controlled lamps to not light and others to stay lit. It took me a while to solve this...

#3 64 days ago

Definitely reflow the solder on the header(s) on the display boards. Those connections always crack.

Jeff

#4 64 days ago

And it's very possible the high voltage filter capacitor on the solenoid driver board has failed.

#5 64 days ago
Quoted from KenLayton:

And it's very possible the high voltage filter capacitor on the solenoid driver board has failed.

Is that not the 11,700 ufd one that I replaced?

Quoted from Grangeomatic:

Definitely reflow the solder on the header(s) on the display boards. Those connections always crack.
Jeff

I'll do this as a precautionary / maintenance thing, but since the displays all started flickering together, and I've juggled which displays are plugged in, I don't think this is causing the issue at the moment.

Quoted from mkdud:

Check to make sure you don't have any slight grounding of any GI . I had a similar issue. It turned out one of the coin door lamp wires had worn insulation, and would occasionally rub on the coin door causing some crazy score display issues, as well as a bunch of controlled lamps to not light and others to stay lit. It took me a while to solve this...

Good idea. I popped the GI fuse to remove them from the equation and no change. I'll try with all fuses that aren't 5V or display power later to do the same in case of ground shorts with a coil somewhere.

Are displays fed with some signal from MPU? I wonder if there could be a failure on the MPU somewhere.

#6 64 days ago

Seen this on several machines. Can you check the frequency of the display interrupt on the MPU. Can't remember what it should be but I'm sure that it can be found on the web. This is what is responsible for generating the interrupt that controls the refresh speed for the displays. It consists of a 555 timer chip, some resistors and a couple of capacitors. I have seen the caps and IC go bad and it can be in the battery damage area.

#7 63 days ago

320Hz BTW

#8 63 days ago

Edit, I got my Hantek USB oscilloscope drivers / software installed and working. I'll do some studying tonight to figure out how to read it. I found this, am I looking to see 320 Hz on each of the highlighted pins? And can I compare to ground braid (easier to attach to than pin 1), or must I compare pin to pin within the chip?

pasted_image (resized).pngpasted_image2 (resized).png

#9 63 days ago

Check pin 3 for 320Hz. Yep, use the ground braid as reference, or better, clip your scope ground to TP4 on the MPU.

#10 63 days ago
Quoted from EvanDickson:

Good idea. I popped the GI fuse to remove them from the equation and no change. I'll try with all fuses that aren't 5V or display power later to do the same in case of ground shorts with a coil somewhere.

You can't remove the solenoid fuse game won't boot up without the zero cross (or run, for that matter)

Look for a diode on the coin door shorted. Boot the game up with the coin door connector disconnected to check if there's something amiss there.

Quoted from EvanDickson:

Is that not the 11,700 ufd one that I replaced?

Nope, that's your +5 filter cap. HV filter cap is the one at the top with the 350v rating on it. What is your display voltage set to? <190vdc and some displays can flicker. (I know "everyone" says to set your display voltage to 170-175, but factory spec is 190)

Also try one display hooked up only, see if you get the same results.

#11 63 days ago
Quoted from slochar:

You can't remove the solenoid fuse game won't boot up without the zero cross (or run, for that matter)
Look for a diode on the coin door shorted. Boot the game up with the coin door connector disconnected to check if there's something amiss there.

Nope, that's your +5 filter cap. HV filter cap is the one at the top with the 350v rating on it. What is your display voltage set to? <190vdc and some displays can flicker. (I know "everyone" says to set your display voltage to 170-175, but factory spec is 190)
Also try one display hooked up only, see if you get the same results.

I've tried disconnecting all but one display (and switched to a different display after to make sure I didn't leave the one faulty one on). Steady and seemingly clean 178 V on the displays. I bumped it up to 192V and there was no difference.

Tried removing the coin door connector with no change.

Will try replacing that other cap later today, along with checking frequency on the display interrupt. Thanks!

#12 63 days ago

It's C26 the 160 uf @ 350 volt axial one that needs to be replaced. That one is the high voltage filter capacitor.

I've replaced lots of them over the past few years.

#13 63 days ago
Quoted from Bakerman:

Check pin 3 for 320Hz. Yep, use the ground braid as reference, or better, clip your scope ground to TP4 on the MPU.

On Bally games with 7 digit displays, the display interrupt generator actually runs faster at 430Hz. These MPU boards have a different resistor at R21 for the 555 timer. See this thread with photos:
https://pinside.com/pinball/forum/topic/two-bits-mpu-with-flickering-strobing-displays-#post-5408121

Quoted from EvanDickson:

I bumped it up to 192V

If you bumped the HV to 192V and the pot was near the middle, it says the HV 160uF 350V capacitor is possibly ok. Measure how much AC ripple you're getting.

Check the "blanking" signal to the displays. Pin 10 on the display connector. It should be pulsing.

#14 63 days ago
Quoted from Quench:

On Bally games with 7 digit displays, the display interrupt generator actually runs at 430Hz. These MPU boards have a different resistor at R21 for the 555 timer. See this thread:
https://pinside.com/pinball/forum/topic/two-bits-mpu-with-flickering-strobing-displays-#post-5408121

If you bumped the HV to 192V and the pot was near the middle, it says the HV 160uF 350V capacitor is possibly ok. Measure how much AC ripple you're getting.
Check the "blanking" signal to the displays. Pin 10 on the display connector. It should be pulsing.

To measure AC ripple, I just check for / measure AC Voltage at one of the poles of the cap?

#15 63 days ago
Quoted from EvanDickson:

To measure AC ripple, I just check for / measure AC Voltage at one of the poles of the cap?

Yes, meter set to AC high voltage and a meter probe on each leg of that HV capacitor at the top left corner of the solenoid driver board.

#16 63 days ago
Quoted from Quench:

Yes, meter set to AC high voltage and a meter probe on each leg of that HV capacitor at the top left corner of the solenoid driver board.

At the positive end of the cap, it oscillates beween 0.35V and 0.15V AC. At the negative end, it sits around 0.01V AC.

#17 63 days ago
Quoted from EvanDickson:

it oscillates beween 0.35V and 0.15V AC

Sounds ok to me. Anyway if you've got another Solenoid Driver Board, try it to rule the current SDB out.

Get a logic probe on that blanking signal I mentioned at the end of post #13.

#18 63 days ago
Quoted from Quench:

Check the "blanking" signal to the displays. Pin 10 on the display connector. It should be pulsing.

I'd missed that line.

Is that pin 10 on any one of the display boards' connectors? Or is that a connector on the mpu, or lamp board, or aux lamp board? The manual isn't as detailed as newer machines.

#19 63 days ago
Quoted from EvanDickson:

Is that pin 10 on any one of the display boards' connectors?

Pin 10 at any of the displays. That signal is daisy chained from one display to another. It comes from the MPU board at J1 pin 10 and is an orange wire.

#20 63 days ago
Quoted from Quench:

Pin 10 at any of the displays. That signal is daisy chained from one display to another. It comes from the MPU board at J1 pin 10 and is an orange wire.

Yes, pulsing, square wave < 200 kHz, if I'm reading my probe instructions correctly.

#21 63 days ago
Quoted from EvanDickson:

Yes, pulsing, square wave < 200 kHz, if I'm reading my probe instructions correctly.

It should briefly pulse high at around 430Hz which is every time a digit is written to the displays. If you're getting 200kHz, that's way over the top.

What frequency did you measure on the pin 3 output of the 555 timer at U12? It should be around 430Hz.
BTW does this have a Bally MPU board or some aftermarket board? If it's a Bally board, what's the silkscreened date on the bottom right corner?

#22 63 days ago
Quoted from Quench:

It should briefly pulse high at around 430Hz which is every time a digit is written to the displays. If you're getting 200kHz, that's way over the top.
What frequency did you measure on the pin 3 output of the 555 timer at U12? It should be around 430Hz.
BTW does this have a Bally MPU board or some aftermarket board? If it's a Bally board, what's the silkscreened date on the bottom right corner?

Sorry, that was logic probe. I don't have an exact frequency. But 430 < 200,000. Will try with the oscilloscope when I figure out the software this evening.

Silkscreened date (original Bally MPU) is 10-79.

#23 63 days ago
Quoted from Quench:

It should briefly pulse high at around 430Hz which is every time a digit is written to the displays. If you're getting 200kHz, that's way over the top.
What frequency did you measure on the pin 3 output of the 555 timer at U12? It should be around 430Hz.
BTW does this have a Bally MPU board or some aftermarket board? If it's a Bally board, what's the silkscreened date on the bottom right corner?

If I've stumbled through using the oscilloscope partially correctly, then I'm getting 343 Hz on pin 3 of U12, and 113 Hz on the blanking signal for the displays. So the timer is running slower than expected. Perhaps I should replace U12, R21, R22 and C16.
Capture (resized).PNGcapture 2 (resized).PNG

#24 62 days ago
Quoted from EvanDickson:

Silkscreened date (original Bally MPU) is 10-79.

Quoted from EvanDickson:

then I'm getting 343 Hz on pin 3 of U12

Your MPU board dated 1979 came from an older game with 6 digit displays so is running the display frequency generator (555 timer) at a slower speed. Some people will notice the displays on your game strobing a little and it might annoy them. Others less sensitive won't notice.

Quoted from EvanDickson:

113 Hz on the blanking signal for the displays.

That blanking speed doesn't sound right. The blanking signal is software controlled by the display interrupt service routine that happens every pulse out of the 555 timer (i.e. in your case blanking should be around 343Hz).
I'll have to confirm this on one of my games later - can't do it at this minute.

With reference to the video in your first post, it shows all the displays flickering badly. Is this how it really looks on the game or does the flickering appear worse in the video due to camera sync artifacts?

#25 62 days ago
Quoted from Quench:

Your MPU board dated 1979 came from an older game with 6 digit displays so is running the display frequency generator (555 timer) at a slower speed. Some people will notice the displays on your game strobing a little and it might annoy them. Others less sensitive won't notice.

That blanking speed doesn't sound right. The blanking signal is software controlled by the display interrupt service routine that happens every pulse out of the 555 timer (i.e. in your case blanking should be around 343Hz).
I'll have to confirm this on one of my games later - can't do it at this minute.
With reference to the video in your first post, it shows all the displays flickering badly. Is this how it really looks on the game or does the flickering appear worse in the video due to camera sync artifacts?

It looks pretty bad, the video is accurate. When I first got this thing home last weekend, the displays were fine, there was no flicker. Then they all started flickering intermittently - all displays would flicker or all would be fine. Now they all flicker.

#26 62 days ago

Did you replace that high voltage capacitor yet?

#27 62 days ago
Quoted from KenLayton:

Did you replace that high voltage capacitor yet?

Not yet. I placed a digikey order last night for other stuff, but they had nothing like that capacitor in stock. Even GPE doesn't have a suitable substitute.

I guess I should get on the google and figure out the calculations to combine capacitors to achieve the correct specs, I'm assuming there are formulas, like for combining resistors.

#28 62 days ago
Quoted from EvanDickson:

When I first got this thing home last weekend, the displays were fine, there was no flicker. Then they all started flickering intermittently

What changed between it working and when the flickering started?
If the flickering is intermittent, have you wiggled connectors to see if there's any bad connections?

#29 62 days ago
Quoted from Quench:

What changed between it working and when the flickering started?
If the flickering is intermittent, have you wiggled connectors to see if there's any bad connections?

Nothing notable changed. While playing the first few games, they flickered a bit. And the flicker became permanent. I've wiggled all the things lots of times, and reflowed anything that was recommended so far.

#30 62 days ago

Great Plains Electronics # CEA-150UF-350V-RMD is the cap you need.

https://www.greatplainselectronics.com/products.asp?cat=108

#31 62 days ago
Quoted from KenLayton:

Great Plains Electronics # CEA-150UF-350V-RMD is the cap you need.
https://www.greatplainselectronics.com/products.asp?cat=108

It's ok to go below spec on the capacitance? I thought you could substitute with either higher capacitance or higher voltage rating, but not go lower?

#32 62 days ago
Quoted from EvanDickson:

It's ok to go below spec on the capacitance? I thought you could substitute with either higher capacitance or higher voltage rating, but not go lower?

You won't find a 160uF capacitor. It's a non-standard capacitance value - strange choice back in the day really.
The part Ken listed is fine. I install 180uF, 105 degree radial caps since I'm not in the US.

While you wait for the replacement HV capacitor, reflow the current HV capacitor solder joints in case they're cracked - does happen. And while you're there implement the ground mod for it if it hasn't been done yet.

#33 62 days ago
Quoted from Quench:

You won't find a 160uF capacitor. It's a non-standard capacitance value - strange choice back in the day really. I install 180uF radial caps.
While you wait for the replacement HV capacitor, reflow the current HV capacitor solder joints in case they're cracked - does happen. And while you're there implement the ground mod for it if it hasn't been done yet.

I'm a dummy for not thinking of just using a radial.

I'll follow up here on my progress when I've done stuff. Probably next weekend. It won't be til mid or late next week that the parts arrive.

Thank you muchly!

#34 62 days ago
Quoted from EvanDickson:

I'm a dummy for not thinking of just using a radial.

Order the capacitor Ken mentioned. It's a large physical size so should have a long life. Smaller physical capacitor size doesn't equal better.

#35 62 days ago

Since you have an oscilloscope, you can visualise the ripple on the high voltage rail to the displays which will tell you if the HV capacitor is your problem.
First check if your oscilloscope can handle 200 volts and take care.
Make sure to set the X1/X10 switch on the oscilloscope probe to X10 so you don't damage the oscilloscope (this divides the input voltage by 10).
Set the oscilloscope to AC and highest voltage setting then adjust until you can get a reading.

Here are snapshots from my 6 digit display machine. Display Blanking and Display Interrupt generator speed are about the same (allow for variance/error in reading on my ancient oscilloscope).

Last picture is the high voltage ripple - I get around 300 millivolts. My high voltage is set to 180 volts.

DisplayInterruptGenerator.jpg
DisplayBlankingSignal.jpg
DisplayHighVoltageRipple.jpg

#36 58 days ago

My stuff arrived today. I replaced the solenoid / power driver cap 160 uF / 300 V with 180 uF/350V. No change to the displays.

I replaced U12 timer555 with NE555P which supposedly replaces timer555. I also replaced polyester capacitor C16 .1 uF 100v with a polyester 0.1uF 400V. No change to displays. But my attract mode ilights are moving at super high speed. I'm not sure if it was like that before or if I'm just noticing it now. Lights seem to move at a normal speed during game play (eg when the lights at the saucer point to the upper or lower playfield before the ball kicks out).

Resistor R21 is 24K.

I guess next I start repinning all the connectors?

#37 57 days ago
Quoted from EvanDickson:

I replaced the solenoid / power driver cap 160 uF / 300 V with 180 uF/350V. No change to the displays.

The AC ripple you previously measured indicated the old cap was doing its job. At least you have a fresh minty cap there now to last for years to come.

Quoted from EvanDickson:

I also replaced polyester capacitor C16 .1 uF 100v with a polyester 0.1uF 400V.
...
Resistor R21 is 24K.

You should be getting around 430Hz out of the 555 timer now.

Can you measure the voltages at the test points on one of the displays BUT use TP3 on the display as your ground connection for the multi-meter.
Do test points TP1 (5V) and TP2 (190V) measure ok?

#38 57 days ago
Quoted from Quench:

The AC ripple you previously measured indicated the old cap was doing its job. At least you have a fresh minty cap there now to last for years to come.

You should be getting around 430Hz out of the 555 timer now.
Can you measure the voltages at the test points on one of the displays BUT use TP3 on the display as your ground connection for the multi-meter.
Do test points TP1 (5V) and TP2 (190V) measure ok?

The last thing I did last night before bed was reflow all the header pins on the board. After that, it does the 'weewoo' siren thing longer than usual, and the game fails to boot, even though I get seven flashes. Displays remain off. It might be that the MPU is just too old and fatigued. I ordered an Alltek replacement. Hopefully that does the trick, I'll check these things once it arrives next week and I'll report back if that still doesn't help.

#39 57 days ago
Quoted from EvanDickson:

The last thing I did last night before bed was reflow all the header pins on the board. After that, it does the 'weewoo' siren thing longer than usual, and the game fails to boot, even though I get seven flashes. Displays remain off. It might be that the MPU is just too old and fatigued. I ordered an Alltek replacement. Hopefully that does the trick, I'll check these things once it arrives next week and I'll report back if that still doesn't help.

I'm wrong. With fresher eyes I get 6 flashes. There's no solenoid power at TP3. I think some sloppy wiring by me grounded the new capacitor to the backbox. Ungrounding it doesn't fix the issue, I'll test/troubleshoot the power issue tomorrow night.

#40 57 days ago
Quoted from EvanDickson:

I'm wrong. With fresher eyes I get 6 flashes. There's no solenoid power at TP3.

Might have blown the solenoid fuse on the rectifier board. The MPU board needs to see solenoid power at J4 pin 15 and uses it as the "Zero Crossing" signal. The last MPU LED flash on power up tests the presence of this signal.
If you reflowed that connector carefully check there's no accidental solder bridges on the pin header.

#41 56 days ago
Quoted from Quench:

Might have blown the solenoid fuse on the rectifier board. The MPU board needs to see solenoid power at J4 pin 15 and uses it as the "Zero Crossing" signal. The last MPU LED flash on power up tests the presence of this signal.
If you reflowed that connector carefully check there's no accidental solder bridges on the pin header.

My rectifier board was missing a few voltages. I had a failed BR. Replaced it and I'm back to 6 flashes, a siren, and GI lighting up, but it's still not fully there. TP1 on the rectifier board is now 7.8V, which is a little high, and TP5 is 20V instead of 43. The other BR checked as ok but I might pull it and replace anyways, I'll leave that for tomorrow. This is becoming a farce. :p

Edit: connector wiggling got TP5 up to 43V. TP voltages are all perfect on solenoid board. Missing TP2 (12V) on MPU. Trace it back to the rectifier board, pin 8 of J2 reads 0 on the rectifier board (fuses are all good, tested out of sockets). Will look at the board side of the connector tomorrow for cold solder / evil elves, and replace the other BR that tests ok in case it's not really ok and they failed in tandem somehow. There's so little happening on a rectifier board that I'm sure this should be simple.

#42 56 days ago
Quoted from EvanDickson:

TP1 on the rectifier board is now 7.8V, which is a little high

The feature lamp voltage is known to be a bit high in these games. As lamps illuminate (presuming you're using incandescents) the voltage will drop under load.

Quoted from EvanDickson:

TP5 is 20V instead of 43.

Half voltage is usually indicative of an open circuit diode inside the bridge. You need to disconnect the 600 ohm resistor (R1) that's across the bridge to test the bridge properly, otherwise you'll get misleading readings.

#43 56 days ago
Quoted from EvanDickson:

My rectifier board was missing a few voltages. I had a failed BR. Replaced it and I'm back to 6 flashes, a siren, and GI lighting up, but it's still not fully there. TP1 on the rectifier board is now 7.8V, which is a little high, and TP5 is 20V instead of 43. The other BR checked as ok but I might pull it and replace anyways, I'll leave that for tomorrow. This is becoming a farce. :p
Edit: connector wiggling got TP5 up to 43V. TP voltages are all perfect on solenoid board. Missing TP2 (12V) on MPU. Trace it back to the rectifier board, pin 8 of J2 reads 0 on the rectifier board (fuses are all good, tested out of sockets). Will look at the board side of the connector tomorrow for cold solder / evil elves, and replace the other BR that tests ok in case it's not really ok and they failed in tandem somehow. There's so little happening on a rectifier board that I'm sure this should be simple.

Something simple- make sure fuse clips have good tension and not tarnished

1 week later
#44 44 days ago

Did you ever get to the bottom of this?

#45 43 days ago
Quoted from Bakerman:

Did you ever get to the bottom of this?

I did and I didn't. I cleaned up everything else in the machine, had all the right voltages to the MPU, and still had the flicker. I rebuilt the timer circuit with a new 555 chip, and the related polyester resistor and other resistors, and still had the flicker. I replaced with an Alltek MPU and the flicker cleared up.

So maybe a problem with one of the other IC on the MPU further upstream from the timing circuit. At some point I'll read this all again with fresh eyes and see what else I can do to bring the MPU back to 100% as a spare.

Promoted items from the Pinside Marketplace
$ 40.00
Gameroom - Decorations
The Flipper Room
$ 279.95
Lighting - Led
Pin Stadium Pinball Mods
$ 259.95
Lighting - Led
Pin Stadium Pinball Mods

Hey there! Got a moment?

Great to see you're enjoying Pinside! Did you know Pinside is able to run thanks to donations from our visitors? Please donate to Pinside, support the site and get anext to your username to show for it! Donate to Pinside