• Please review our updated Terms and Rules here

The curse of fake parts out of China.

shirsch said:
I'm doubting what you say, but I cannot find any information on how to read that code. I have 3 or 4 different device programmers and none of them offer that feature. Internet search turns up nothing for UV EPROM.
Click to expand...
What are you doubting that I am saying exactly? So you dont own a programmer that has a Chip ID read function? Again I have a minipro TL866A. IT can read eprom ID's no problem. What do you have?

see photochiid.jpgIf I go to the top menu "device" then click on "READ ID" I get the ID. This is a cheap 27c64 eprom I got from china with an ST label, but that ID comes up with the info below so its been relabled. They all seem to be relabeled ST brand for some reason.
9B0827C64AEurotechnique27C64
 
The worrying thing for the batch of bad eproms I got, the programmer thought they were ok and the devices verified and their byte file looked normal, in the programmer when the 21v programming voltage was applied as per the part number. They failed to operate normally in the actual circuit where they were deployed as a character generator on a video card. But who knows, in another application they may have worked. And the borderline condition they suffered from cleared with the higher programming voltage. It could be that some of these sorts of parts we think may be good, are not, but we don't find out unless the application tasks them in some way, such as higher frequency operation than in the programmer, that they cannot support.
 
VERAULT said:
What are you doubting that I am saying exactly? So you dont own a programmer that has a Chip ID read function? Again I have a minipro TL866A. IT can read eprom ID's no problem. What do you have?

see photoView attachment 1269188If I go to the top menu "device" then click on "READ ID" I get the ID. This is a cheap 27c64 eprom I got from china with an ST label, but that ID comes up with the info below so its been relabled. They all seem to be relabeled ST brand for some reason.
9B0827C64AEurotechnique27C64
Click to expand...
I meant to say I'm not doubting what you say. And, I believe your programmer has that feature as you show. I just cannot find anything that explains the low level bit twiddling that makes this happen.
 
shirsch said:
I meant to say I'm not doubting what you say. And, I believe your programmer has that feature as you show. I just cannot find anything that explains the low level bit twiddling that makes this happen.
Click to expand...
Oh.. Gotcha! Cant help you with that.
 
shirsch said:
I'm not doubting what you say, but I cannot find any information on how to read that code. I have 3 or 4 different device programmers and none of them offer that feature. Internet search turns up nothing for UV EPROM.
Click to expand...
See for example this National Semiconductor app note - the second page describes the process required to read the manufacturer code from EPROMs that support this feature:


There is of course a "trick" to reading this data (requiring application of 12V to the A9 pin) so this information is hidden during normal operation of the EPROM.
 
Learn something new every day. I wonder why my device programmers don't mention or support this feature? One of them (Andromeda Research) is quite well supported by the vendor. I'll drop them a note and see if this feature is there by some other name - or perhaps on a hidden menu.
 
VERAULT said:
If I go to the top menu "device" then click on "READ ID" I get the ID. This is a cheap 27c64 eprom I got from china with an ST label, but that ID comes up with the info below so its been relabled. They all seem to be relabeled ST brand for some reason.
9B0827C64AEurotechnique27C64
Click to expand...
I recently bought a couple of SST27SF256 from China, to verify operation (with adapter) on an IBM 5150 motherboard.
My EPROM programmer automatically alerted me to the fact that the device ID read from the supplied SST27SF256's was not as expected.
I then ran the 'ID' option of my programmer, and that informed me that the device ID read was for Windbond W27E257.

( I added the discrepancy to my list at [here]. )
 
Apparently there are some caveats involved. I asked John Dumont at Andromeda and he told me:

We do not support Silicon Signature. The reason is that it applies a high voltage to an address line to read the
data. This damages any part which does not support the function.
Click to expand...
So it sounds like you need to be careful with older parts.
 
shirsch said:
I'm curious about the EPROM id code you mention. I know that EEPROMs and other contemporary devices have this capability but I've never seen any information to suggest that legacy UV-eraseable EPROMs contain such a thing.
Click to expand...
Device id isn't likely to work on anything earlier than late issue 2764 devices. From memory reading the id involves feeding "out of range" voltages to the A9 line amongst other things. "Out of range" meaning something outside the normal 0 to +5V. A device datasheet will give you the details.
 
I think I might have figured out what is wrong with the 2732A ROMs .

My theory now is they have been over-erased.

I know already that they were used ROMs that were cleaned up and had new pins attached to them and the pins re-Tin plated. They break off with a small amount of force. They would have been erased too, probably in a large high power UV chamber to do a lot of IC's at once. A few of them also have a damaged "cell" here and there and won't program.

The higher than standard required program voltage suggests to me that there has been some global shift in the charge distribution in the die.

I was reading that it is possible to damage a UVeprom by over erasure, but the nature of the damage was not described. But since the difference between a 0 and 1 represents a difference in the amount of charge in a particular physical location with respect to another, and the process of erasure and programming represents having the charges in two different locations and it is now requiring a higher programming voltage, I think somehow they have been over erased. The behavior in the character generator, looking at the characters with lines of pixels coming and going, suggests that a number of locations in the ROM are acting as though they are on the border between a 0 and a 1 when the programming voltage was 21v, but not 25V.

Does anyone think this makes sense ? And does anyone know if the over erasure problem is permanent or reversible ?

I'm wondering now if this "damage" is reversible, or not. In that if I over programmed them a few times with bytes all 00 at a 25V voltage and a brief as possible erasure to get all FF's, if I could shift the charge distribution back the other way and get them to program and work properly with the 21V programming voltage that they were designed for. Could be an interesting experiment to help verify the theory.
 
Last edited:
Altered parts are just bad, period.

I purchased some NEC V20 chips from a supplier that had good feedback and the pictures showed original, correct markings on the chips. And of course what I got was different. The top of the chips had been sanded with something; you could see the grooves with an eye loop. New markings were applied that looked like they were done with a laser, not ink.

The parts ran in a 5Mhz machine but I ordered 10Mhz parts. With no way to tell if these were actually 10Mhz parts before they were cleaned up, I rejected them and filed a claim on eBay. It was pretty simple - the pictures didn't match what was shipped, and the sanding grooves did not belong there.

And of course some clown on Reddit tried to convince me that everything was fine with the chips. Sure, whatever - you can use them in your own project buddy ...

Link: Fake/Altered NEC V20
 
shirsch said:
So it sounds like you need to be careful with older parts.
Click to expand...
My programmer knows what devices support 'device ID'. And so I expect that after I inform my programmer as to what chip I am going to insert, it knows whether or not it can interrogate the chip for device ID.

( The device list for my programmer, which is a text file, is at [here]. Presumably, ID="XXXX" means either 'not known by programmer' or 'not supported at all'. )
 
Hugo Holden said:
I think I might have figured out what is wrong with the 2732A ROMs .

My theory now is they have been over-erased.

I know already that they were used ROMs that were cleaned up and had new pins attached to them and the pins re-Tin plated. They break off with a small amount of force. They would have been erased too, probably in a large high power UV chamber to do a lot of IC's at once. A few of them also have a damaged "cell" here and there and won't program.

The higher than standard required program voltage suggests to me that there has been some global shift in the charge distribution in the die.

I was reading that it is possible to damage a UVeprom by over erasure, but the nature of the damage was not described. But since the difference between a 0 and 1 represents a difference in the amount of charge in a particular physical location with respect to another, and the process of erasure and programming represents having the charges in two different locations and it is now requiring a higher programming voltage, I think somehow they have been over erased. The behavior in the character generator, looking at the characters with lines of pixels coming and going, suggests that a number of locations in the ROM are acting as though they are on the border between a 0 and a 1 when the programming voltage was 21v, but not 25V.

Does anyone think this makes sense ? And does anyone know if the over erasure problem is permanent or reversible ?

I'm wondering now if this "damage" is reversible, or not. In that if I over programmed them a few times with bytes all 00 at a 25V voltage and a brief as possible erasure to get all FF's, if I could shift the charge distribution back the other way and get them to program and work properly with the 21V programming voltage that they were designed for. Could be an interesting experiment to help verify the theory.
Click to expand...
That is an interesting thought. I imagine UV EPROMs do have finite number of write/erase cycles, just as flash and EEPROMs do. I'm more interested in the thought of overdriving the erasures or writes to shorten or lengthen their lifespan or number of cycles they can go through. I will definitely do some research on this.
 
Interesting video about identifying fake chips:
, may be relevant for this thread
 
I mean I guess I should have known when I ordered "Original 10PCS W27C512-45Z W27C512 IC DIP EEPROM 512KBIT 45NS Winbond EEPROMs" from eBay, originating in China. 10 arrived, each bagged in plain old pastic bags (definitely no ESD safety there), half were on sort-of pink/red foam(think very thin camping / yoga mat), the other half white.... slight differences in appearance between the two also. I did give the benefit of doubt as maybe sourcing from multiple batches etc. I didn't test more than one of each, but I'm pretty sure the pink/red foam read as expected in the TL866(empty). The other didn't read. Kind of gave up on that batch, mainly due to lack of confidence in the safety of the packaging, haven't yet ordered any more. As a side note I've had better luck sourcing old logic IC's from Europe and HK, they at least turn up in tubes and usually ESD safe bags, but the cost is usually much higher. I tend to avoid Rockwell labels on anything.
 
You must log in or register to reply here.
Back
Top