Gyruss Needs to "Warm Up"

[DZA]

@mecha (I have seen your posts about games like Gyruss needing extra power to the boards, so I figured I would list you in this post)

When I start my Gyruss "cold", it boots to garbage (actually a screen full of upside down "A" characters). Eventually, it will work its way out of that and play fine. It takes about 2-5 minutes.

I could tell that the voltage was low at spots on the board. So I ran another set of wires to the lower board from my power supply (as @mecha has mentioned in the past). I figured that would fix it. Unfortunately I am still having issues. It still seems to take awhile to "warm up" the boards.

I have the voltage such that I get 5.10 volts at the supply, 5.07 volts at the header on the top board (I cleaned it), 5.05 volts at a ROM on the top board, and 5.05 volts at a ROM on the bottom board right out of the gate and it still does not get going. It eventually settles at 5.22V, 5.18V, 5.14V, and 5.14V respectively. If I turn the game off and then turn it back on, it works instantly. If I wait 10 minutes or so, the process starts again. I am not sure that these values are super important at this point since I can turn my supply up, get higher voltages all around the board, it is still does not boot right away.

I also recapped the major caps on the board initially thinking this might be the issue but I saw posts about directly hooking up the lower board and figured that had to be my problem.

Anyway, I am guessing it is some component issue at this point but would be open to ideas. The game plays fine but it bothers me that this does not work and I worry it is indicative of another more serious failure to come.

 

[KenLayton]

Sounds like bad capacitors in the power supply.

 

[DZA]

Sounds like bad capacitors in the power supply.

Interesting. Yeah, I'll try another one and report back.

 

[ArcadeTechGW]

Rebuild your power supply sir! Get the right caps and semiconductors and do a sweep.

 

[andrewb]

Those voltages are fine. If it's a bad PS, then you should be able to measure that to confirm. (And as a result, I don't buy that idea.) As long as you're getting at least ~4.80V on the board, the logic should be happy, from a DC perspective.

That said, you should measure the AC voltage on all PS voltages, to see if there could be a bad filter cap in the system, which a DC measurement won't pick up. But if it's ok, I wouldn't waste your time on PS parts. You want to measure, find, then replace, when it comes to troubleshooting.

What's more likely is a marginal logic chip on the board (could be RAM, or anything else, but RAM is a good place to start). Or perhaps a cracked joint somewhere (which is closing once the board warms up).

If so, I'd use the ice cube trick to test all of the chips (by putting an ice cube in a ziploc bag, and touching it to each chip, once the game has warmed up, to cool each chip down.) I'd also inspect the board (first) under a bright light, and look for cracked joints. Note that you'll want to separate the two board to do this, and inspect the solder sides.

I'd also remove all socketed chips, clean the legs with a Dremel with a wire wheel brush, paint some DeOxit D5S on the legs, and reseat. This is something you should do to any board as a starting point for any debugging, as they're all oxidized and dirty after 40 years, and you could be getting a bad connection in any socket. Replacing the sockets is not necessary. You always want to establish a good foundation (good power, good CPU, good ROMs, good connections) before doing more in-depth troubleshooting.

If you want to test the board, you can send it in, as I have a Gyruss cab. It'll cost you ~$20 round trip shipping, if you just want me to check it in my cab.

 

[parism]

In addition to Andrew's suggestion, you can flex or put some pressure on the board after it has warmed up and running. A bad/cold solder may cause the game to fail while pressure is applied, after it is warmed up.

Borderline working ICs are difficult to identify. But, since you have a way to reproduce the symptoms consistently, you are halfway there to figuring out the culprit.

 

[DZA]

Wow! Thanks for all the replies.

I am happy to report that it indeed was a bad power supply. When I took it off and looked at it, I could actually see inside and the caps were split open. I put another one in and it works great.

Again, since the voltages were within range just thought the supply was fine. However it did always have a very deterministic warm up time that made me think it was literally something building up a charge. That is why I replaced the caps on the board. Turns out I fixed the wrong caps.

Thanks all and now back to my MCR power supply - https://forums.arcade-museum.com/threads/tapper-resetting.477705/

 

[DZA]

P.S. I actually got this game a while ago and am just getting to it now. I actually got it for cheap as it was "not working" as the owner never would wait the 5 minutes for it to warm up (just assumed it was scrambled). Always nice when the fix is swapping in a $30 supply...

 

[andrewb]

Did you measure the ripple voltages per above? That would have told you if the caps were bad, and on what supply.

It would have been good to get a baseline for that a 'bad' ripple measurement is in one of these. (I know the values for Atari supplies, but not these.)

 

[demogo]

Did you measure the ripple voltages per above? That would have told you if the caps were bad, and on what supply.

It would have been good to get a baseline for that a 'bad' ripple measurement is in one of these. (I know the values for Atari supplies, but not these.)

The voltages seemed okay so what could have been wrong with the p/s other than AC ripple? Not supplying enough current?

 

[DZA]

Did you measure the ripple voltages per above? That would have told you if the caps were bad, and on what supply.

It would have been good to get a baseline for that a 'bad' ripple measurement is in one of these. (I know the values for Atari supplies, but not these.)

I did not. I still have the supply. I could do that and report back for the sake of others in the future.

 

[andrewb]

The voltages seemed okay so what could have been wrong with the p/s other than AC ripple? Not supplying enough current?

The DC measurements only give you part of the picture. They tell you the average voltage value, but don't tell you anything about how clean it is. If any supply has a lot of AC noise on it (which happens when the filter caps aren't filtering properly), the board can be affected, as the voltage isn't actually the stable voltage that the DMM is reporting.

A DMM is a low-resolution device. It's really just measuring the average voltage value, when in DC mode. The voltage could be bouncing around between 0 and 10V, and the average will still be 5. But the systems on the board need consistent, stable voltages to work properly. That's why you really want to always do some kind of AC measurement (depending on the game) when testing power, as the AC value tells you how well the filter caps are filtering.

I'd have to look at the Gyruss schematics, but sometimes there are power-on reset circuits (as Atari games have) that are powered by other voltages (like 12V), so you can have ripple on just that supply, that will cause the board to not boot. That's actually a common issue on Atari games when the big blue goes bad. But knowing how much ripple there is on each voltage will give more insight as to how the ripple is causing the board to not boot. (It could just be as simple as the 5V not being clean enough.)

 

[lilypad19]

I just picked up a dead Gyruss this past Monday. I was surprised the original PS was a switcher (PSR-703AE). It was dead (no +5, -5 or +12). A little testing and R8 was dead. It sits between the bridge rectifier and pretty much everything else. New 330K resistor and it's 100%. I did test for AC ripple and there is nothing on mine. Not applicable here - but may be useful in the future for someone.

 

[DZA]

I just picked up a dead Gyruss this past Monday. I was surprised the original PS was a switcher (PSR-703AE). It was dead (no +5, -5 or +12). A little testing and R8 was dead. It sits between the bridge rectifier and pretty much everything else. New 330K resistor and it's 100%. I did test for AC ripple and there is nothing on mine. Not applicable here - but may be useful in the future for someone.

Thanks for sharing. Any chance you have a pic of the power supply. I am curious what it looks like.

Another Gyruss in MA saved

 

[VectorCollector]

The biggest take-away here, is people need to own a quality RMS DMM and get used to testing both DC voltage and AC ripple on the DC supplies.
This gives the best understanding that either a LINEAR or a SWITCHING power supply are providing in-spec voltage rails to the logic boards.

 

[Interlock]

AC ripple can screw all kinds of things up. It can do some bizarre things.

Have a Gottlieb system 1 pinball machine that would play fine except when the right drop target bank needed to reset, at which point the board would freak out and do a full reset. It was the ripple capacitor on the power supply. These circuit boards tolerate it to a degree (and weird shit can happen), until it is untolerable, and that is when the really weird shit happens!

 

[lilypad19]

Here it is - its pretty nice. I still have it opened up. I have the 1/4w I tested with installed - waiting on a 1/2w 330K 1% to show up. Once I replaced it - everything worked. My old DMM was pretty good, but I demoted it and got the Fluke. Well worth the money. Auto ranging alone is worth it to me.
All the caps look to be original, but I get no AC on any of the outputs so I'm not touching them. The ones that would test in circuit correctly with my ESR meter were well within spec.

My game board fired up on the spare switcher (before I fixed this one) - with bad graphics. I popped, cleaned and tested all the eproms and 3 check bad. Waiting on 2764's so show up to confirm that is all. It plays - but most of the screen (except the planet in attract mode) is messed up.

 

[DZA]

The biggest take-away here, is people need to own a quality RMS DMM and get used to testing both DC voltage and AC ripple on the DC supplies.
This gives the best understanding that either a LINEAR or a SWITCHING power supply are providing in-spec voltage rails to the logic boards.

Agreed. I will be adding this to my testing repertoire in the future.