Home Forums General Discussion Technical Questions EPROM chip copying and reading

I don't know much in this area and want to know if you can use a universal chip burner/copier to look at a EPROM from a drive and see if it's corrupted. If it's not, can it be copied and burnt onto a new chip. Ive come across a couple drives now where it says "EPROM Fail" or something along those lines, and it'll shut down. But if you go through and reset it will run for a few weeks before shutting down on a same or similar fault. My though is that the EPROM is old and some part of the code is flickering or not being read correctly intermittently so it shuts down. So if you can run a check on a chip to see if it's reading properly and copy it if it is then maybe a new chip won't have that issue...
Once again I know almost 0 on this topic and I'm just pissing in the wind with that theory.
Also what set up are people using for chip readers?

Thanks

From my software engineering experience long back, many times you will find that an EPROM is socketed. Sometimes there is oxidation on one or more legs of the EPROM or socket. Sometimes I'd take the EPROM out and shiny up the legs with a pencil eraser or ScothBrite pad and then put it back in. Sometimes that fixed the problem.

There are EPROM readers and burners. Sometimes an EPROM will have an anti-duplicate feature, but I doubt that will be the case here.

You can get an EPROM burner from Amazon for less than $200. It probably won't be very friendly to use, but it will get the job done. If you need an EPROM, I'd go to a quality supplier like DigiKey.

Thanks, would the software for the reader/burner be able to tell if the chip is corrupted?

Maybe and maybe not.

If the chip is electrically damaged, the reader detect that and tell you.

If the data inside is corrupted, such that a "1" is now a "0" or vice versa, it will not be be able to detect that. Typically, the data inside the EPROM will have a checksum that was generated by the development software that is also stored along with the data inside the EPROM. This checksum is then recalculated when the computer starts up by reading the entire EPROM and compares it to the stored valve also in the EPROM. If they don't match, it will give an error. The location of the stored checksum is not necessarily the same, but will typically be at the end of the EPROM or the end of the data stored inside the EPROM.

There are many types of checksums, but the most common, especially for this period, would be just adding all the bytes together and limiting the sum total to the least significant 4 bytes of all the bytes added up in that are in the EPROM.

So if it has a checksum stored inside and a 1 goes to 0 then wouldn't the checksum be off?
Sorry this is fairly new to me and I'm just trying to wrap my head around it

Yes, exactly. If one of the bits inside the EPROM ends up changing, the calculated checksum won't match the stored checksum.

Now, there are finer points that can be dependent on the checksum algorithm that is used. A machine from the 80's or older probably just used a simple sum of bytes checksum. With that simple algorithm, two bit flips could cancel each other out and the simple checksum would not pick it up.

Ahhh that makes sense thank you. These drives are probably late 90's early 2000. Magnetic dsd412

Back to the OP's original post, my first question is to ask what the ACTUAL error message is. Is it really "EPROM Fail"? If you can give the exact details, the members of this forum may be able to provide better clues about what to do.

The latest DSD 412 Manual from the Columbus McKinnon website is

https://www.cmco.com/globalassets/inriver/resources/cs00407-rv12.pdf

It does not mention an error about the EPROM (refer to table on manual page 82 which is PDF file page 84)

There is a DSD 412 Application Note "DSD 412-118"

https://www.cmco.com/globalassets/inriver/resources/dsd412_app118_nvram.pdf

It mentions that the NV-RAM device stores a copy of the EPROM checksum, and a failing NV-RAM battery can result in it being incorrect, which then causes the NV-RAM to be initialized. That doesn't seem to align with your "go through and reset it will run for a few weeks ..." approach though.

NOTE: The full list of documents for the DSD 412 can be found at

https://www.cmco.com/en-us/products...levator-drive/#documents-pdp-all-columns


Regarding EPROMs, you may find that it is hard to get an exact replacement EPROM part for an older drive like the DSD 412. Many technology changes have happened over the years since the DSD 412 was released - everything from memory access timing (now faster) to packaging (now surface mount instead of DIP chips) to memory capacity (it's hard to find "small" EPROMs). In some cases the only alternative may be to use a new part with an adapter PCB. Or maybe Columbus McKinnon can sell you a "genuine spare part" or offer some technical support.

Web sites like Jameco.com seem to have better stock of old DIP EPROMs than others like Digikey.com do. I also noticed that Mouser.com has some EPROM parts packaged in all-plastic packages without a clear "window" for erasure. This effectively makes them One-Time Programmable Read Only Memory (OTPROM) parts rather than Erasable Programmable Read Only Memory (EPROM). You can use these OTPROMs, but be aware that if you program them with corrupted data extracted from your current EPROM, you can't erase an OTPROM part and will have to try another one.

Good luck in your quest! Please let all of us know what you eventually discover to be wrong.

I had a look at some of those documents. In this one...
https://www.cmco.com/globalassets/inriver/resources/cs00407-rv12.pdf

...on PDF page 98, there is a circuit board layout diagram. A callout for U50 is there on the right side of the print and it points to U56 on the circuit board (not sure why the discrepancy). Anyway, U56 is the NV-RAM and the docs say to repalce every 8 years and it is rated for 10 years.

That sound very much like a device that was popular in that time frame from a company called Dallas Semiconductor. They had NV-RAMs that had embedded, non-replaceable, lithium batteries.

Depending on which one it is, there may be a replacement. I've also seen YouTube videos of folks hacking them apart and putting an external battery on it.

This could be your problem as well.

Would be good to know exactly the sequence of events and the error you are getting.

So I looked the faults I had up on one of those documents, and they're slightly different info then what I had.
I had several faults (only one per shut down) that fell under the misc faults.


Faults :13-17, 21-26,
100-104, 112-114,
220-234, 240-254,
926

MISC FAULTS - Problem with either hardware or software on the Drive Control
PCB. See list at Section 4.5.
Check and/or replace U13, U14, U39, U40 (EPROM’s).
Replace Drive Control PCB A1.


Also this is a fairly new job for us and it seems like they've had drive issues in the past because there's 2 boxes with drive boards in the machine room one looks like a newer version and another looks old. Neither have chips on them and they were in a pile of other boards that some were marked "NG" and others "GOOD". But it didn't surprise me that the NG boards were left behind because the machine room was so filthy the former service guy had to put shoe covers on to walk in so when he left he didn't track grease and carbon dust all over the building