Burning EPROMS

[SkiDragon]

I need to be able to burn an EPROM to restore an arcade game. What sort of equipment would I need?

[FABombjoy]

To burn an eprom, you need:
An eprom burner
An eprom(s)


...ok seriously, what size/type eprom?

The 2 popular budget burners are the Willem and the Pocket Programmer. I have the latter and it works quite well.

[SkiDragon]

All I know is that they are 24 pins. They are for a Warlords arcade game, supposedly they are for "graphics".

[FABombjoy]

Now you need to figure out the equivalent EPROM part number. For starters, what size are the rom files?

[Captain Wrong]

Interestingly, I was going to start a similar topic to this. Is there an EPROM for dummies out there somewhere? LOL

I'm looking to do various BIOS changes on boards (Neo-Geo, ST-V) and just be able to repair things if I need. What would I need to get started and is it really as simple as hooking the burner up to the PC and let 'er rip?

[FABombjoy]

I'm looking to do various BIOS changes on boards (Neo-Geo, ST-V)

If you do get a burner, and go with the Pocket Programmer II, you'll need to factor in the cost of the EP-16 adapter to do Neo Geo stuff (not sure on ST-V). It's silly expensive, and I considered getting a Willem w/16-bit adapter in addition to my Pocket Programmer. But I figured that I'd rather have less clutter on the workbench and bought the adapter instead.

[SkiDragon]

Size: 2.00 KB (2,048 bytes)
Size on disk: 4.00 KB (4,096 bytes)

[FABombjoy]

Hmm... I can't find anything that actually confirms the type of Warlords EPROMs. I don't know enough about that era of boards to safely say what EPROM type, but it's probably a (non TI) 2716, which are cheap & programmable in the Pocket Programmer II. You may want to post a question on rec.games.video.arcade.collecting.

[SkiDragon]

There are 8 rom chips in total. 2 are missing. See this image:

http://www-personal.umich.edu/~adonald/warlords/Warlords2.JPG

This shows the location of the missing chips, E6 and F/H6.

The existing chips all say on the sticker:
Atari
8113
03715X
01E

And on the chip:

AM4716DC
8111WP
AMD 1979

[SkiDragon]

Ok, I got the chips, plugged them in to the corresponding spaces, plugged everything else in, and plugged it in. All I got was a big green square that slowly appeared over the course of 30 seconds or so.

http://www-personal.umich.edu/~adonald/warlords/postchip/green.bmp

Also, the LED on the PCB lit up. What course of action is best to take now? Remember, I am missing the coin return and action buttons. There are a couple plugs inside that arent connedted to anything. I assume these would go to the coin return. Also, there is a jack on top of the power brick that is unused.

http://www-personal.umich.edu/~adonald/warlords/postchip/brick.bmp

What should I try to do now?

[SkiDragon]

I never plugged in the machine without the chips. It was apparently broken before the chips were removed, however.


Is your +5v present (under load)?

Where is this?

I will check the contacts and stuff now.

[SkiDragon]

I took all the chips out and reinserted them one at a time. I even replaced a couple of the sockets. I cleaned the pcb edges with alcohol like I would an NES cartridge. Still no luck.

[DogP]

There should be a few test points on the PCB for testing +5 and GND. Just put the multimeter on DCV and test between those points. If you can't find those test points, you can test between the corners of the chips (middle and last pin of the chip, like pins 12 and 24 of a 24 pin chip), just be careful not to short any of the pins. Anyway, you should get between 5V and 5.1V. I'd also look closely over the board for any gouges where it could have broken through the trace.

Other than that, I'd just keep an eye out for a replacement PCB, since they don't seem to be too expensive, and unless you're really familiar with troubleshooting PCBs it'll probably be pretty tough to get it fixed. Oh... from that pic, it looks like the green square is probably nothing... either a white screen (and the monitor is just adjusted badly), possibly a green screen, or it could be no signal and the monitor is just adjusted badly (a lot of times when a game isn't working, people will turn the brightness up and adjust colors, etc. trying to see something).

If you do want to dig into it deeper, I'd suggest getting a logic probe, and first thing to check would be that you have a good clock to the CPU. Other than that, print out the schematics and keep working from there.

DogP

[SkiDragon]

Yes, there is a red LED lit up on the PCB.

[SkiDragon]

All the different voltages on that board seem to check out ok.

Also 5V on the main PCB.

What else could cause my problem?

[SkiDragon]

Well, it was reccomended that I check the crystal oscillator for damage. I removed it, and although it's hard for me to tell, it looks broken. The real question is whether it was previously broken or if I broke it while removing it. Guess it doesnt matter now. Is there any way to test the crystal using an oscilloscope or anything? Where can I buy one? The only writing on the crystal says:

12.0960
ECS

Edit: I just read the manual. Apparently it is a 12.096 MHz, ±.005% Crystal. I havent had a chance to check any part distributors for this yet. To tell you the truth I am not too familiar with crystals in general. Can I reproduce the effect temporarily (for test purposes) with a function generator?

[DogP]

You should be able to find the crystal from lots of electronics stores, but yeah... if you have a function generator that can output a TTL square wave at 12.096MHz, that should work fine too.

Actually, more than likely you can stick just about any crystal slower than 12MHz and get something if that's the problem, it'll just run too slow, and possibly glitchy if there's other timing to sync with.

DogP

[SkiDragon]

I'll have to find a crystal eventually. Do these have a polarity? Probably not. Are you sure that the crystal creates a square wave? I thought they just made a sinusoid, but what do I know? If I am able to get my hands on a function generator, at what amplitude should the wave be?

[DogP]

Right... a crystal doesn't have polarity (the crystal with two legs, not an oscillator, which usually has more). It does generate a square wave, although can drive VERY little current, so it always goes into a buffer that drives the rest of the logic.

If you get a function generator, you should output TTL voltages... that'd be a square wave going from 0V to 5V (if it doesn't have TTL, you could do a 5Vpp with a 2.5V DC offset).

DogP

[googlefest1]

this is the sucker i use - its pretty cheap

http://www.ustr.net/epromer5/index.shtml

just need to make sure you can program the type of chip you want

i used this one for atari chips

[SkiDragon]

I am going to assume that a 14.318 KDS 9A crystal is too far away from what I am looking for to work.

Is there a way to test the crystal using an oscilloscope?

Can I connect multiple crystals somehow to get the right frequency?

Can I use a LC circuit of some sort to replace the crystal?

[DogP]

The 14.318 is probably too fast to use permanently, but if you popped it in for a few minutes it wouldn't damage anything. It should be close enough that the game will work, just probably play too fast/maybe be glitchy.

You can't really test the crystal easily, since the load of the scope will likely be too much for the crystal itself, but you can look at the schematics and see which chip is the buffer for the crystal and test the output of the buffer. If there's nothing, either the crystal is bad, the buffer is bad, or there's a bad connection/broken trace.

I've never heard of connecting multiple crystals to get a certain frequency, and I don't see how it would work, and I can't think of a way to make an LC circuit do the same thing.

DogP

[SkiDragon]

I am starting to doubt now whether the crystal is in fact the problem. Putting in the faster crystal only made the "green square" become larger and gave it a pattern that looked like interference. Curious, I took out the crystal and powered it up with nothing. This caused nothing to appear on the screen, i.e. no green square whatsoever.

[DogP]

Yeah, if you remove the crystal and it acts differently than with it in there, more than likely the crystal is doing it's job (usually when they fail they stop completely, not sorta fail).

DogP

[SkiDragon]

That what I was thinking, because the size of the green square seems to be proportional to the frequency of the crystal. Unfortunately, now I am back to "sqaure" one.

[SkiDragon]

Just an update: I bought a new crystal and used that just in case. The same green square appears, although I swear its brighter now, but its probably just my imagination. I know several people have suggested buying a new board or sending it in for repair, but I really want to fix it myself. If anyone has any other suggestions, please pass them along. Thanks.

[SkiDragon]

Another update. I was talking to the guy who was trying to fix it previously. Apparently he was able to get some gibberish on the screen, which is more than I've been able to do. He did mention that he saw the green square when a rom was missing. Well, I have all the chips now. It seems that some are either defective or arent making a good contact. I have been slowly swapping the 24 pin sockets with new gold plated ones. I have two left to do, and I'll see what happens then.

[SkiDragon]

Im sure nobody cares by now, but I found a dead fuse and replaced it, but nothing changed and I still dont see anything except green on the monitor.

[NoahsMyBro]

I'm still curious and interested. Keep posting your progress.

[SkiDragon]

UPDATE:
I decided to replace the two electrolytic capacitors on the main PCB with new ones. The new ones are physically smaller but have the same capacitance and tolerances. Yes, I put them in the right way. Unfortunately, this did not seem to make any difference.

Something I had neglected was the sound. There is one speaker in the machine, and I realized it was giving out a faint hum. I adjusted the volume and was able to get a louder hum out of it, and when I put my ear right up to it I thought I heard other sounds. So, I turned off the machine from the switch on the bottom, and then turned it back on. Then I definitely heard a faint sound, almost like game sound, and definitely not outside interference. I tried this several times, but it was not consistent. The pattern of sounds changed slightly, and sometimes I heard no sound. One time I got a constant repetition of beeps that continued until I shut it off, but it didnt repeat that. I also noticed some static in the monitor occasionally. I fired up the ROM image using mame but the sounds weren't the same.

This is strange and makes me think that the machine is not totally dead. Is it possible that both the monitor and sound are bad and that the game is actually doing something?