(Topic ID: 296801)

Solved: Gottlieb System 80 - various booting issues

By sparky672

14 days ago


Topic Heartbeat

Topic Stats

  • 18 posts
  • 4 Pinsiders participating
  • Latest reply 13 hours ago by sparky672
  • Topic is favorited by 2 Pinsiders

You

Linked Games

Topic Gallery

View topic image gallery

Screen Shot 2021-07-26 at 4.29.51 PM (resized).png
IMG_8794 2 (resized).JPG
Screen Shot 2021-07-21 at 6.36.31 PM (resized).png
Screen Shot 2021-07-21 at 5.49.53 PM (resized).png
Screen Shot 2021-07-14 at 7.18.16 PM (resized).png

#1 14 days ago

This is a continuation of this old thread that described the following boot issue on my 1981 Gottlieb Volcano (System B, Gen 2). I erroneously thought I had found and solved the issue 8 years ago. Then I left the game sit and only recently got back into it.

Randomly...

1. Turn on switch.
2. Clunk... ball release and ball kicker solenoids engage and stay engaged. Both on same fused circuit. Both also have the under playfield transistors that are triggered by lamp drivers. (I've since added the recommended pull-up resistor to these). The red "extra ball" lamp near the drain lights up.
3. Simultaneous to #2 above, credit display brightly shows "0" and score displays remain blank.
4. Wait a couple seconds, nothing changes; and it would stay like this until fuse blows or solenoids smoke.
5. Turn off switch as quickly as possible.
6. Go back to step 1.

It's getting worse. Some days I never see this happen and other days I have to repeatedly turn the game off/on before I get a normal boot up and can play. During play, it's perfectly fine however. It's only an issue when first turning on the machine.

Today, I dug into the MPU again and although it had some minor battery damage more than 10 years ago that a previous owner fixed. I found and cleaned off more corrosion which led to the following repairs...

1. Replaced the standard socket for Z2 with a strip socket and verified good solder connection on top and bottom side of board. The old chip socket was full of green corrosion.
2. Removed original Z3 and added a strip socket and a new 7404 chip here.
3. The legs of Q4 were almost corroded completely through and cracked off when cleaning. Replaced with a new 2N3906.
4. Resistor R6 cracked off where the lead connects to resistor body. Replaced.
5. Removed, cleaned, and re-soldered the clock crystal and resistors R3, R4, and R5. Physically, the crystal looked fine with only minor corrosion on the bottom ends of the legs where they were soldered.
6. I carefully checked and cleaned off any corrosion, and found no more broken leads or traces.

The booting problem persists... again, if I toggle it off/on several times, I can get a good boot and play normally.

I went through Clay's guide for checking a MPU with a logic probe and discovered that the crystal only shows a clock signal on the bottom leg. The top leg is showing a steady low. For some reason I'm having trouble with the grammar or wording of the guide and just not 100% sure if I'm supposed to see a clock pulse on both legs of the crystal? I did not see a clock signal on pin 1 of Z3 which makes sense since this is connected directly to the leg of the crystal.

clock logic circuit

Resistors R3, R4, and R5 tested good when I had them off the board. I did not replace because I did not have those exact values on hand.

Following more tests in Clay's guide, I am seeing the clock pulse at the other pins as indicated... U4, U5, U6, etc. using the logic probe.

I assume the crystal is supposed to show a clock pulse on both legs... could something like a half-dead crystal explain why it boots sometimes and not others? Timing issues on the power-up sequence? Looking at the schematic, I imagine that half the clock's square wave might be missing. My Fluke has a frequency setting, but 200 kHz is as high as it will read; and I think the buffered clock output at pin 6 of Z3 is supposed to be something like 900 kHz, so I have no idea. I just don't understand how it still plays normally if that's the case. Does this make any sense and can anyone shed any more light on this? Thanks.

#2 14 days ago
Quoted from sparky672:

I assume the crystal is supposed to show a clock pulse on both legs

I don't think this is the case.
IIRC, I could only ever see a pulse from one leg, and only with a scope.
Once you get to Z3-6 you should always be able to see the pulse.
Pin 37 of the 6502 will be pulsing too. Pin 39 provides the phase 2 pulse.

Since some of the parts in the reset section were corroded, and you removed them, I'd scrape that entire section per the PinWiki and replace it with a DS1811.
See: https://www.pinwiki.com/wiki/index.php?title=Gottlieb_System_80#Using_a_Reset_Generator_for_the_CPU_Reset_Section

You'll still need to make sure that Z2/Z3 and the three resistors are doing their jobs in the reset section.
--
Chris Hibler - CARGPB #31
http://www.ChrisHiblerPinball.com/Contact
http://www.PinWiki.com - The Place to go for Pinball Repair Info

#3 14 days ago

Thank you for responding Chris!

Quoted from ChrisHibler:

I don't think this is the case.
IIRC, I could only ever see a pulse from one leg, and only with a scope.

According to Clay's guide, I should also see the clock pulse on pin 1 of the Z3 chip, but I do not. Pin 1 of Z3 is low and directly connected to the same leg of the crystal that shows no clock signal.

At this point, based on what you're saying I don't even know anymore if I have a clock problem. You're saying I should only see a pulse on one leg as opposed to what Clay's guide is claiming (pin 1 of Z3).

Quoted from ChrisHibler:

Once you get to Z3-6 you should always be able to see the pulse.

Yes, confirmed.

Quoted from ChrisHibler:

Pin 37 of the 6502 will be pulsing too. Pin 39 provides the phase 2 pulse.

I can double-check. But a clock signal was present at all locations as per Clay's guide, except pin 1 on Z3.

Quoted from ChrisHibler:

Since some of the parts in the reset section were corroded, and you removed them, I'd scrape that entire section per the PinWiki and replace it with a DS1811.

I'm hesitant to do this modification since the game is playable ... maybe after exhausting other avenues. EDIT: I found the DS1811 equivalent (Microchip Technology's MCP130-460DI/TO) on Amazon and will have it on hand soon. Mod looks fairly simple and will probably do it just to eliminate a lot of questionable board components.

Either the clock is not working correctly (acc to Clay's guide) or maybe it is ... how to determine? I respect you both as experts on these machines.

BTW, the clock logic probe test results are identical when it boots correctly to when it does not boot correctly... rule out the clock?

Quoted from ChrisHibler:

You'll still need to make sure that Z2/Z3 and the three resistors are doing their jobs in the reset section.

Everything was scraped down to shiny copper in this area. Z3 is a brand new 7404 chip (socketed) and Z2 (socketed) checks out good as per tests in Clay's guide. The three clock resistors measured good values on the bench. There is continuity through this assembled circuit and I see the pulse at pins 2 through 6, but Clay's guide says I should see it at pins 1 through 6. I also see the corresponding signals at the connection points at each resistor, so I know there are no breaks in the circuit here. I also did continuity testing when I re-assembled just to verify same.

I have a new crystal coming from Marco in a few days. When I change it, I'll also change the resistors with some new ones.

Clock aside, in your opinion is there anything else that can cause this persistently intermittent boot issue I've described above? Unplugging the credit display and driver board from the MPU to keep from burning out the display and solenoids, the boot issue can still be intermittently reproduced.

PS was rebuilt in 2012 and the connectors replaced. 5 volts is solid and other voltages check out reliably. Ground mods done as well as most of the other recommended mods such as the pull-up resistors. Displays have been repaired/replaced as needed. MPU display segment driver chips have been replaced as needed. Various broken/flaky connections fixed and vias stitched through. None of these clean-ups, fixes, and improvements have mattered... this intermittent boot issue is the one single issue that has plagued this machine since the day I got it, and the main reason it sat unplayed from 2013 until now.

Thanks again for any additional ideas you may have.

#4 14 days ago

It is quite normal that you dont see pulses on the other leg of the crystal. This is because the probe and lead capacitance causes additional load on the crystal and it stops oscillating. If you see good pulses on Z3 pin 6, the oscillator circuit is OK.

I suspect the reset circuit. Either the reset pulse is bad or does not last long enough to let +5V stabilize before letting the CPU start.

#5 14 days ago
Quoted from Tuukka:

It is quite normal that you dont see pulses on the other leg of the crystal. This is because the probe and lead capacitance causes additional load on the crystal and it stops oscillating. If you see good pulses on Z3 pin 6, the oscillator circuit is OK.

Thank you! This seems to make sense to me. And Clay was using a logic probe with a mini scope display so that's probably why his guide does not mention this.

Quoted from Tuukka:

I suspect the reset circuit. Either the reset pulse is bad or does not last long enough to let +5V stabilize before letting the CPU start.

Yes, as per Chris's suggestion I ordered the DS1811 equivalent for the reset circuit simplification. Physically, I see no remaining damage in this area, however, it's probably smart to get rid of most of these older components. Maybe the Z1 CMOS chip is flaking out and became sensitive to temperature?

Today I cannot even troubleshoot further as the game is now booting up perfectly every single time I try.

#6 14 days ago

So it started happening again and I tried troubleshooting the reset section as per Clay's guide.

Using a DMM I checked voltage on pin 40 of U1 while cycling power to the machine. This is a difficult test since the DMM is slow to respond, however, here are my observations:

When the machine boots normally, I see the DMM very quickly jump from 0 to 4.6 volts with nothing flashing in between (most of the time).

When the machine fails to boot, I still see it jump to 4.6, however, I see a quick flash of a lower intermediate voltage before it gets to 4.6 volts (every time). Example: 0... 3.4... 4.6 I know a DMM is not ideal but I repeated this many times. Every time it failed to boot properly, I saw a flash of an intermediate voltage before it got to 4.6 volts. It's all very fast and happening within a second, but happens in all cases as best as I can tell.

As per Clay's guide, I checked pin 34 of the U4, U5, and U6 chips and all are showing the same thing as above.

#7 14 days ago

I am not very familiar with System 80 reset circuit, but System 1 sure was buggy. And the System 3 often needed an additional capacitor on the CPU reset pin to delay CPU starting. So I would not be surprised if there is something glitchy on the sys 80 reset circuit.

TTL crystal oscillator may take some time to start, and also the power supply might take some time until voltage is OK. I think the reset signal should stay low for at least 500 ms at power on.

#8 13 days ago
Quoted from Tuukka:

So I would not be surprised if there is something glitchy on the sys 80 reset circuit.
TTL crystal oscillator may take some time to start, and also the power supply might take some time until voltage is OK. I think the reset signal should stay low for at least 500 ms at power on.

Unfortunately, I have no practical way to measure frequency or timing without my oscilloscope that is nowhere to be found.

Anyway, I looked at the Z2 chip again (clock circuit). It was originally socketed, and I replaced the socket because it was very green and corroded inside. Since the legs of Z2 looked fine, I just re-socketed for Z2 and stupidly put the original chip back. Really stupid, especially since I have a pile of perfectly good 7474 chips sitting over here. I just put a new 7474 into Z2... so far, it seems to be more reliable for now. Since I have a week or two wait before my reset circuit parts arrive, I should know for sure by then. In a couple days, I'll have a new crystal just in case. I'll keep this thread updated.

#9 13 days ago

There is nothing wrong with the clock circuit. You can rule that out.
The reset circuit is suspect.
You can manually ground pin 40 briefly when the game fails to boot. If the game boots after that (every time) then the reset circuit becomes a strong suspect.

I almost always scrape the parts and use the DS1811 equivalent (noted in PinWiki). At the very least, it reduces quite a few parts down to a single component.
--
Chris Hibler - CARGPB #31
http://www.ChrisHiblerPinball.com/Contact
http://www.PinWiki.com - The Place to go for Pinball Repair Info

#10 13 days ago
Quoted from ChrisHibler:

There is nothing wrong with the clock circuit. You can rule that out.
The reset circuit is suspect.
You can manually ground pin 40 briefly when the game fails to boot. If the game boots after that (every time) then the reset circuit becomes a strong suspect.
I almost always scrape the parts and use the DS1811 equivalent (noted in PinWiki). At the very least, it reduces quite a few parts down to a single component.

I had nearly the same problem with Mars: God of War a few months ago. It would only boot about half the time. I determined it was the reset circuit by grounding the reset circuit as described by Chris. Installing the DS1811 was easy and fixed the problem.

#11 13 days ago

Thanks to you all in this thread, there's been some progress today....

Quoted from ChrisHibler:

There is nothing wrong with the clock circuit. You can rule that out.

I swapped out the Z2 (7474) chip today in the clock circuit and it made a HUGE difference. It's booting up more than 95% of the time now. I might explain it away as coincidence, except the failure symptoms have completely changed. When it occasionally fails to boot now, it quietly manifests itself as very bright single random character in the displays... no longer activating any solenoids. By doing nothing else but changing that Z2 chip.

The three original resistors I left in the clock circuit are going to get replaced for sure now.

Quoted from ChrisHibler:

The reset circuit is suspect.

No argument there.

Quoted from ChrisHibler:

You can manually ground pin 40 briefly when the game fails to boot. If the game boots after that (every time) then the reset circuit becomes a strong suspect.

This is an excellent idea but I might not be able to try that now. After swapping that Z2 chip, it's booting up successfully so many times in a row, I will not easily be able to reproduce the failure condition. And when it does fail, I can't leave the displays going that brightly long enough to get to that pin.

Unless I try putting my old 7474 back into it as a test. Maybe later.

Quoted from ChrisHibler:

I almost always scrape the parts and use the DS1811 equivalent (noted in PinWiki). At the very least, it reduces quite a few parts down to a single component.

Yes, I have the DS1811 parts on the way. After studying the modification, I really like the idea of eliminating many of these suspicious parts and getting rid of an entire chip. Thank you for that suggestion.

Quoted from uncivil_engineer:

I had nearly the same problem with Mars: God of War a few months ago. It would only boot about half the time. I determined it was the reset circuit by grounding the reset circuit as described by Chris. Installing the DS1811 was easy and fixed the problem.

Thank you... I plan on doing this. BTW, when yours failed to boot half the time, what symptoms did you see? Displays, solenoids, sounds, etc? Were these the same most of the time?

#12 13 days ago
Quoted from ChrisHibler:

You can manually ground pin 40 briefly when the game fails to boot. If the game boots after that (every time) then the reset circuit becomes a strong suspect.

It failed to boot twice with the display in a safe state (once dark and once with normal brightness) and no solenoids energized. Therefore I had enough time to manually trigger a good boot-up both instances by touching pin 40 of U1 to ground. What a great suggestion. Thank you so much!

#13 13 days ago

When mine failed to boot, I would get nothing on the displays and only random controlled lights coming on. Sometimes it would hold in a coil, but not always.

#14 12 days ago

UPDATE:

I've since installed a new crystal and all resistors in the clock circuit. So in addition to the new Z2 and Z3 chips, there are no original components left in the clock circuit. I also replaced C1 on the MPU.

After two days of messing around, playing pinball, booting and rebooting... two things have become apparent:

1. The booting problem is still there but less frequent and with much different symptoms. The clock circuit is as good as it's going to get.

2. When it fails to boot, I can force it to boot by touching pin 40 of U1 to ground.

In a couple weeks, I'll do the reset circuit modification that will eliminate/replace 20 old components with just 1 plus a few jumpers.

#15 7 days ago

While waiting on my reset circuit chip, things got worse real quick here...

Something on Pinwiki said if the background sound continues to play after game is over (mine does that if I go from a game into test mode) to make sure DIP 25 is working... it needs to be ON. Mine was ON, but I toggled it a couple times and left it ON. I'm sure I did this while the game was powered off.

Then after changing some capacitors on the sound board and doing some playfield work, like adding the capacitors to the slingshot solenoids, I booted up successfully a few times and then bam... nothing.

It comes up to all zeros as if the tilt switch was held closed. All three tilt switches are verified open.

After a bit of troubleshooting...

1. Still have a clock.
2. I have the Reset signal. U1 pin 40 goes high within ½ second. Touching pin 40 to ground does not force game to boot.
3. I have the RDY signal. U1 pin 2 goes to 5 volts.
4. IRQ never goes to pulse. U1 pin 4 stays at 5 volts.

Thinking about DIP 25, I started checking the board around this area and discovered Z15 has a bad gate:

Z15

Tested with a logic probe.

All inputs are the same for all gates.

Input pins 1, 4, 10, 13 are all high.
Input pins 2, 5, 9, 12 are all pulsing.
Output pins 3, 6, 8 are high... but missing any output on pin 11.

Tested Z15 with diode mode of DMM and red lead on ground. Black lead shows ~0.6 on all pins except pin 11, which is showing ~0.01.

While waiting for a new 74LS32 to arrive, could a bad Z15 chip cause a failure to boot as described?

EDIT:

Using the DMM to diode test more chips, the gate at pins 11, 12, 13 on Z14 is also bad.

Z14 and CR28

I tested the diodes for corresponding DIP switches by turning them all off to ensure isolation from the circuit. They all tested good.

#16 2 days ago

Thank you Chris Hibler for the suggestion. I successfully completed the reset modification using the equivalent chip (MCP130-460DI/TO from Microchip Technology). Additionally, there is no doubt in my mind the reset section was causing the most troubles. Three resistors literally just fell apart on removal.

Reset Modification

However, now when I click the power button, the first digit of the credit display very brightly flashes zero for maybe a second before booting normally.

bright zero on credit display

I also recently did the flippers.com anti-boot-thunk modification. However, that doesn't seem to be the cause since it's the same with the Driver board unplugged... a bright flash of zero on power up. In the past, I had only seen something similar on failure to boot.

ChrisHibler, is this what's normally supposed to happen after the reset modification or do I have another issue to track down? Thank you!

#17 15 hours ago
Quoted from sparky672:

ChrisHibler, is this what's normally supposed to happen after the reset modification or do I have another issue to track down? Thank you!

I wouldn't worry about the hot zero.
It is being displayed prior to the CPU getting hold of the board circuits.
The modern reset generator takes about 1/2 second to provide the RESET signal, and that is why the hot zero is lingering. No biggie.
--
Chris Hibler - CARGPB #31
http://www.ChrisHiblerPinball.com/Contact
http://www.PinWiki.com - The Place to go for Pinball Repair Info

#18 13 hours ago
Quoted from ChrisHibler:

I wouldn't worry about the hot zero.
It is being displayed prior to the CPU getting hold of the board circuits.
The modern reset generator takes about 1/2 second to provide the RESET signal, and that is why the hot zero is lingering. No biggie.

Ok thank you! And thanks again for pushing me to do this. No more booting issues.

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