any tricks to retrieve a NEC 23256 ROM data ?

Don't think so.. Sorry.. Not only that, I think you may have even exceeded the programming voltage! :(

It might be possible, but it would probably be very expensive.

How much is the data worth?

of

wow, very interesting. I would need a pretty good %99.999 off coupon to use that service

robb

I think you're finally out of luck.

Is the manufacturer of that board / equipment still around ?

Graham

Yes. For masked ROMs you can decap the part and read it optically.

Decap is easy for ceramic; you can just break it open. For plastic packaged parts, it involves fumic nitric or sulfuric acid, so you'd be best advised to send it to a lab.

Once you can see the die, under a microscope you can distinguish the one bits from the zero bits. Generally that is the presence or absence of a metal contact. Whether the presence represents a one or a zero depends on the specific part number, and might vary within the array (i.e., it might be presence=1 for the left half of the array and absence=1 for the right half).

You also will need to determine the layout of the bits. The address lines are divided into those that address rows and columns. Study of the decoder circuitry on the perimeter of the array can help determine this.

If you can take a sufficiently high-resolution photomicrograph of the die, once you've determined the mapping you may be able to write software to extract the bits, so that you don't have to do it by hand.

None of this is easy, but there is an existence proof that it is possible. Peter Monta optically dumped three ROMs from the HP-35 calculator. These stored 2560 bits each, and were made in 1972 using ten micron process geometry, The 23256 is a much newer part in a smaller geometry, which means you'll need better resolution for the photomicrograph, but it still should be quite possible.

Or you can just buy another unit of the model the 23256 came from.

Eric

WHAT programming voltage? It's a mask programmed ROM

Too late now, so you'll just have to live with the lobotomy. :(

Oh sorry Mr. Terrell your holiness, I obviously over looked that part, I just assume it was a prom or even UV-prom..

I'll try not blunder so badly next time. I wouldn't want to be the cause of any health issues on your end.

IF you drive the logic inputs from CMOS, it's possible that internal protection diodes will sufficiently power the chip from the input logic. As long as at least one address or select bit is HIGH, there's a chance the outputs will go to the right voltage level for readout.

This will work best if (1) the chip is cold, (2) the outputs are very lightly loaded, or disconnected completely, (3) the only fault that resulted from overvoltage was a burnt-out power wire, (4) the readout is done slowly (a millisecond delay for each readout should suffice).

(compatible

pin.

internal

input

readout.

are

fault

readout

thanks for idea, are there ROM programmer/readers that allow you to specify reading to that degree ?

are there pin electrical reading/ checks that need to be performed to verify that this is possble ?

thanks for help and ideas, robb

of

optically.

plastic

you'd

distinguish

presence or

one

half

address

Study

of

do

is

HP-35

using

for

from.

thanks eric, i think your last idea is more possible/probable and thrifty :)

thanks again for help, robb

Not bloodly likely. The chip will be damaged much more quickly than the supply (or ground) bond wires will open.