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: 228155)

Bally controlled lamp flicker


By semicolin

1 year ago



Topic Stats

  • 22 posts
  • 8 Pinsiders participating
  • Latest reply 1 year ago by semicolin
  • Topic is favorited by 2 Pinsiders

You

Linked Games

Topic Gallery

There have been 6 images uploaded to this topic. (View topic image gallery).

44BAFF1E-57E3-403C-9F4F-42A2C69D7FDF (resized).jpeg
IR_2411 (resized).jpg
A53D2802-7AE1-4DB5-BBCB-D489AF84A0B3.jpeg
8333957B-28DD-4C1C-B89B-50584322BE79.jpeg
FBC424A2-5D99-4AA9-82B2-DC4E3A01C5F3 (resized).jpeg
D3A9A24D-30BE-402E-86FA-136110C3CEE5 (resized).jpeg

#1 1 year ago

So some of the controlled lamps on my Centaur 2 are misbehaving. When they are supposed to be off, they flicker in a repeating pattern. They work fine when they should be on. This doesn't only happen with these two lamps; other lamps are affected depending on what's happening in the game.

I recorded a video of this happening.

#2 1 year ago

And you've added a resistor to the sockets or alltek board? Those are LEDs right?

#3 1 year ago

Are the lights LEDs?

#4 1 year ago

The lamps are LEDs, but they have resistors. Either way, these two aren't actually "on" right now, so the SCR holding current should not be the issue. When the lamps are actually in their "on" state, there is no flicker at all.

#5 1 year ago

The scr's might be leaking a small amount of current before they fail and are stuck on. Replace one and see.

#6 1 year ago
Quoted from jj44114:

The scr's might be leaking a small amount of current before they fail and are stuck on. Replace one and see.

Ordinarily I'd be inclined to agree, but there are a couple of other lamps that also do this, and it's all intermittent. Most of the time, it's not doing this.

#7 1 year ago

Then I would suspect connectors or cracked solder joints on header pins on the lamp driver/aux driver boards.

#8 1 year ago

Pretty sure I've ruled out headers A5J4 and A4J1; a good poke/wiggle/shimmy doesn't affect the end result.

Every lamp on the playfield independently goes ON when it's supposed to fire, and OFF when it isn't. No crossed wires. The problem is that some fire occasionally when they aren't supposed to, and this changes depending on which other lamps are lit.

Here is an additional example:

I'm not inclined to think it's bad 4514 decoders considering that multiple decoders are involved, but that they're getting bad data. If there was a broken header or trace on the data/address bus between the MPU and driver board, I feel like it would impact multiple lamps in unison and far more intrusively than this. It could be a problem with the 6820/1 PIA but I don't know if that would manifest itself in this way. Outta my experience here.

#9 1 year ago

Check your voltage going into the controlled lamps, low voltage on the lamp bus can cause some wacky things with LED's and load resistors.

-Hans

#10 1 year ago
Quoted from HHaase:

Check your voltage going into the controlled lamps, low voltage on the lamp bus can cause some wacky things with LED's and load resistors.

Hans, you may be on to something here. 3-4VDC is all that is available on that bus; 3.75V between TP1 and TP2 on the lamp driver board. Going to have to look into that further.

#11 1 year ago

Solved! Two things that had to be fixed:
1) Shorted bridge rectifier on the rectifier board was causing only a half wave to hit the lights.
This fixed a few things:
-increased brightness of all of the controlled lamps
-increased stability of all the lamps: no dimming or flicker
-no more hum from transformer. The failed diode was likely causing the transformer to short out during 1/2 of the wave, collapsing the field and making an audible vibration in the core
-no hum from audio amplifier circuit. The DC lamps were operating on a half-wave, so they were making interference that the amps were picking up

So: lots of problems solved, except no impact on the blinking lights, which actually got worse.

Tested the output from the LM353 on the solenoid/regulator board. Rock solid 5V. Tested it at test points on the MPU, Aux driver, squawk and talk, main lamp driver. Rock solid 5V. Tested it at Vdd of one of the 4514 latching decoders: 3V. Bingo! Same on all four. Turns out the trace was bad after the test point on the lamp driver board.

So the 4514 decoders *were* failing because of low voltage, and I happened to find and fix a bunch of other issues on the way.

Thanks for your help, everyone.

Thanks for being a part of this. It was a slice.

(And check out the before and after. I think it might have got hot. New one is a bit beefier.)
D3A9A24D-30BE-402E-86FA-136110C3CEE5 (resized).jpegFBC424A2-5D99-4AA9-82B2-DC4E3A01C5F3 (resized).jpeg

#12 1 year ago

Excellent work there!

What model bridge was the old one vs the new one?

If I were you I would put a more substantial heatsink on that bridge rectifier. The original design had the bridges mounted underneath the PCB to the large metal base plate to dissipate heat.

#13 1 year ago
Quoted from Quench:

Excellent work there!
What model bridge was the old one vs the new one?
If I were you I would put a more substantial heatsink on that bridge rectifier. The original design had the bridges mounted underneath the PCB to the large metal base plate to dissipate heat.

The old one looks to have been a GBPC2506W, no heatsink, driving incandescent lamps. The new one is a GBPC3504W, driving LEDs. I would have liked to have mounted it closer to the board but I needed access to the top traces for soldering. That heatsink is the largest I've got, but considering the reduced draw and the fact that Rottendog ships theirs with no heatsink at all, I think it should be okay.

That said, I'm curious, so tonight I'll play a few games, get it up to temperature, and throw the FLIR at it.

Speaking of FLIR, check out R11 on the MPU. Man that's cooking!!! Probably could use a three watter in there instead of a 1W...
8333957B-28DD-4C1C-B89B-50584322BE79.jpegA53D2802-7AE1-4DB5-BBCB-D489AF84A0B3.jpeg

#14 1 year ago
Quoted from semicolin:

The old one looks to have been a GBPC2506W, no heatsink, driving incandescent lamps.

No wonder it failed.

Quoted from semicolin:

Speaking of FLIR, check out R11 on the MPU. Man that's cooking!!! Probably could use a three watter in there instead of a 1W...

Indeed. Capacitor C23 on the solenoid driver board will lift the 12V rail closer to 17V. That means R11 on the MPU board could be dissipating around 1.6W
Measure the voltage across R11.

#15 1 year ago
Quoted from Quench:

No wonder it failed.

Indeed. Capacitor C23 on the solenoid driver board will lift the 12V rail closer to 17V. That means R11 on the MPU board could be dissipating around 1.6W
Measure the voltage across R11.

IR_2411 (resized).jpg

Max temperature 45.3°C after several hours under load. I would say that's a major improvement!

Dropping 9.7VDC across R11. Assuming that it's still 82Ω (haha) then it's dissipating 1.15W. Just a little out of spec for this revision one board which called for a 1W resistor. I almost don't want to replace it given that it's been chugging along fine for so many years, but then again getting it off the board entirely by replacing it with this chassis mount power resistor doesn't feel irrational: https://www.digikey.ca/product-detail/en/te-connectivity-passive-product/THS1582RJ/A131973-ND/2366910

#16 1 year ago
Quoted from semicolin:

Max temperature 45.3°C after several hours under load. I would say that's a major improvement!

That's fine. What's the ambient temperature?
How warm/hot is the bridge rectifier getting?

#17 1 year ago
Quoted from Quench:

That's fine. What's the ambient temperature?
How warm/hot is the bridge rectifier getting?

Ambient is approx 20 degrees C. The bridge rectifier and heat sink are both selected in that box and they're running at the same temperature, 45 degrees. There's good thermal conductivity between them and that the temperature itself should be stable. 45 degrees is right in the sweet spot for this component.

44BAFF1E-57E3-403C-9F4F-42A2C69D7FDF (resized).jpeg
#18 1 year ago
Quoted from semicolin:

45 degrees is right in the sweet spot for this component.

If you're running LEDs you shouldn't have a problem. Remember that rises in ambient temperature will increase the heat that the component is running at. So when measuring temperature consider worst case ambient.
You should be ok though.

#19 1 year ago

Cool heat camera. I kinda want to get one.

That 'VUAX' circuit dropping 12v down for the u8 ram chip seems a bit silly and that resistor is toasted often. As the resistor drifts the voltage to ram drifts too and probably contributes to the plague of 5101 RAM failures in these boards. With NVRAM you can put the U8 right on the main VCC and I am pretty sure the 5101 SRAM would be fine like that too. The two 1n4148 diodes, CR5 and CR7 get turned to jumper links and the 82R resistor gets cut. That puts U8 on main VCC. That also puts Q5 on the main VCC but it doesn't seem to effect the the reset sequence.

Voltage dividing resistors R113 and R16 are bit under rated too. I have found boards where R16 has burned open and U14 is dead, probably from over voltage. If the solenoid voltage is at 45v the 2K resistors would be right at the 1/4W rating.

I have noticed not all bridge rectifiers are the same either even if they have the same part number. A Comchip GBPC3506W does not have as substantial of a case as a Fairchild/On Semi GBPC3506W despite having same current rating.

#20 1 year ago

I'm an electrician by trade and my gig for the last year has been as a thermographer. I've had access to some nifty tools, although the FLIR camera I've been using is an antique. Definitely something I'll miss when I leave this shop this week.

Switching to NVRAM probably wouldn't be a bad idea. It's not expensive and would make the thing last a lot longer.

#21 1 year ago
Quoted from semicolin:

I'm an electrician by trade and my gig for the last year has been as a thermographer. I've had access to some nifty tools, although the FLIR camera I've been using is an antique. Definitely something I'll miss when I leave this shop this week.
Switching to NVRAM probably wouldn't be a bad idea. It's not expensive and would make the thing last a lot longer.

Get the FLIR camera for a phone and write it off since you are in the trade
amazon.com link »

#22 1 year ago
Quoted from eh97ac:

Get the FLIR camera for a phone and write it off since you are in the trade

Don't think I haven't thought about it... that said, the FLIR one might be adequate for some folks but not me. Even though it's well worth the price, the resolution (80x60) and range would be impossible to get used to for someone like me trained on the big guns. I've been ruined by $10K+ cameras and I'll never be able to downgrade.

Promoted items from the Pinside Marketplace
$ 3.50
Various Other Swag
UpKick Pinball
$ 49.99
Electronics
PinballElectronics.com
$ 9.00
Electronics
Yorktown Arcade Supply
$ 85.00

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