Game voltages--is this normal?

[demogo]

OK, this is the follow up message to this earlier thread:

http://forums.arcade-museum.com/showthread.php?t=164260

OK, I just bought a pricey new power supply hoping that by throwing money at the problem I could fix it.

I have tiny leads on the bottom left and top right legs of a game eprom running to my DMM.

Here's what I'm seeing now. Is this normal? Should I be concerned?

When I powered the game up I was seeing 5.10 volts for a second or two and then it dropped to about 5.03 as the game booted. It stayed at 5.03 volts for a few minutes while it ran through its demo a time or two. Game is happy. I went around to the front to double check the voltage readings at the CP lines running to the buttons. Same numbers there.

At some point a few min later I come back and check the voltages at the buttons again and the game is getting 4.83 volts. WTF? I go around to the EPROM again and I'm also getting 4.83 volts there (it fluctuates a tiny bit around 4.83 - 4.85 while the demo is going). Game is still happy.

I power game off and back on again. Voltage dropped to 5.03 after a second or two and stayed that way for about 30 seconds and then dropped back down to 4.83 and then stayed right there. Game is happy.

I changed the battery in my DMM. Didn't change anything.

Watcha think?

So should I be concerned about this, especially given the weird behavior I was seeing with the previous power supply???

 

[demogo]

Bumpety bump

 

[modessitt]

Seems to be running lower once the load gets bigger. If it's running fine, i'd say leave it alone. If you just HAVE to mess with it - and the PS is adjustable - then wait until it stabilized a 4.83, then bump it up to 5.00

 

[demogo]

I'd rather not mess with it if it's not going to cause me any grief.

The thing that concerns me is that for the first few minutes after it's turned on it'll be getting close to 5.20 volts until it settles down. Not sure I like that idea.

Honestly, this explains some of the misbehavior I was seeing with the last power supply with the voltage dropping. OTOH, with the last one it WAS crashing with the voltages dropping too low and that was w/o any adjustments over time so I think something really was wrong with it.

 

[demogo]

Daytime bump.

 

[YellowDog]

That seems to be a normal operational feature of switching power supplies (vs linears). Under load they will drop voltage in order to supply sufficient amperage. In modern applications (i.e. newer computer boards), the boards don't draw that high an amperage, so the power supplies remain more stable. With older video games and especially pinball games they draw a lot more power compared to the modern boards.

Also, I think the Gauntlet uses a video card with a fan, if it is the same setup as the Gauntlet DL. I have noticed that the CPU fan and the video card fan don't kick on immediately so they may be demand driven. It takes a few minutes to warm up and then the fans kick in. Since they are powered off the same power supply, they may be increasing the power draw and lowering the voltage. Just a thought.

ken

 

[whammoed]

What is happening with the voltage at the power supply connection? Does it fluctuate too? If it fluctuates as well, then it's not regulating the voltage very well. If it is only fluctuating at the pcb, then maybe something is changing in resistance as things heat up like a poor connection or cold solder joint.

 

[demogo]

Thanks guys.

I still have the game opened up -- I'll check at the power supply tonight and post an update later tonight.

 

[demogo]

OK, I took a 6 month break on my arcade repair activities because I was burning out on them and wanted to restore the fun factor.

I have 3 Jamma extension cables in this system, one hooks up to each of the game boards (NBA Jam and Gaunt Leg) and they run to the front under the CP, one hooks up to the jamma harness and runs to the front under the CP. I can switch games by powering things off and simply swapping cables under the CP and all of the signals switch perfectly. At least in theory.

So what I noticed in the past is that I'd set the voltage and then over time (sometimes 3-4 sessions later, sometimes 15 min later) both NBA Jam and Gauntlet Legends will reset and when I check the voltage it's dropped from 5.0 to 4.8 or less so the reset is as-expected. The voltage is screwed up but the reset isn't surprising.

Replaced the power supply; no change. Replaced marquee light/ballast/starter; no change. Went through this issue a number of times and it's very much pissing me off.

So fast forward 5.5 months and I decided to get methodical to track down this problem. Here's what I discovered:

Disconnected all Jamma extension cables, adjusted to 5.00 volts and ran GL by itself for a number of sessions, just as if the game was unmodified and everything ran great.

Hooked up Jamma extension cable 3 to NBA Jam and ran it to the cab jamma harness, adjusted the voltage to 5.00, and ran it for quite a few sessions over a few weeks and everything was fine. Voltage fluctuated between 5.00 and 5.04 but was almost always around 5.02. Switched my DMM hook up points to a game ROM legs for maximum accuracy. Just bought a new Extech DMM and hooked it up at the same time as my cheapo Fry's DMM and both DMMs were close (Extech read .02 volts lower at all times, even as the voltage fluctuated slightly).

Tonight I unhooked Jamma extension cable 3 and swapped in Jamma extension cable 2 in its place and w/o adjusting the power supply I immediately got 5.15 volts at startup. Only change was swapping cables. Aha -- I have it cornered!

Then I swapped out Jamma extension cable 2 and swapped in cable 1 and got 5.13 at startup. What??? Does this mean the problem was with cable 3 all along?

So I swapped cable 3 back in the exact same way that I had it before and at startup I was getting 5.11 volts.

WHAT?

I also noticed something suspicious that may be related. Someone had layered solder and built it up a bit on a 5v finger on the jamma connector on the GL board, apparently fixing a burned or bad/scratched finger. That can't really explain the identical issues on the NBA Jam board though unless the hack repair is somehow damaging the jamma extension connectors or the jamma connector for the cab.

Thoughts anyone as I continue to attack this problem? What's the next thing to try?

I'll go test continuity on the jamma extension cables as I wait to hear back from you.

 

[demogo]

OK, a few more details.

I checked continuity on my jamma extension cable #3 with special focus on the power section and I got tone on everything.

The pins/connectors looked okay with a visual inspection.

It appears that the cabinet jamma cable power section only has 3 +5 pins hooked up; one is missing and not obviously clipped off or anything.

Thoughts? Ideas on what to check next?

 

[demogo]

Here's what the power connectors on one side of Jamma extension cable #2 look like:

Note the blotches. Not sure if this is corrosion, dirt, scrapes, or what, or if anything needs to be done with it. I get a good continuity tone on those with my DMM.

This may be a side issue since harness #2 wasn't involved in all of the testing noted above but it's still interesting.

 

[mecha]

the Gauntlet Legends "repair" on the +5 fingers is a very common problem with Vegas boardsets. it was actually so bad on my Sportstation I had to replace the input board ($100). I too tried the solder method on the old board first, but that's not going to do dick because it's not actually connected to anything, it's still going to be making shitty connection underneath the layer of solder.

now what I also observed was that I would make voltage adjustments at the power supply and they would in fact change when I was using the bad input board, and the power supply never got knocked around or moved, and it was a brand new unit back in November, so it's not a bad +5 volt adjustment pot. I don't know if it tries to compensate for the increased resistance built up on the +5 lines on the JAMMA edge or what. for fun, I would meter what your power supply is adjusted to at the molex plug, then the JAMMA edge itself, then the hard drive plug (I've found that you will get the same exact voltage between a boot rom and the hard drive plug, so meter there for simplicity). I had a difference between 5.4V at the JAMMA edge and 4.79V at the components on the board, that's how bad the 5 volt fingers were.

no degree of power supply adjustment would bring the voltages high enough to where the game wouldn't eventually reset, and this was after I pencil erased the fingers and was able to get satisfactory results for awhile before the fingers started to lift.

JAMMA extensions are also bad because they'll create longer travel for power to reach the board, just more resistance. I wouldn't be surprised if you had to have the power supply cranked to get 5.02V or whatever you're shooting for at the board. I personally try to make sure I get 5.10V at board components, that'll allot for the voltage tolerances and drops (particularly when a lot of amplifier activity is going on I think?) and your game should never reset. most games will work fine as low as 4.9-ish volts, really just depends on operating voltage tolerances of components.

see how much of a drop you're seeing on that Gauntlet Legends first. maybe try NBA Jam on all 3 of your extensions alone to see if there's any change. are all 3 of your extensions running the same gauge wire on the power sections? there's some that will be like 18 or 22 depending on who makes them.

 

[whammoed]

Are you testing voltages just at the game boards or at the power supply also? If it is dropping at the power supply too, maybe it's a bit overloaded and can't regulate properly?

 

[YellowDog]

One thing I am not sure about from your description, are you running both games at the same time and just switching the controls? If so, that may be your problem, you are overloading the power suplly. When you mount dual (or more) boards, each should be either powered separately or the power should be applied serially. This is commonly done by either moving the power cables, or mounting 2 separate power supplies and switching the AC to each via a three position toggle switch (on-off-on), so that left powers board 1, center is off, right is board 2.

ken

 

[demogo]

the Gauntlet Legends "repair" on the +5 fingers is a very common problem with Vegas boardsets. it was actually so bad on my Sportstation I had to replace the input board ($100). I too tried the solder method on the old board first, but that's not going to do dick because it's not actually connected to anything, it's still going to be making shitty connection underneath the layer of solder.

now what I also observed was that I would make voltage adjustments at the power supply and they would in fact change when I was using the bad input board, and the power supply never got knocked around or moved, and it was a brand new unit back in November, so it's not a bad +5 volt adjustment pot. I don't know if it tries to compensate for the increased resistance built up on the +5 lines on the JAMMA edge or what. for fun, I would meter what your power supply is adjusted to at the molex plug, then the JAMMA edge itself, then the hard drive plug (I've found that you will get the same exact voltage between a boot rom and the hard drive plug, so meter there for simplicity). I had a difference between 5.4V at the JAMMA edge and 4.79V at the components on the board, that's how bad the 5 volt fingers were.

no degree of power supply adjustment would bring the voltages high enough to where the game wouldn't eventually reset, and this was after I pencil erased the fingers and was able to get satisfactory results for awhile before the fingers started to lift.

JAMMA extensions are also bad because they'll create longer travel for power to reach the board, just more resistance. I wouldn't be surprised if you had to have the power supply cranked to get 5.02V or whatever you're shooting for at the board. I personally try to make sure I get 5.10V at board components, that'll allot for the voltage tolerances and drops (particularly when a lot of amplifier activity is going on I think?) and your game should never reset. most games will work fine as low as 4.9-ish volts, really just depends on operating voltage tolerances of components.

see how much of a drop you're seeing on that Gauntlet Legends first. maybe try NBA Jam on all 3 of your extensions alone to see if there's any change. are all 3 of your extensions running the same gauge wire on the power sections? there's some that will be like 18 or 22 depending on who makes them.

Thanks -- will check voltages at the molex plug and jamma connector although I think I'll have to check at the jamma connector with it unplugged and no load on the PS. I tried checking it once before with the connector half unplugged so I could reach the contacts to read them but I got flaky results from it being half-unplugged.

I definitely have to turn up the voltage when using the jamma extension cables. I know that's expected but it should be consistently higher; not apparently stable and then suddenly different.

The jamma extension harnesses are from Bob Roberts so they are high quality and I think 18ga power. I tried some jammaboards extension harnesses first and they were complete shit.

Regarding your suggestion of trying all of the cables with NBA Jam, that's exactly what I'd set out to do. Got cable 3 set, p/s adjusted and it was very stable for a while. Then readings were radically higher with cable 2 and cable 1 and then cable 3 itself again with no other changes.

This smells like a physical problem to me -- either a bad connection or a loose wire or something...

Thanks for the tips -- will try them when I return from vacation.

 

[demogo]

Are you testing voltages just at the game boards or at the power supply also? If it is dropping at the power supply too, maybe it's a bit overloaded and can't regulate properly?

Haven't tested at the p/s in a long while; been testing on rom legs on the nba jam game board lately.

Will check it out at the power connectors and start tracking the values there too.

 

[demogo]

One thing I am not sure about from your description, are you running both games at the same time and just switching the controls? If so, that may be your problem, you are overloading the power suplly. When you mount dual (or more) boards, each should be either powered separately or the power should be applied serially. This is commonly done by either moving the power cables, or mounting 2 separate power supplies and switching the AC to each via a three position toggle switch (on-off-on), so that left powers board 1, center is off, right is board 2.

ken

Ken,

No, just one game at a time is connected to the cab jamma cable. I believe the power feeds directly from the power supply to the cab jamma cable and then a game (or extension harness) plugs directly into that.

From the cab jamma cable (lately) I've been testing with a jamma extension cable to nba jam.

I did have multiple cables in play before but am trying to simplify my test cases to nail this down.

 

[demogo]

OK. 4.99 at the rom, 5.19 at the molex power connector, 5.16/5.17 on the female cab jamma connector plugged into a jamma extension harness, and 5.08v where the extension harness plugs into the NBA Jam board.

This is after I turned the p/s back down to get back to 5.00v (it was reading about 5.15 earlier with the extension harnesses and no other differences).

I guess I'll use this as my baseline and we'll see what happens.

 

[R3LL1K]

Have you considered switching to a computer PSU?

 

[Hewitson]

Why would anyone want to replace a proper adjustable arcade power supply with a computer one?