EPROM dumping trouble

[Dziadek_Mroz]

I'm trying (with no luck) to read contents of eprom from Keithley multimeter (Edit: Meter is Keithley model 175). Ic is M5L2764K by Mitsubishi (8K x 8 ), and its datasheet states that its interchangeable with Intel 2764 (which I checked and seems to be 100% true). Using TL866 series programmer, I set it in software to Intel chip, cause there's no mitsubishi chip of that kind in library, hit read and nothing. FF from top to bottom. I tried other 2764 chips from library, but all of them behave exactly the same. I thought that maybe IC is bad, but meter works fine. Programmer should be ok too, since there was no problems with reading contents of second eprom (from GPIB board). Google doesn't want to help in any way and I ran out of ideas what might be problem here, so any help would be appreciated.

[amyk]

Is there a read speed option? Try a slower setting. A scope or LA on the pins will also help.

[Shock]

Check what software/firmware you are running (the latest for the original TL866 is 6.85 with firmware 3.2.86) and if you have a genuine programmer. Remove any excess solder if there is any and give the pins a good clean. You can also try toggling the "check id" as well. There is a large thread where most TL866 problems are discussed in, your post might reach a wider audience there.

http://www.autoelectric.cn/EN/note.html
https://www.eevblog.com/forum/blog/eevblog-411-minipro-tl866-universal-programmer-review/

[Dziadek_Mroz]

Checked xgecu website and I believe that what I have is genuine product (with newest software and firmware). Eprom in question is socketed on the board (forgot to mention, meter is Keithley model 175), so there's no solder on pins. I tried cleaning all pins with IPA, but nothing changed. I also tried changing USB cable or port, thinking that maybe IC draws too much current, but that's also dead end. ID chceck has to be "off" for programmer to at least try doing something, because ID always reads as FF FF, which I guess is somehow related to the main problem.

There is a large thread where most TL866 problems are discussed in, your post might reach a wider audience there.

Thanks, didn't know that, I'll check it out

Is there a read speed option? Try a slower setting. A scope or LA on the pins will also help.

Couldn't find anything for read process, only for write. But I'm going to look for alternate software that might have more options. Scope checking has to wait a while for day when I have more time, and I highly doubt that HP1630D will be fast enough to logic probe read process.

[james_s]

Does your device read other EPROMs successfully? Are you sure you have it inserted correctly in the socket? Those chips should be 100% compatible for reading.

[Dziadek_Mroz]

I'm sure that I inserted it correctly. After first attempt I was horrified by FFs, thinking that I fried the IC somehow, but reinstalling it into DMM showed that it works just fine. And there was no problems with reading contents of 2732 type eprom at GPIB board. I tried some eproms found in drawer (24 to 40 pins) and most of them reads fine, others I suppose are just blank after few years without sticker.

Those chips should be 100% compatible for reading.

Well, I'm aware of that, which makes it even harder for me to understand what's wrong. Sadly, I don't have any other 2764s in stock to check different chip for same behaviour. Guess I'll order some and see what happens.

[m k]

Was there a content in any of your 28pin chips?
If not then check that pin 14 actually is GND.

[james_s]

Are you sure the *entire* chip is FF's? FF is the default for a blank EPROM and I have seen ROMs that are mostly empty space used in equipment.

[amyk]

Couldn't find anything for read process, only for write. But I'm going to look for alternate software that might have more options. Scope checking has to wait a while for day when I have more time, and I highly doubt that HP1630D will be fast enough to logic probe read process.

That's a 100MHz LA. Definitely more than fast enough for this purpose.

[Dziadek_Mroz]

Was there a content in any of your 28pin chips?
If not then check that pin 14 actually is GND.

Most of other chips of this size read fine, some read as blank, but as I said, I would blame time for that.

Are you sure the *entire* chip is FF's? FF is the default for a blank EPROM and I have seen ROMs that are mostly empty space used in equipment.

Yes, scrolled through whole "contents".

That's a 100MHz LA. Definitely more than fast enough for this purpose.

I thought it's 25MHz. Maybe looked up wrong table in manual. First I'll test this with scope, next I'll try this one then.

[Dziadek_Mroz]

It's possible that I'm doing something wrong, because for one time it managed to read that eprom (contents saved, safe in archive ) but then again nothing upon next try. Maybe bad ZIF? As long as meter works, I'm not going to fight with that one anymore. Thanks for help everybody

[james_s]

I thought it's 25MHz. Maybe looked up wrong table in manual. First I'll test this with scope, next I'll try this one then.

Even that would probably be more than enough. Most stuff that uses these old EPROMs runs at a few MHz, maybe 10MHz tops.

[m k]

Manual says that M5L2764K access time is 250ns, 200 for M5L2764K-2.