MV1C problem with battery

[littlecharlie]

Did you try replacing your 74HC32?

Not yet. I have the components but haven't got the time to make the replacements (I have never replaced smd components so it doesn't help)
First I'll try the 74HC32 and then the back up RAMs.

Another point is that, out of the 3 mv1c that I have, the 2 in which I added the unibios (with the neobiosmasta) are the ones that drain faster the battery. Will try to find if it is related.

 

[BilboBaggins]

Having a similar problem, but I believe my issue is different. Have a thread here.

4 slot board in a 2 slot cab. Installed coin cell battery and can't get anything to save - high scores, calendar settings etc. I shouldn't have to let it charge since it's coin cell right? How do I diagnose this? Thanks.

 

[shadowkn55]

Having a similar problem, but I believe my issue is different. Have a thread here.

4 slot board in a 2 slot cab. Installed coin cell battery and can't get anything to save - high scores, calendar settings etc. I shouldn't have to let it charge since it's coin cell right? How do I diagnose this? Thanks.

Did you remove the 470 ohm charging resistor?

 

[BilboBaggins]

Did you remove the 470 ohm charging resistor?

Thanks yes I did do that. I may have installed the battery backwards...

Going to retry. Curious though - would buying a memory card circumvent the need for the backup ram? Or does a memory card only service high scores / game settings etc.?

 

[shadowkn55]

Thanks yes I did do that. I may have installed the battery backwards...

Going to retry. Curious though - would buying a memory card circumvent the need for the backup ram? Or does a memory card only service high scores / game settings etc.?

No. Game settings are only saved in backup ram.

 

[BilboBaggins]

Got it working thanks!

Sent from my SM-N920V using Tapatalk

 

[jlogo]

Just an update on this in case anyone has the same issue. I ended up finding a radial diode lying around that I installed on my board in the D4 position and my battery draining issue has been completely resolved! Next time I place a parts order, I'll get the proper SMD type, but for now, it's working great. Thanks to all who provided suggestions and ideas.

Hi, could you tell me the specs of the radial diode did you install?

 

[ledfrog]

Hi, could you tell me the specs of the radial diode did you install?

Ideally you would want a surface mount diode to make a clean install. I'm not too proficient with diodes, but I assume you just need one that is rated for at least 3.6v. The one I used was bought for a retro Game Boy project I was working on and the model number is: MUR160G. I found it on Mouser and according to the spec sheet, it's rated for 600v. But like I said, this one has leads sticking out, so it doesn't look too pretty the way I installed it!

 

[jlogo]

Ideally you would want a surface mount diode to make a clean install. I'm not too proficient with diodes, but I assume you just need one that is rated for at least 3.6v. The one I used was bought for a retro Game Boy project I was working on and the model number is: MUR160G. I found it on Mouser and according to the spec sheet, it's rated for 600v. But like I said, this one has leads sticking out, so it doesn't look too pretty the way I installed it!

Thank you.

I changed the D4 diode for a 1N4004 and everything worked but the the battery leak still was there.

At the end I fixed the battery leak removing C3. after pull out that capacitor the leak was gone.

 

[ledfrog]

Thank you.

I changed the D4 diode for a 1N4004 and everything worked but the the battery leak still was there.

At the end I fixed the battery leak removing C3. after pull out that capacitor the leak was gone.

Yeah it seems like this battery circuit has a bunch of different things that can go wrong with it. Hopefully this thread will prove useful for anyone in the future. Glad you got it working.

 

[joala]

Finally got around to looking more into this. I think you're right. As a reminder, when my board is powered on, I'm getting 5v on the + battery terminal with no battery inserted. After removing D4, the 5v was no longer present on the empty battery holder so it looks like I've rooted out my problem. Of course now, with D4 removed, the battery is also not able to provide power to the backup ram, so I need some help.

Can anyone report back the exact diode that's supposed to go in D4? I was able to make out "S.4 97" written on it, but I'm not too sure what that means...I don't have much experience with diodes.

I've been having another crack at this and testing out your initial issue and I'm still getting 5v on the + terminal, powered on with no battery even AFTER removing D4 - any ideas what else could be causing the 5v here?

I also tested the rest of the diodes, they're all good but D3 reads voltage in both directions when soldered to the board but when I removed it to test, it only reads in one direction as expected. I swapped it out for a new one and get the same reading....

 

[ledfrog]

I've been having another crack at this and testing out your initial issue and I'm still getting 5v on the + terminal, powered on with no battery even AFTER removing D4 - any ideas what else could be causing the 5v here?

I don't remember if it was mentioned early in this thread or if I read it somewhere else, but I've heard that these boards are multilayered and that this voltage leak could be caused by a short on one of the other layers.

 

[Beharius]

What is the value of D3?

 

[Xian Xi]

A friend and I eventually nailed down this issue, remove C3 and the two bypass caps on the BRAM. No more parasitic drain. You can replace the bypass caps on the BRAM but they aren't really needed.

 

[Beharius]

My symptoms are as follows:
I can get around 4-5 volts from the battery terminals when there is no CR2032 battery installed. (Charging circuit disabled)
I also noticed that the 12th and 13th pins of the HC32 outputs around 5volts when the system is running. So it seems I have to replace the HC32 and 0.1uf caps around to solve this issue.

 

[Beharius]

So I tested the HC32 with TL866 programmer and it's failed. Battery terminal still shows ~5v although HC32 and all the caps removed .

UPDATE : Replacing the D4 diode seems to solve the issue. Thanks to everyone...!

 

[noir]

A friend and I eventually nailed down this issue, remove C3 and the two bypass caps on the BRAM. No more parasitic drain. You can replace the bypass caps on the BRAM but they aren't really needed.

Are the bypass caps PC2 and PC4 on the top of the board?

 

[avramce]

Came across this thread as well, I have a handful of boards with the same issue, where after getting a full charge would drain the battery from 3.4V to 1.2 in the matter of 1-2 minutes. I removed C3 and PC2,4 as recommended, and the circuit appears to be retaining charge a better (drained from 3.1V to 2.8V over the course of 30 minutes).

Is it work replacing any of the removed caps? Or should I be checking the HC32? What's the typical length you'd expect the coin cell to hold a charge, as the batteries on multi-slot boards can easily go months and still hold voltage in my experience.

 

[Xian Xi]

I’ve had coin cells last for 1.5 years before because the drain was so low. Sometimes the type of RAM used affects the drain as well. But as long as you get 6-12 months from a non-rechargeable, I think that is pretty good.

The only thing in the circuit that can drain the battery are diodes, RAM and capacitors. I wouldn’t worry about the HC32 since that’s only used when the power is on so it doesn’t run off battery power.

 

[lithy]

Bumping an old thread just since most of the info I'm using is in here.

I'm working on an MV-1A that was not holding settings.

It shows some evidence of previous work (a repair wire in the back-up area, a few repaired traces elsewhere, and some scratched solder mask here and there) so it's not the cleanest board overall, but it has worked well and I'm using it as a kind of dedicated board to use with my Vortex cart and using real carts on my 4 slots.

I currently have it with a coin cell holder and an ML2032 with the charging circuit left intact. So I ran the board for a couple hours to put some charge into it. Then brought it upstairs and checked with a multimeter and sure enough it was ticking down 0.01V every handful of seconds.

I know most of this thread relates mostly to 1C rather than 1A boards, but I figure many of the components are the same. So I started by removing C3. I got immediate improvement in the speed of the drain. Went from 2.45V to 2.37V in an hour

So my current questions. Do I need to replace the component (I don't even know what it is lol) at C3? And next, even 0.08V drained in an hour means the battery would be dead in a day if it wasn't used so what should be my next move? I could keep removing components but I'd like to troubleshoot if I can. Appreciate any suggestions here as to what to test or look for.

Thanks!

Some photos for reference

Top side of back-up area showing battery and wire repair



Bottom side of back-up area showing now removed C3 and some previously removed solder mask



Outside the back-up area, I looked at PC2 and PC4 in case they are the same ones being mentioned on the 1C, but it looks like I don't have a component at PC2 in place. I did not remove this so if it was ever there it has already been removed. Same for PC26 and PC31 (both not pictured) and PC36. The MV-1A board scan on the dev wiki seems to show a component at PC31 and PC36 but also doesn't show one at PC2 or PC26. That's all I noticed, in case any of that is relevant.



Also, I found this thread with a similar issue but limited info so I tossed my post in here.

MV1-A & MV1-FZ battery mod

Hi, I am Vendest newly registered. I am facing an issue with the battery type CR2032 installed on 2 nos. 1 Slot MVS boards that I have consolized recently. I have followed thoroughly the tutorial and disable the recharging function by removing the necessary components. The battery does not...

www.neo-geo.com

 

[lithy]

Alright well I took PC4 off just because, what the hell.

Then I stuck it back on the cab and let it run on KOF98 attract mode overnight from Thursday until Friday morning. About 12 hours running.

I pulled it off around 7am Friday morning and got a first reading of 2.90V. From there it dropped a bit quickly, taking off 0.12V in the first hour. So I figured I would see how long it takes to get down to 2.0V and then test again to see if it saved the clock/calendar.

Well as the days have gone by, the rate seems to have slowed (maybe there is some equalization among components like backup ram and the battery immediately after taking it off 5V? Just guessing).

Saturday morning it was at 2.44V. So as of this morning, it was down to 2.30V, a total drop of 0.6V, but a 24 hour drop of just 0.14V.

And just now, about 12 hours after the Sunday morning value it is at 2.26V, only down 0.04V in half a day. Will it continue to slow the rate of drain before mostly settling somewhere for a long while?

If I can get at least a week of saved settings between uses, that will be good enough for my purposes. Using 0.04V/12-hours from this point, I would hit 2.0V around early Thursday morning, about 6 days after removing it from power. If the rate slowed even another 0.01V, that would stretch another day out of it and that's assuming 2V is even the baseline for when it will quit saving settings.

Just some personal notes, will update after a few more days.

 

[lithy]

Day 0 - 2.90V
Day 1 - 2.44V
Day 2 - 2.30V
Day 3 - 2.22V
Day 4 - 2.17V
Day 5 - 2.13V
Day 6 - 2.10V
Day 7 - 2.07V
Day 8 - 2.05V
Day 9 - 2.03V
Day 10 - 2.02V
Day 11 - 2.01V
Day 12 - 2.00V

Here is the info for my first 'leakdown' test. You can see the rate continued to slow. It is currently back in the cab and it held the Unibios and calendar settings. So I consider this issue resolved for my purposes.

But I am going to run it back to a full charge and then pull it again for one more test. That will be to see how long it takes to get down to 1.50V or 3 months max and see if it is still holding the backup RAM.

At the 0.01V/day rate I saw at the end of the first test, that would be get me through 2 full months if it holds the settings in RAM. I'll probably take daily measurements for the first week, then switch to weekly readings.

Just dumping info here in case it is useful to anyone in the future.

 

[Neo Alec]

There's a thread going about this on AP right now too. I don't know why people insist on doing this on hard mode. If you follow the instructions documented online for a non-rechargeable CR2032 you end up with something that lasts practically forever. What data are you afraid of losing on an MVS board, anyway? It will clear itself automatically when you change games enough times, or inevitably have to reset the backup ram. You can even backup high scores to a memory card with the Unibios these days.

Anyway, as far as I'm concerned, at this point this is for science. Godspeed!