I hope what I am seeing isn't what I think it is.

Xian Xi

Xian Xi

JammaNationX,
15 Year Member
Joined
Dec 1, 2005
Posts
27,750
For the past 4 months I have been noticing an increasing rate of people complaining about the AES or MVS all of a sudden randomly resetting during gameplay for no reason. Clean cart, clean slots and boards normally working fine.

Since because of the recent "Reveal" we found out that some of the SNK custom chips were made by Fujitsu. Fujitsu TTL chips made in the 90s are known to lose internal connectivity for no reason.

I hope and pray that this is not what is happening especially with the SNK customs.

It just might be a coincidence that all these resetting systems are just surfacing at one time but I find it a little suspicious.

So far in the past year I have had 2 MV1Cs that had bad NEO-YSA2 QFPs that I had to swap out. I don't know if they were made by Fujitsu.

5 months ago I had to swap out faulty Neo-D0, Neo-E0, Neo-C1 and Neo-B0 chips.

This could mostly just be caused by the constant powering on and off of the consoles/systems. Arcade stuff was meant to be left on. These power spikes on the 5v line could be the culprit.

This of course is just speculation.
 
68k

68k

Tung Fu Rue's Prize Student
20 Year Member
Joined
Oct 3, 2006
Posts
6,780
Powering on and off wouldn't cause a system to fail. There is no "spike" in power.

Aging capacitors could be a contributor.
 
A

Alpha Skyhawk

Windjammers Wonder
Joined
Feb 28, 2012
Posts
1,383
How can we fix it in the long run? PLCs? Surely we know enough about these chips through emulation in order to be able to recreate them, but I'd imagine the main problem in creating new ones would be cost.
 
FalcomAdol

FalcomAdol

Fio's Quartermaster
Joined
Feb 24, 2007
Posts
483
Better to find an approach to prevent overspike on the 5v line at poweron? Could save some cycles.

AES should surely be designed to be powered on and off, but of course, these things were probably not designed/built with a 20 year life cycle in mind. Amazing they're still going now.
 
Rotten Zombie

Rotten Zombie

n00b
Joined
Aug 31, 2010
Posts
25
The system get hot, then get cold, then get hot again and so on. The thermal stress may be the culprit behind the problem.
Anyway,with the brutal cost of the system back in the '90s, one would think the systems were designed for long-last, with the best components they had by the time... Nope!.
 
N

NEO-GEO man

Fake account, shill troll for Christophe Ratel. Us
Joined
Aug 26, 2006
Posts
2,050
The parts used are more than sufficient to last a very long time.

There will be no voltage spike on the 5v supply cause it is supplied from a regulator. On a 5v system there may be an issue if a non regulated power supply is used, but the 5v should be from a regulated supply.

Unless the regulator becomes faulty, voltage from it should not exceed its design.
 
channelmaniac

channelmaniac

Mr Neo Fix-it
15 Year Member
Joined
Nov 29, 2005
Posts
4,275
NEO-GEO man said:
The parts used are more than sufficient to last a very long time.

There will be no voltage spike on the 5v supply cause it is supplied from a regulator. On a 5v system there may be an issue if a non regulated power supply is used, but the 5v should be from a regulated supply.

Unless the regulator becomes faulty, voltage from it should not exceed its design.
Click to expand...

Actually there is an overshoot phenomenon with voltage regulators where the voltage initially spikes above the regulated value (in this case 5v +/-5%) before the regulation kicks in. It's momentary, but it's there.
 
N

NEO-GEO man

Fake account, shill troll for Christophe Ratel. Us
Joined
Aug 26, 2006
Posts
2,050
Yes, but +/- 5% is still going to be within the component tollerance. Its sure going to be alot better than a straight up 5 volt supply unregulated, which under no load could likely be up to 7 or 8 volts with a 3amp supply. Ive seen loads of 9 volt unregulated supplies put out 13-15 volts with no load.

I cant remember for sure cause its a long time since ive been in, but the regulator i think is before the power switch ( i cant remember for sure ), so switching off the power in this case shouldnt be an issue, unless youre shutting it off at the wall to change carts, as the regulator will still be doing its job. ( again assuming its prior to the switch. )
 
Xian Xi

Xian Xi

JammaNationX,
15 Year Member
Joined
Dec 1, 2005
Posts
27,750
From what I remember the switch is actually parallel to the regulator. So it waits for the voltage reference before it adjusts of course. When you first power on some AES units especially the early ones you see it go slightly above 5.3v sometimes then drops to about 4.8v. On the newer ones I don't think the spike is that big though. But that little jolt above 5.25v can hurt some chips in the long run.
 
dragonpt

dragonpt

Over Top Auto Mechanic
10 Year Member
Joined
Oct 19, 2008
Posts
867
Xian Xi said:
For the past 4 months I have been noticing an increasing rate of people complaining about the AES or MVS all of a sudden randomly resetting during gameplay for no reason. Clean cart, clean slots and boards normally working fine.

Since because of the recent "Reveal" we found out that some of the SNK custom chips were made by Fujitsu. Fujitsu TTL chips made in the 90s are known to lose internal connectivity for no reason.

I hope and pray that this is not what is happening especially with the SNK customs.

It just might be a coincidence that all these resetting systems are just surfacing at one time but I find it a little suspicious.

So far in the past year I have had 2 MV1Cs that had bad NEO-YSA2 QFPs that I had to swap out. I don't know if they were made by Fujitsu.

5 months ago I had to swap out faulty Neo-D0, Neo-E0, Neo-C1 and Neo-B0 chips.

This could mostly just be caused by the constant powering on and off of the consoles/systems. Arcade stuff was meant to be left on. These power spikes on the 5v line could be the culprit.

This of course is just speculation.
Click to expand...

Your probably right, it can't be an coincidence that some of my Megadrives started break down at the same time + - ...
Same for both my GameGear's , i was forced to recap all the capacitors, and BAM problem solved for the Megadrives and Gamegears.

Time is an factor on all electronics components, sadly but true
 
ChuChu Flamingo

ChuChu Flamingo

We have purposely, trained him wrong, ...as a joke
10 Year Member
Joined
Nov 23, 2010
Posts
2,786
Could be a lot of things contributing to the death of these ICs. On the arcade side, I know a lot of people don't really check their 5V voltage to make sure it is at 5.0-5.10. On the console side, there are still lots of people not using quality surge protectors.

One must also consider that not all ICs are created equal. Over time, there will be ones that last considerably longer than their brethren. Whether that is due to that batch being better made or just luck we will never know.

I hope it isn't what you say James, only time will tell.
 
N

NEO-GEO man

Fake account, shill troll for Christophe Ratel. Us
Joined
Aug 26, 2006
Posts
2,050
Surge protectors will make no difference to power supplied after a regulator.
 
ChuChu Flamingo

ChuChu Flamingo

We have purposely, trained him wrong, ...as a joke
10 Year Member
Joined
Nov 23, 2010
Posts
2,786
NEO-GEO man said:
Surge protectors will make no difference to power supplied after a regulator.
Click to expand...

But they can before which will consequently affect after and do some damage.

I've seen a lot of computers get shit blown out on the pcb motherboard due to this whether it be overvoltage or undervoltage (which you need a UPS for).

What type of voltage regulator is the neo geo? I know the early models used it in the power supply, but not the latter ones. Isn't it just like a standard ATX PSU?
 
Last edited:
N

NEO-GEO man

Fake account, shill troll for Christophe Ratel. Us
Joined
Aug 26, 2006
Posts
2,050
No, its not like an ATX power supply.

The regulator inside the console operates within a given nominal range, and outputs a constant voltage regardless of input voltage, provided it stays within the nominal range. The input range goes quite high.

I cant remember the regulator model in the later home consoles, but i think its an LM2576
 
Last edited:
dragonpt

dragonpt

Over Top Auto Mechanic
10 Year Member
Joined
Oct 19, 2008
Posts
867
NEO-GEO man said:
No, its not like an ATX power supply.

The regulator inside the console operates within a given nominal range, and outputs a constant voltage regardless of input voltage, provided it stays within the nominal range. The input range goes quite high.

I cant remember the regulator model in the later home consoles, but i think its an LM2576
Click to expand...
wow i didn't know that ...

I always thought the internal regulators work in the same way like PC power supplys.

Thanks for the info, i'm always learning .
 
N

NEO-GEO man

Fake account, shill troll for Christophe Ratel. Us
Joined
Aug 26, 2006
Posts
2,050
Pc power supply operates from mains voltage down to 3 different voltages.
 
distropia

distropia

SouthTown StreetSweeper
10 Year Member
Joined
Apr 28, 2010
Posts
1,075
What about quality PSU's that holds the power until everything is regulated and then power on? I use MeanWell ones that works that way, or at least they say that.

Besides that, certainly we should be thinking on how to find a replacement for those custom ICs.
There's people out there that have reproduced 8-bit era customs, like the Sinclair Spectrum's ULA (graphic and I/O chip) in a PLD. But a very basic chip like that could be barely compared to a YSA, not a highly complex GRZ.

Its a shame that we can not count on SNK-P for a "reprint".

@neogeoman & dragonpt: pc power supplies give +5, -5, +12, -12, and +3.3. An entire psu is not just a regulator.
 
Last edited:
distropia

distropia

SouthTown StreetSweeper
10 Year Member
Joined
Apr 28, 2010
Posts
1,075
I'm afraid I have to confirm that custom chips are increasingly dying. Some years ago most repairs were broken traces, bad ram or easy stuff. Now I can't find enough Neo-C1 for repairing customer's boards.
This is real drama, not that childish discussions on the news.
 
You must log in or register to reply here.
Top