(seriously) Sick AES- *updated*

I just picked up a broken AES. When I got it, it had a failed Debug BIOS install (piggyback method). One of the Debugs legs was snapped off right at the plastic. I figured, oh man! Easy!

Nope.

I snipped the useless Debug IC off of original BIOS, cleaned up the solder and tested. I get a consistent garbled screen. Same garbles with a cart in, as with no cart in. It also, makes a steady( every 3 seconds) quiet 'click' noise audible through the TV speakers, coinciding with a screen flash.

I figure, well, lets pop out the old BIOS, pop in a socket, and pop in my old AES BIOS from my AES (I pulled mine out of my AES, and socketed in the Debug when I modded mine), as it's was working when I popped it out.

Soldered old BIOS out, soldered in a socket, and popped the old working AES BIOS in. Same Garbled screen, same 3 second cycle flash and click, no change at all. It's like putting another BIOS was not the problem at all.



Got out the meter. I verified continuity from each BIOS IC leg to the underside socket pin. 100% continuity. Then I started testing continuity from each socket pin, to it's trace destination on the PCB. I verified all of the ones I could, by following the traces (about 37 of them were followable), and they all had continuity, so it seems no broken traces.

That's just great. Not. No solder spatters. No obvious blown capacitors, broken or missing components. The worst looking part is gone, once I took out the messily soldered debug BIOS.

Other things to note- the unit has not had a video or stereo mod attempted to it, as all of those IC's and board areas are clean as a whistle. It also came with a third party (Ratshack) AC adapter. It was set properly, and seems fine, but it is a variable voltage AC. For all I know, it may have once been plugged in set too high, or polarity reversed. What does that do to a Neo? Are these symptoms similar?

So I'm thinking something else was damaged, perhaps when the Debug mod was attempted, something got shorted, and blew an IC?

If anyone has anything to offer, such as experience with AES's that had an improper AC used with them, or bad Debug BIOS mods, I'm all ears.

I'm not ready to call this AES 'parts' just yet.

dude did you buy that AES from lithy? cause i had attempted a debug bios mod on an AES and he offered to buy it when i couldn't get it working. it could be the old boy.

I had one like that before, it died during the mod for no reason whatsoever.. I wasn't able to find anything, checked all you did too, no go. I've done loads of mods on NGs and other machines, it was definately not lack of skill or tools.

I have a mvs one slot which does the same as your aes clicking sound and a flashing screen, is your screen flashing green blocks?

Mostly green blocks, but there is a jumble.

Yes, I got it from lithy. I figured the original modder might stick his head up.

Worst case scnenario, she is parts. However, I'm not giving up yet.

is that is so... please keep my name down for a AES shell as mine is cracked

i'm interested in a few parts, if you should decide to rip it apart.
might also buy the whole thing, if the shell is not in a too bad condition.

The shell is not cracked, but it is totally logo faded. Totally. Someone did a nice gold silkscreen of "NEO-GEO" right where it's supposed to be, but that's it. Everything else is gone logo wise.

The vultures are circling! Shoo- Shoo! It's not dead! I need help, not vultures!

I wonder if it could be the slot. It looks fine, and it's a 60000 series serial.

From what you mention regarding; same Garbled screen, same 3 second cycle flash and click, its most likely due to the bios not installed properly.

Did you trace only the line from the bottom where you solder the legs?
Trace the line from the top too, there are connections from the top too.

Mostly likely the line or the rings are broken on the top if theres no problems at the bottom.
The lines and rings are not visible due to the socket/bios blocking it.
If possible, remove everything including the socket and check both top and bottom rings/ trace lines before you solder the socket in.

Check them out.

Hopes it helps,

Will do. I did not check the top ones. I'll re-do the continuity tests with nothing there, and see what I get.

More desoldering. sigh.

I've repaired a lot of botched mods and most of the time it's a busted trace (or a handful of them). On some occasions, something not even related to the bios gets messed up just because people are not careful.

The behaviour you're having is from the system not booting. If it was in the gfx or sound, it would at least boot, so there is something wrong around the bios/processor area - focus your efforts here. You may have checked all the pins to the processor, but you may need to check the traces that go through the bios chip area that are not connected to the bios. There are a couple traces between pin 1 and pin 2 that are like this and I've had a couple people bust these ones.

Any chance you could send me a scan or picture of the board? You can't always go by serail number, but if you can send me something to look at, I'll let you know the proper voltage supply for the board.

I also offer a free evaluation if you decide to give up. I have a pretty good success rate, so chances are I can solve it for you. I also keep the costs to a very minimum as I don't like to see a Neo die because of a couple $$... If I can't fix it, all I ask is that you cover return shipping. You can send just the board - no cables, accessories, or shell needed.

Jeff
jmkurtz@columbus.rr.com

I have a cpuple of things on the go right now, with repairs for others right now (Ps2's and Xboxes). I'll try to take a pic, and get it up soon.

Thanks for the helpful tips all. If it's a trace from the BIOS, I'm in business I bet.

Time to go sloooow.

omnedon said:

The shell is not cracked, but it is totally logo faded. Totally. Someone did a nice gold silkscreen of "NEO-GEO" right where it's supposed to be, but that's it. Everything else is gone logo wise.

The vultures are circling! Shoo- Shoo! It's not dead! I need help, not vultures!

I wonder if it could be the slot. It looks fine, and it's a 60000 series serial.

Click to expand...

*Circules around omnedon's AES*

try dabbing some more solder on the bios socket pins...

I once had a socketed AES that would boot both the original and debug bioses fine.

But when i put the uni 1.3 in, i get the green screen of death.... I could interchange any of the other two bioses and it would work fine. but pop in the 1.3 and get the green screen of death.....

Finally i added some solder to all of the pins on the bottom of the socket... low and behold:

Do dooo dooo do na na na na...... I get the neo boot screen on my tv with the 1.3 in it....

Pics! I took 'em in quasi hi-res, so I'll just link 'em up.

http://members.arstechnica.com/subscriptors/x/omnedon/Neo board 002.jpg

http://members.arstechnica.com/subscriptors/x/omnedon/Neo board 006.jpg

http://members.arstechnica.com/subscriptors/x/omnedon/Neo board 007.jpg

http://members.arstechnica.com/subscriptors/x/omnedon/Neo board 008.jpg

The iron burn near the mem card slot is NOT my doing. Looking at the underside shot, itlooks like visible holes in my solder job, but those are shadows. I've looked at it under my magnifyer light, ans the solders seem solid. maybe not the prettiest, but solid. I hope they suck out nice and easy....

Looks like a decent board no? I probably won't be able to get time to desolder the socket for a day or two.

Some of these traces (especially the underside ones) were HARD to follow. They go a long long way in many cases, with traces being one of a great many in line with eachother. Does anyone have a schematic that could make my following of the traces more accurate (and easier)? I'm no wiz with schematics, so if someone has some to offer, a tip or two as to where to look on it would be helpful as well.

I don't ask for much do I?

omnedon said:

The shell is not cracked, but it is totally logo faded. Totally. Someone did a nice gold silkscreen of "NEO-GEO" right where it's supposed to be, but that's it. Everything else is gone logo wise.

The vultures are circling! Shoo- Shoo! It's not dead! I need help, not vultures!

I wonder if it could be the slot. It looks fine, and it's a 60000 series serial.

Click to expand...

holy shit, that is the AES i tried to mod, gave up on and sold to lithy, though when i had it the second pin on the debug bios was still attached.

and the melted memory card slot...dude that is my old AES!

Make sure you are using a 5V supply with that board.

Looking at the bottom left side of the board you will see that the LM2576 is missing, which is used on 9V boards. On 5V boards, this spot will be empty (like on yours) and you will have the power transistor (just left of the reset button -- green thing sticking up).

Jeff

omnedon said:

Pics! I took 'em in quasi hi-res, so I'll just link 'em up.

http://members.arstechnica.com/subscriptors/x/omnedon/Neo board 002.jpg

http://members.arstechnica.com/subscriptors/x/omnedon/Neo board 006.jpg

http://members.arstechnica.com/subscriptors/x/omnedon/Neo board 007.jpg

http://members.arstechnica.com/subscriptors/x/omnedon/Neo board 008.jpg

The iron burn near the mem card slot is NOT my doing. Looking at the underside shot, itlooks like visible holes in my solder job, but those are shadows. I've looked at it under my magnifyer light, ans the solders seem solid. maybe not the prettiest, but solid. I hope they suck out nice and easy....

Looks like a decent board no? I probably won't be able to get time to desolder the socket for a day or two.

Some of these traces (especially the underside ones) were HARD to follow. They go a long long way in many cases, with traces being one of a great many in line with eachother. Does anyone have a schematic that could make my following of the traces more accurate (and easier)? I'm no wiz with schematics, so if someone has some to offer, a tip or two as to where to look on it would be helpful as well.

I don't ask for much do I?

Click to expand...

JMKurtz said:

Make sure you are using a 5V supply with that board.

Looking at the bottom left side of the board you will see that the LM2576 is missing, which is used on 9V boards. On 5V boards, this spot will be empty (like on yours) and you will have the power transistor (just left of the reset button -- green thing sticking up).

Jeff

Click to expand...

Damn this man is good! I wonder if he ever did a mod before.



(This is for the people that this joke passed up...ITS A JOKE)

Desoldered off the socket. Started continuity tests along the top of the board.

The bottom row of pins was 100%.

The top row... oh my. Right off the bat a visible lifted trace that runs under the IC. Plus a junction of a few more, on the first from the right pins, an entire swatch of them are burn damaged, from the initial mod attempt it seems to me.

The traces are soo goddamned long! It's so easy to lose track when following! Any tips? I'm taking my time with this. I have stuff to do for other people that can't wait, and I think doing this right is going to take time. I think this initial swath is the worst part.

Once I'm certian where traces are supposed to lead, I'll repair with mostly bridging, and a couple look like they should take with a painted line of defroster repair solution.

I'll keep yall posted.

I spent an hour and a half today, painstakingly following traces. I have 3 confirmed broken traces, all topside. I went through all 40 pins TWICE, making tiny marks with a sharpie once I confirmed a trace as functional.

I need to go get a new socket tomorrow. I think these broken traces are great news. I have hope now.

Well, it's still not fixed. No change.

Here is what I did- I bridged 4 points that were CONFIRMED not contacting with their destination points. The bridges are visible in the pics. All of the other 36 points were confirmed and marked twice each, top and bottom.

Then I bridged the broken ones when I installed the brand new socket.

http://members.arstechnica.com/subscriptors/x/omnedon/Neo board bridged 001.jpg

http://members.arstechnica.com/subscriptors/x/omnedon/Neo board bridged 002.jpg

http://members.arstechnica.com/subscriptors/x/omnedon/Neo board bridged 003.jpg

http://members.arstechnica.com/subscriptors/x/omnedon/Neo board bridged 005.jpg

The third pic is the end screen result, with no change in regards to what I described in my earlier post.

I'm done for today anyway.

I'm wondering if the repeated solders have damaged the board, and will do trace connection tests with the pins I can follow and test, leaving the socket in. Pic 2 really shows what the heat has done to the board, but in person, with a mag glass, it really seems OK to me. A couple of random trace tests showed everything fine, including the bridged ones.

It is a possibility that the BIOS chip is bad. It's from my AES (an earlier version), and it came out reasonably easily. I could take my AES apart and try some swaps, but I really do think it's a longshot. I found a local place that can burn me eeproms, so I may get a uni bios or two done up. One for my AES anyway. Then I can eliminate that remote possibility.

*sigh*

It's broke.

A happy ending!

In the end, I took JMKurtz up on his offer. As it all turned out, it was a matter of broken traces. I missed one or two, and I misrouted a couple of the ones that I did locate. Compared to my personal 'early model' AES, the BIOS traces go a long ways in some cases, and I made mistakes.

JMKurtz found the errors, fixed them, and fixed some other questionable traces as well. I received it today, I popped a uni 1.3 into the socket, and it's golden. I'm going to S-vid and A/V mod this unit, and sell it to a buddy.

A public Thank You to JMKurtz, and props to his experience with the Neo, and his willingness to use it to benefit the community, at a very reaonable price.

Another Neo saved!