Reading a soldered in EPROM

[R-1125F]

I am working on a project where I need to read a TC57256AD-20 UV erasable EPROM. This is on a borrowed board so I cannot take any chances destroying it. The chip is soldered in, and I cannot remove it because it is a borrowed board.  My EPROM programmer is a T48 (TL866 3G).  The closest match they have listed is a TC57256D.  From what I have been able to find on the internet, the code (9845) for the TC57256AD-20 and TC57256D is the same.  The programming voltage is different, but I only have to read this chip.  Since I cannot take any chances with this chip, is there any danger to the contents by reading the TC57256AD-20 as a TC57256D in this programmer?

[calibrationfixture]

Hi,

The Circuit will load the Reader/Programmer. A nice Article (and Products):

https://www.arlabs.com/incircuit.html

Succes,

Calibrationfixture

[pcprogrammer]

Not only that, it will most likely also power the circuit and the eprom wlll be in its normal use.

A bit odd that an UV erasable eprom is soldered in though.

[fenugrec]

You might not be able to read it in-circuit if some lines (i.e. OE / RD) are never Hi-Z even if you manage to hold everything else in reset.

I've been in a similar situation a few times and another (tedious) solution is to use a logic analyzer with sufficient channels to cover at least all the address bits , OE+RD lines, and as many data bits as possible (it's always possible to do multiple captures and reconstruct the actual bytes). If you're lucky, the firmware does a full ROM checksum at some point during boot, where you are guaranteed to see every location read from at least once.

On another board where it was impractical to do so without custom riser boards, I just took a chance and desoldered the EPROM. I got some serious cold sweat when I installed it in my tl866 and couldn't read it -- I had to disable "validate chip ID" or whatever the option is that attempts to identify the IC, and was then able to dump the contents.

[R-1125F]

I went ahead and tried it.  It seems to work.  I had to disable the chip ID to make it work.   The load on the programmer did not seem to be a problem.  The only readable text is the copyright notice.  That is a good sign.   

This board only has the EPROM and a RAM chip and two other TTL chips, so there is no processor running that will confuse things. 

The programmer did not have an exact chip ID so I used the closest one I could find.   The chips are nearly identical, with the only difference being the programming voltage.   Since I am not programming a chip, I hope this is not an issue. 

Thanks to all for the help.

[Swake]

How did you connect the programmer to the soldered chip? Did you us a clip (which one?) or some other method?

[R-1125F]

I used mini grabbers to connect to the chip and pin headers to connect to the programmer.  Photo attached.  I had to use what I had on hand.  I would have preferred to use a chip clip (logic clip?) to connect all the pins at once. 

[Swake]

Very nice. You where a bit lucky having enough room to fit them 
Thanks for the pic.