(Topic ID: 85292)

Bally/Stern AS-2518 Club !

By mof

10 years ago


Topic Heartbeat

Topic Stats

  • 3,868 posts
  • 368 Pinsiders participating
  • Latest reply 6 days ago by tatman9999
  • Topic is favorited by 287 Pinsiders

You

Linked Games

Topic Gallery

View topic image gallery

IMG_1013 (resized).jpeg
IMG_1804 (resized).jpeg
IMG_1847 (resized).jpeg
IMG_1849 (resized).jpeg
IMG_1856 (resized).jpeg
IMG_2083 (resized).jpeg
IMG_2088 (resized).jpeg
IMG_2092 (resized).jpeg
IMG_2091 (resized).jpeg
IMG_2093 (resized).jpeg
IMG_2095 (resized).jpeg
pasted_image (resized).png
IMG_2852 (resized).JPG
IMG_2571 (resized).jpeg
IMG_2849 (resized).JPG
pasted_image (resized).png

There are 3,868 posts in this topic. You are on page 26 of 78.
#1251 5 years ago
Quoted from Coyote:

FTR, this is not how you test the zero crossing. The zero crossing is high/low, pulsing. You would need a multimeter that can report frequency. Better yet, an oscilloscope to see the high/low pattern.
However, since everyone else is sending you off on strobe hunts, I'll be quiet.

Well we're pretty much done with the strobe hunting, although we did see that interesting behavior (the FUTURE letters not pulsing) with the other strobe. Not sure what that means though.

Ok, I have a multimeter and an o-scope (more of a student o-scope on my computer). If someone could tell me what leads go to what for the o-scope, I can take some readings and share them here.

#1252 5 years ago
Quoted from jsa:

Well we're pretty much done with the strobe hunting, although we did see that interesting behavior (the FUTURE letters not pulsing) with the other strobe. Not sure what that means though.

Can anyone explain to me why, when the strobe lines are reversed, lamps would no longer pulse? And why, specifically, three lamps would pulse during their off state?

Also can anyone tell me what I need to connect leads to to properly check the frequency of zero crossing? Thanks!

#1253 5 years ago

The only thing I can think of is some capacitance on the data line or strobe line, so that the strobe lines are too far advanced (or the data lines are too far behind) the strobe, causing the decoder chips to erroneously think they were being addressed, up until the datalines catch up.

#1254 5 years ago

Have you disconnected the Auxiliary lamp board as a test?

#1255 5 years ago
Quoted from Quench:

Have you disconnected the Auxiliary lamp board as a test?

I believe so, early on, and also swapped those boards. I’m going to try it again today though.

@quench if you wanted to see frequency and voltage readings from anything in this chain right now, what would you look for? I’m also going to inspect the zero crossing waveform this afternoon.

Finally, would any measurements taken at the socket help?

#1256 5 years ago
Quoted from jsa:

@quench if you wanted to see frequency and voltage readings from anything in this chain right now, what would you look for? I’m also going to inspect the zero crossing waveform this afternoon.

Hi,
Can you post a video showing the whole playfield (both in attract mode and lamp test mode) so we can see all the feature lamps that are flickering. Please remove the fuse for the GI lamps when you do this - the LEDs in your GI are causing heavy strobing making it hard to see what's going on.
We need to try and build a full picture to see if there's a pattern.

The 22.5VDC you measured at the Zero Crossing TP6 test point on the MPU board indicates it's unlikely to be your problem. An issue here should have resulted in a much lower voltage. To confirm, measure TP2 on the Alltek MPU board - it should be around 43VDC. The Alltek MPU board divides the voltage at TP2 in half which is where TP6 is connected.

If you want to hook up your oscilloscope to visualise it, set the voltage scale on your oscilloscope to something like 10V/division and the time scale to something like 2ms/division.
Hook up the black lead to ground and the probe tip to TP2 at the MPU board.
You should see a signal that looks something like the lower "Output Voltage" waveform in the image pinfixer posted here:

https://pinside.com/pinball/forum/topic/bally-as-2518-club/page/24#post-4607509

Adjust the voltage and time scales until you can see a good overall view of the waveform. Post a picture of it.

#1257 5 years ago
Quoted from Quench:

Can you post a video showing the whole playfield (both in attract mode and lamp test mode) so we can see all the feature lamps that are flickering.

Absolutely. Context, In this case, I have the Alltek MPU board, an original Bally LDB w/ Siegecraft connectors. I have LEDs again in the sockets and Siegecraft boards with 6.3v rail properly verified and connected. GI fuse is out, displays disconnected from the bus, aux lamp driver board disconnected and backbox lamps disconnected from the LDB.

I can replace all of these with incandescents and re-do videos if you request. To date, to test the incandescent flickering, I've "spot replaced" LEDs with incadescents (or groups of them) so I abandoned that line of troubleshooting but can go back if you think it's a good idea.

I've made regular and high speed videos of both attract and lamp test per your request. I find high-speed is easier for me to see with crystal clarity the solid vs. flickering lamps. Be sure to set your YouTube settings to HQ.

First, attract.

Regular speed:

High speed:

Next, lamp test.

Regular speed:

High speed:

Quoted from Quench:

To confirm, measure TP2 on the Alltek MPU board - it should be around 43VDC. The Alltek MPU board divides the voltage at TP2 in half which is where TP6 is connected.

I measure 45.5VDC.

Quoted from Quench:

If you want to hook up your oscilloscope to visualise it, set the voltage scale on your oscilloscope to something like 10V/division and the time scale to something like 2ms/division. Hook up the black lead to ground and the probe tip to TP2 at the MPU board. Adjust the voltage and time scales until you can see a good overall view of the waveform. Post a picture of it.

Screen Shot 2018-10-04 at 12.08.53 PM.pngScreen Shot 2018-10-04 at 12.08.53 PM.png

This is what I get with the probe tip on the Zero Crossing test point:

Screen Shot 2018-10-04 at 11.58.29 AM.pngScreen Shot 2018-10-04 at 11.58.29 AM.png

Let me know if you want me to make any adjustments to that o-scope. I can't get to 10V division, must be a limitation on my student scope.

#1258 5 years ago

I can't get the exact time and volts as your setup on my hand held scope but seems like yours spending a bit more time flat line all the way at the top than my test fixture. Extra capacitance somewhere? What is the value of the capacitor with one leg touching the 21v test point (original MPU would be C1 @ 820pF).

Probe digital zero crossing interrupt signal going to the PIA chip. It should sit slow mostly and every 8.3ms will be a small square.

Your video to me looks like the typical SCR not staying latched type problem.

#1259 5 years ago
Quoted from barakandl:

Extra capacitance somewhere? What is the value of the capacitor with one leg touching the 21v test point (original MPU would be C1 @ 820pF).

I want to answer your question. Do you mean the 470pF one on this portion of the schematic of the Alltek? TP6 is the "Zero Crossing" (22.5VDC) test point.

Screen Shot 2018-10-04 at 4.11.35 PM.pngScreen Shot 2018-10-04 at 4.11.35 PM.png

Just a reminder, the problem exists with either the Alltek or the original MPU, in precisely the same fashion (and either LDB).

Quoted from barakandl:

Probe digital zero crossing interrupt signal going to the PIA chip. It should sit slow mostly and every 8.3ms will be a small square.

Can you confirm we're talking about the 4049 at U12, and what pin I should be testing?

Quoted from barakandl:

Your video to me looks like the typical SCR not staying latched type problem.

Could LEDs on the same address wire that have non-ghosting circuitry in them somehow interfere with the latching? I have no idea how non-ghosting circuitry works, but those are the bulbs I had so those are in there. Alternatively, could the load reduction on OTHER parts of the same lower addressing cause these lamps not to latch?

#1260 5 years ago
Quoted from jsa:

I want to answer your question. Do you mean the 470pF one on this portion of the schematic of the Alltek? TP6 is the "Zero Crossing" (22.5VDC) test point.
[quoted image]
Just a reminder, the problem exists with either the Alltek or the original MPU, in precisely the same fashion (and either LDB).

Can you confirm we're talking about the 4049 at U12, and what pin I should be testing?

Could LEDs on the same address wire that have non-ghosting circuitry in them somehow interfere with the latching? I have no idea how non-ghosting circuitry works, but those are the bulbs I had so those are in there. Alternatively, could the load reduction on OTHER parts of the same lower addressing cause these lamps not to latch?

Probe at U12C P6 in your picture. Should be small squares 8.3ms apart. No random blips and all squares should be the same length. The wrong type of 4049 causes the squares to sometimes be the wrong length (if i remember right).

The alltek schem shows 470pF (471 mark) instead of 820pF like original MPU which is probably fine. Just make sure that value is installed into the PCB and not 470nF or something else bigger. Too much capacitance will break the zero cross.

SCR should "latch" and stay on. If they don't latch they turn back off until the CPU refreshes it again and blinks and goes back off. If the pulse is too short or there is not enough load the SCR may not stay on. Also do note that there is some flickering that is unavoidable because the power supply has no capacitor on the feature lamp voltage. It would be 120hz and that is faster than your eye can see... but your super slow mo camera will notice LEDs flickering at the 120hz regardless of a flicker free solution.

Every lamp position is kind its own thing. One logical address / data combo makes is one lamp turned on by one SCR. The way a LED compat lamp driver would effect other lamps is for the feature lamp bus 6.3vdc to be not connected to the lamp driver board leaving sneak paths to ground. I don't think this is your problem based on your videos.

#1261 5 years ago
Quoted from barakandl:

Probe at U12C P6 in your picture. Should be small squares 8.3ms apart. No random blips and all squares should be the same length.

Does this look right? I think this is what you're looking for.

Screen Shot 2018-10-04 at 6.43.49 PM.pngScreen Shot 2018-10-04 at 6.43.49 PM.png

Quoted from barakandl:

The alltek schem shows 470pF (471 mark) instead of 820pF like original MPU which is probably fine. Just make sure that value is installed into the PCB and not 470nF or something else bigger. Too much capacitance will break the zero cross.

I'm pretty sure that's C78. As far as my failing eyesight can see, it isn't marked with a number:

c78.jpgc78.jpg

However, again, none of this MPU stuff should matter since I've swapped MPUs back and forth with identical results. The problem MUST be in the wiring or connectors, but I haven't figured out how yet.

Quoted from barakandl:

SCR should "latch" and stay on. If they don't latch they turn back off until the CPU refreshes it again and blinks and goes back off. If the pulse is too short or there is not enough load the SCR may not stay on. Also do note that there is some flickering that is unavoidable because the power supply has no capacitor on the feature lamp voltage. It would be 120hz and that is faster than your eye can see... but your super slow mo camera will notice LEDs flickering at the 120hz regardless of a flicker free solution.

Agreed, but I'm talking about visible blinking, not something barely perceptible to the naked eye.

FWIW, I decided to do a little experiment. I replaced with all the FUTURE upper playfield lamps, the FUTURE SPA bottom lamps, and the SPA target lamps with "47" incandescents (basically all the blinking lamps save one). Weird results. The lower playfield FUTURE lamps stopped blinking (greater draw?). The SPA target lamps blinked worse, and very dimly. The upper FUTURE lamps changed...The two Us, R and E were solid, the rest blinking.

I do think this must somehow be SCR related, but w...t...f...

#1262 5 years ago

Your first picture shows the zero cross pulses 16ms apart like it is missing every other zero cross. The squares are also big. Does your scope have some kind of time multiplier? Otherwise i think that looks wrong.
20181004_175800 (resized).jpg20181004_175800 (resized).jpg

#1263 5 years ago
Quoted from barakandl:

Your first picture shows the zero cross pulses 16ms apart like it is missing every other zero cross. The squares are also big. Does your scope have some kind of time multiplier? Otherwise i think that looks wrong.

Maybe I’m looking at the wrong thing. I’m not sure I understand the U12C nomenclature. I’m probing U12 pin 6?

#1264 5 years ago

Your videos show the first 4 outputs of each of the lamp decoders (4514 chips on the LDBs) are flashing. Logic wise it's not a data/address thing, more likely timing.

Your first two oscilloscope readings look fine (time wise) - the periods of those waveforms are around 8.3ms (the time in which they repeat) which translates to a frequency of 120Hz and is right. Does your oscilloscope probe have a "X1 / X10" switch? If yes, put the switch on X10 which basically divides the input voltage by 10. You'll then be able to see the full height of the DC waveform at TP2.

Anyway, as barakandl has mentioned, the zero crossing detector circuit on your Alltek is outputting the wrong frequency (presuming you've probed the correct pin - U12 pin6). It should be the same frequency as TP2 and TP6. The wave shape will look different but the time taken to repeat should be identical.

Have you got a working Bally MPU board you can hookup?

#1265 5 years ago
Quoted from Quench:

Anyway, as barakandl has mentioned, the zero crossing detector circuit on your Alltek is outputting the wrong frequency (presuming you've probed the correct pin - U12 pin6). It should be the same frequency as TP2 and TP6. The wave shape will look different but the time taken to repeat should be identical.
Have you got a working Bally MPU board you can hookup?

I’ll make sure I’ve set this probe correctly and try again, and also I have a working Bally MPU. I’ve swapped it in before with the same results, but I’ll do it again and probe it as well.

#1266 5 years ago
Quoted from jsa:

also I have a working Bally MPU.

The Zero Crossing detector circuit output to probe on the Bally MPU board is pin 4 of U14.

BTW, I think you mentioned somewhere that before the game restoration there was no issue with the lamps. Exactly what's changed (hardware wise)?

#1267 5 years ago
Quoted from Quench:

The Zero Crossing detector circuit output to probe on the Bally MPU board is pin 4 of U14.
BTW, I think you mentioned somewhere that before the game restoration there was no issue with the lamps. Exactly what's changed (hardware wise)?

I've replaced all the sockets. The harness and electronics are the same, except I've re-terminated and re-connectored everything and replaced the rectifier board that was a Frankenstein hack job with a new one from @barakandl.

I keep repeating the U12 pin 6 experiment and keep getting the same result (I feel like I must be measuring the wrong thing). Therefore, I replaced the MPU with the Bally, and measure U14 pin 4. As you can see, here is what we get, 8.3ms:

Screen Shot 2018-10-05 at 8.29.44 AM (resized).pngScreen Shot 2018-10-05 at 8.29.44 AM (resized).png

I'd love to say "Eureka," but nothing changes on my lighting issue. This can't be something on the MPU unless both MPUs are broken, in different places, and creating the same problem, which is super unlikely.

It's got to be something related to the rectifier, backbox wiring/connectors, or the playfield.

When I swapped the strobe wires, the upper playfield "FUTURE" lamps all were lit solid. That told me it's not the playfield, so let's rule that out, which leaves us with backbox wiring, connectors, rectifier. Am I missing any other possibilities?

#1268 5 years ago

One other thing: When all these blinking lamps are replaced with 47s, only the SPA target and upper F T, 2nd U and R blink. Also, the upper U and R seem to stop blinking after about five minutes of operation. That also points to some kind of heat -> latching issue, but I can't find anything in common with those lamps! They are all on AD0, but so are other lamps that aren't blinking (also with incandescents).

#1269 5 years ago

Did you build the rectifier board or was it one barakandl pre-built?

Lets strip this down to bare minimum.
Completely disconnect the playfield from the game - I mean every playfield connector.
Disconnect any connectors to the display/lamp hinged panel in the head box.
Measure the voltage on TP1 of the rectifier board and if it's around the 6.5VDC range, grab a spare lamp socket and connect the base to that TP1. Connect the other end of the lamp socket to pin 18 of J1 header at the lamp driver board (remember the lamp driver board pins are numbered upside down from 1 at the bottom and count upwards). Pin 18 of J1 is the output for the centre target "S" lamp that's currently flickering. Power on the game and go to lamp test mode. Does this lamp still flicker with an incandescent and LED?

#1270 5 years ago
Quoted from Quench:

Did you build the rectifier board or was it one barakandl pre-built?

Pre-built, but as you know, the soldering job of the transformer to the rectifier is all me.

Quoted from Quench:

Measure the voltage on TP1 of the rectifier board and if it's around the 6.5VDC range, grab a spare lamp socket and connect the base to that TP1.

Wait, I'm getting 8.22vdc. Should I continue this experiment?

#1271 5 years ago
Quoted from jsa:

Wait, I'm getting 8.22vdc. Should I continue this experiment?

With no load it will be a bit higher, so no big deal.
But can you first hook up the oscilloscope to TP1 on the rectifier board and post the waveform. Adjust the volts/division on your oscilloscope so you get all the amplitude from the waveform to fit in the image - start from around 2 volts per division.

#1272 5 years ago

First, the waveform you requested:

Screen Shot 2018-10-05 at 9.06.26 AM.pngScreen Shot 2018-10-05 at 9.06.26 AM.png

Is that what you wanted? Next, I performed the exact experiment you asked for. Note that this is with an incandescent light since I pulled off the Siegecraft adapter as well. The light blinks exactly the same way:

#1273 5 years ago

I have to say that as frustrating as this whole thing probably is for jsa, I am very fascinated by the methods taking place. Also big kudos to those knowledgeable enough to help troubleshoot this issue. Hopefully the issue is resolved soon enough.

#1274 5 years ago

Wait! I was measuring the waveform wrong, my ground wasn't connected. Here is the proper waveform:

Screen Shot 2018-10-05 at 9.27.24 AM.pngScreen Shot 2018-10-05 at 9.27.24 AM.png

...and one that fits:

Screen Shot 2018-10-05 at 9.28.41 AM.pngScreen Shot 2018-10-05 at 9.28.41 AM.png

Quoted from desertT1:

I have to say that as frustrating as this whole thing probably is for jsa, I am very fascinated by the methods taking place. Also big kudos to those knowledgeable enough to help troubleshoot this issue. Hopefully the issue is resolved soon enough.

These guys are amazing. I feel awful dominating the topic like this but hopefully this is helpful to others.

#1275 5 years ago

Ok, new waveforms look better About what's expected for a rectified DC voltage.
Can you confirm the 27 ohm ceramic resistor on the rectifier board is getting warm?

I know you've triple checked your connectors but can you post clear pictures of them at the following so we can see wire colors:
LDB J4
MPU J1
Aux lamp board connector J1

#1276 5 years ago
Quoted from Quench:

Can you confirm the 27 ohm ceramic resistor on the rectifier board is getting warm?

Getting warm is an understatement, it just burnt my finger.

Quoted from Quench:

I know you've triple checked your connectors but can you post clear pictures of them at the following so we can see wire colors:
LDB J4
MPU J1
Aux lamp board connector J1

LDB J4:

IMG_3174.JPGIMG_3174.JPG

MPU J1:

IMG_3175.JPGIMG_3175.JPG

AUX J1:

IMG_3177.JPGIMG_3177.JPG

#1277 5 years ago
Quoted from jsa:

Getting warm is an understatement, it just burnt my finger.

Sorry

While I check your wires can you do the following:
Lets manually activate the SCR driving that temporary lamp you've connected to the target "S" lamp connection.
Grab a jumper wire and hook up one end to TP3 on the LDB. Touch the other end of the wire on the "gate" leg of SCR Q14. This will manually illuminate that lamp. With the Bally lamp driver board you've got, the gate leg on the small SCRs is the middle leg going upwards.

#1278 5 years ago
Quoted from Quench:

Lets manually activate the SCR driving that temporary lamp you've connected to the target "S" lamp connection.

@barakandl's rectifier has test pads. Any tips on connecting the TP1 connection without holding it there so I can perform the SCR test?

#1279 5 years ago
Quoted from jsa:

barakandl's rectifier has test pads. Any tips on connecting the TP1 connection without holding it there so I can perform the SCR test?

Nevermind I was able to attach the jumper wire without holding it.

The light stays on SOLID.

#1280 5 years ago
Quoted from jsa:

The light stays on SOLID.

Good. We can put to bed any issues with power and rectifier board.

Wire colors at the connectors look right which I'm sure doesn't surprise you.
At this point we have to check continuity from the MPU board to the lamp driver board *after* the connectors.
With the machine OFF, grab your multimeter and set it to resistance mode. If your meter is a cheap non auto-ranging unit, set it to 200 ohms range.
Check for zero ohms resistance between the following:

MPU left leg of R55 (next to J1 pin 15) <---> LDB top leg of resistor R76
MPU left leg of R53 (next to J1 pin 14) <---> LDB top leg of resistor R77
MPU left leg of R51 (next to J1 pin 13) <---> LDB top leg of resistor R78
MPU left leg of R49 (next to J1 pin 12) <---> LDB top leg of resistor R79
MPU left leg of R57 (next to J1 pin 16) <---> LDB bottom leg of resistor R74
MPU left leg of R60 (next to J1 pin 17) <---> LDB top leg of resistor R73
MPU left leg of R62 (next to J1 pin 18) <---> LDB top leg of resistor R72
MPU left leg of R65 (next to J1 pin 19) <---> LDB top leg of resistor R71
MPU left leg of R47 (next to J1 pin 11) <---> LDB top leg of resistor R75

#1281 5 years ago
Quoted from Quench:

Good. We can put to bed any issues with power and rectifier board.
Wire colors at the connectors look right which I'm sure doesn't surprise you.
At this point we have to check continuity from the MPU board to the lamp driver board *after* the connectors.
With the machine OFF, grab your multimeter and set it to resistance mode. If your meter is a cheap non auto-ranging unit, set it to 200 ohms range.
Check for zero ohms resistance between the following:
MPU left leg of R55 (next to J1 pin 15) &lt;---&gt; LDB top leg of resistor R76
MPU left leg of R53 (next to J1 pin 14) &lt;---&gt; LDB top leg of resistor R77
MPU left leg of R51 (next to J1 pin 13) &lt;---&gt; LDB top leg of resistor R78
MPU left leg of R49 (next to J1 pin 12) &lt;---&gt; LDB top leg of resistor R79
MPU left leg of R57 (next to J1 pin 16) &lt;---&gt; LDB bottom leg of resistor R74
MPU left leg of R60 (next to J1 pin 17) &lt;---&gt; LDB top leg of resistor R73
MPU left leg of R62 (next to J1 pin 18) &lt;---&gt; LDB top leg of resistor R72
MPU left leg of R65 (next to J1 pin 19) &lt;---&gt; LDB top leg of resistor R71
MPU left leg of R47 (next to J1 pin 11) &lt;---&gt; LDB top leg of resistor R75

Everything reads 0.00. I'm hoping I'm using the right setting, we're talking the ohms (omega) setting, correct? It goes from "searching" to 0.00 when I make each connection.

#1282 5 years ago

Yep, ohms uses the omega symbol.

You also have zero ohms between the bottom left leg of SCR Q14 on the LDB and the rectifier board under the big ceramic resistor at the solder tab marked "E8"? This is checking that the lamp driver board has good ground.

#1283 5 years ago
Quoted from Quench:

Yep, ohms uses the omega symbol.
You also have zero ohms between the bottom left leg of SCR Q14 on the LDB and the rectifier board under the big ceramic resistor at the solder tab marked "E8"? This is checking that the lamp driver board has good ground.

Yes, zero ohms.

#1284 5 years ago
Quoted from Quench:

Yep, ohms uses the omega symbol.
You also have zero ohms between the bottom left leg of SCR Q14 on the LDB and the rectifier board under the big ceramic resistor at the solder tab marked "E8"? This is checking that the lamp driver board has good ground.

I’m just playing around, but also there is 2kohm resistance between E8 and the TP6 on the MPU, though I assume that’s expected.

#1285 5 years ago

Just for giggles, swapped to the Alltek LDB again to see if it made any difference (and it did not). 6.3vdc rail connected, etc. everything disconnected, same test to the rectifier board, same blinking.

It's like the game code is doing this intentionally just to piss me off.

#1286 5 years ago

Tried more things. Replaced AD0 line with dedicated, lower gauge wire between MPU and LDB. No impact. Swapped MPUs with same. No impact. Ran power cable to different house circuit I know is clean from RF. No impact. Put on aluminum foil hat. No impact.

#1287 5 years ago
Quoted from jsa:

Put on aluminum foil hat. No impact.

But were you extending out your left hand to the sky while your right eye was closed and your right big toe crossed...

I keep expecting you to find the gremlin! I'm watching with much anticipation wishing I had input that would help.

#1288 5 years ago
Quoted from Atari_Daze:

But were you extending out your left hand to the sky while your right eye was closed and your right big toe crossed...
I keep expecting you to find the gremlin! I'm watching with much anticipation wishing I had input that would help.

I mean, let's look at what's not an exact science: My memory, and foolishness.

My memory: I'm claiming Future Spa didn't flash these lamps before the restoration. I don't REMEMBER them flashing, but maybe all Future Spa lamps have flashed this way and I didn't notice before? Doubt it.

Foolishness: Sometimes it's the stupidest things. If the lamp strobe or address lines are somehow getting impacted by noise, or the decoders impacted by noise, who knows? Just because the wiring is perfect, it doesn't mean there isn't noise. You know, like coming from the solenoid board (oh, first thing I did was unplug that sound board).

What is an exact science? What we actually know.

- We know that once we changed all the lower FUTURE SPA lamps to incandescents, they stopped blinking, why? Maybe because draw is higher?

- We know that when we swapped strobe lines, lamps that blinked in the past stopped blinking. One strobe holds them steady, the other...

- We know with the draw of only ONE SINGLE LAMP on the whole thing, we get blinking. This and the power draw idea are compatible. If there is some threshold related to power draw, people owning Future Spa would only notice it if they went full LEDs and had no older, nasty, power sucking sockets. [Edit: I don't think there are many examples of high end restorations like this of Future Spa.]

- We know the LDB SCR isn't the issue. If it's told to hold, it holds.

- We know the MPU isn't broken.

Things I haven't tried:

- Making a new, dedicated harness just for all the address and strobe lines, and any other necessary J1 pins to light it up. I doubt this makes any difference so I'm hesitating to bother.

- Taking all LEDs of any kind out of this game, putting all 44's, and seeing what happens, until the rectifier board (designed only for 10amps) melts or explodes.

- Building a faraday cage around the MPU and LDB boards.

#1289 5 years ago

Is the zero cross output at U12C still 16ms apart after you got the o-scope going? If so that seems like the problem is with zero cross.

Here is my 21v test point wave form on a held held scope. No time really spent flat at the top like your picture showed.
20181004_175440 (resized).jpg20181004_175440 (resized).jpg

#1290 5 years ago
Quoted from barakandl:

Is the zero cross output at U12C still 16ms apart after you got the o-scope going? If so that is the issue.

When I swapped with the Bally MPU and did an o-scope reading on U14 pin 4 and got your results. Lamps still blink. I’m not sure why measuring on the brand new Alltek gives me those readings but the MPU swap pretty much confirms it’s not that, correct?

#1291 5 years ago
Quoted from jsa:

When I swapped with the Bally MPU and did an o-scope reading on U14 pin 4 and got your results. Lamps still blink. I’m not sure why measuring on the brand new Alltek gives me those readings but the MPU swap pretty much confirms it’s not that, correct?

If the CPU zero cross is interrupting at 16ms then there is twice as much time between when the CPU updates the lamp picture which seems like would be a problem. Since your 21v wave picture showed a period of time where the wave flat lines all the way at the top does not match what i see on normal boards and a different replacement MPU. I think that means the cap (too much) bigger? Ultimately the interrupts at 16ms vs 8ms might not matter to the blinky lamps but it doesnt seem right to me.

Check out the theory of operation around page 7 where it talks about interrupts.

http://arcarc.xmission.com/Pinball/PDF%20Pinball%20Misc/Bally%20Theory%20of%20Operation.pdf

#1292 5 years ago
Quoted from barakandl:

If the CPU zero cross is interrupting at 16ms then there is twice as much time between when the CPU updates the lamp picture which seems like would be a problem. Since your 21v wave picture showed a period of time where the wave flat lines all the way at the top does not match what i see on normal boards and a different replacement MPU. I think that means the cap (too much) bigger? Ultimately the interrupts at 16ms vs 8ms might not matter to the blinky lamps but it doesnt seem right to me.
Check out the theory of operation around page 7 where it talks about interrupts.
http://arcarc.xmission.com/Pinball/PDF%20Pinball%20Misc/Bally%20Theory%20of%20Operation.pdf

It's because I was measuring the wrong thing. I swapped back to the Alltek, measure the correct pin, and get this:

Screen Shot 2018-10-05 at 3.22.46 PM (resized).pngScreen Shot 2018-10-05 at 3.22.46 PM (resized).png

Here's another view:

Screen Shot 2018-10-05 at 3.23.59 PM (resized).pngScreen Shot 2018-10-05 at 3.23.59 PM (resized).png

#1293 5 years ago

One more:

Screen Shot 2018-10-05 at 3.26.28 PM (resized).pngScreen Shot 2018-10-05 at 3.26.28 PM (resized).png

#1294 5 years ago
Quoted from jsa:

One more:
[quoted image]

OK, that is what is expected to be.

Guess zero x is ruled out then.

#1295 5 years ago

While we work on this, please only use incandescents in the problematic lamps.

Some waveforms in Lamp Test mode on SCR Q14: These are steady while the lamp is on or off as noted.
My game is a 10 amp feature lamp game opposed to your 20 amp game, hence my lamp voltages are lower.
Do any of yours differ? Are your waves steady while the lamp is ON?

IMG_0021a (resized).jpgIMG_0021a (resized).jpg
IMG_0017a (resized).jpgIMG_0017a (resized).jpg
IMG_0019a (resized).jpgIMG_0019a (resized).jpg

#1296 5 years ago

Ok, @quench forgive I'm I'm new to using this thing.

Here is the first waveform, middle leg. I get very different visuals depending on settings, but I think this is closest to what you're looking for:

Screen Shot 2018-10-06 at 10.31.13 AM.pngScreen Shot 2018-10-06 at 10.31.13 AM.png

Next there is the lower right leg when on or off. Let's start with a view of all of it (on and off, recorded):

Screen Shot 2018-10-06 at 10.35.13 AM.pngScreen Shot 2018-10-06 at 10.35.13 AM.png

Now let's zoom into when it's on. Here's an "on" spike:

Screen Shot 2018-10-06 at 10.39.39 AM.pngScreen Shot 2018-10-06 at 10.39.39 AM.png

Here's it up closer:

Screen Shot 2018-10-06 at 10.40.05 AM.pngScreen Shot 2018-10-06 at 10.40.05 AM.png

Getting much closer doesn't really reveal much, maybe it's my settings:

Screen Shot 2018-10-06 at 10.38.38 AM.pngScreen Shot 2018-10-06 at 10.38.38 AM.png

Now let's look at off mode:

Screen Shot 2018-10-06 at 10.36.23 AM.pngScreen Shot 2018-10-06 at 10.36.23 AM.png

Just for comparison, here is Q4 bottom right leg, zoomed out:

Screen Shot 2018-10-06 at 10.45.09 AM.pngScreen Shot 2018-10-06 at 10.45.09 AM.png

Here is the on condition zoomed in:

Screen Shot 2018-10-06 at 10.47.28 AM.pngScreen Shot 2018-10-06 at 10.47.28 AM.png

Here is the off condition zoomed in:

Screen Shot 2018-10-06 at 10.48.32 AM.pngScreen Shot 2018-10-06 at 10.48.32 AM.png

Let me know if these are useful.

#1297 5 years ago

Hi, sorry I should have mentioned all three snapshots I took were with 2ms/division time base.
You're getting negative voltages and lots of noise which suggests the earth lead on your oscilloscope probe isn't connected to ground on the machine.
BTW, set your voltage offset to "0V". I've positioned my waveforms so zero volts is at the bottom of the screen. Lets just leave yours with zero volts along the centre of your graph.

#1298 5 years ago

SCR Q14 bottom right leg image when the lamp is off should look the same as your second image from post #1274 (with the per division settings). And this is very close to mine.

FutureSpa_FeatureLampVoltage (resized).pngFutureSpa_FeatureLampVoltage (resized).png

#1299 5 years ago
Quoted from Quench:

Hi, sorry I should have mentioned all three snapshots I took were with 2ms/division time base.
You're getting negative voltages and lots of noise which suggests the earth lead on your oscilloscope probe isn't connected to ground on the machine.
BTW, set your voltage offset to "0V". I've positioned my waveforms so zero volts is at the bottom of the screen. Lets just leave yours with zero volts along the centre of your graph.

Ok, I'm playing with this, I think I've got it right. Maybe we're on to something.

I want to duplicate yours, but I can't seem to pull that off. I can't fit 2ms/division time base on a single screen. Are your playfield lamps connected to the LDB when you do the recording? My noise gets lower when I do. Here is middle leg, 5ms/div time base, offset 0v, and .5v/div:

Screen Shot 2018-10-06 at 11.38.01 AM.pngScreen Shot 2018-10-06 at 11.38.01 AM.png

Here is the lower right leg, first at 20ms/division so you can see the on and off pattern:

Screen Shot 2018-10-06 at 11.42.38 AM.pngScreen Shot 2018-10-06 at 11.42.38 AM.png

Now, here's the same zoomed in to the on, 2ms/div:

Screen Shot 2018-10-06 at 11.44.37 AM.pngScreen Shot 2018-10-06 at 11.44.37 AM.png

You can see the off sometimes has two, sometimes one bump. Here's a zoom in of a two:

Screen Shot 2018-10-06 at 11.45.20 AM.pngScreen Shot 2018-10-06 at 11.45.20 AM.png

Here is a zoom in of a one:

Screen Shot 2018-10-06 at 11.45.51 AM.pngScreen Shot 2018-10-06 at 11.45.51 AM.png

Any better?

#1300 5 years ago

The strobe line is responsible for sending the high and low to Q14, right?

I haven't tried a clean strobe line yet. I reterminated it once, at least, but maybe noise is somehow getting into that strobe (though I don't know why it would effect some and not others).

Promoted items from the Pinside Marketplace
$ 12.00
Playfield - Toys/Add-ons
UpKick Pinball
Toys/Add-ons
From: $ 17.00
Lighting
Gameroom Mods
Lighting
$ 18.95
Eproms
Pinballrom
Eproms
$ 859.00
Flipper Parts
Mircoplayfields
Flipper parts
There are 3,868 posts in this topic. You are on page 26 of 78.

Reply

Wanna join the discussion? Please sign in to reply to this topic.

Hey there! Welcome to Pinside!

Donate to Pinside

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/bally-as-2518-club/page/26 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.