Battery backup RAM reading

[jfphp]

Hello,
I would like to read with an ordinary programmer a RAM backed by an external battery (not an internal battery like DALLAS easy  !) on a board (purpose store the cal constants of DATRON calibrators 4000, 4200...). I cannot, of course, disconnect the battery because everything will disappear.  Any idea ? Thank you.

[mojoe]

Use a DIP-clip and some wire leads to attach the SRAM to your device programmer. I've done this with both parallel and serial SRAM.

[VK5RC]

Edit: mis-read qn.,  D'oh

[Le_Bassiste]

did this on one of my HP3468A, but used a logic analyzer, not a programmer.
connected a  LAP-C32128 (hacked to 64Mb) to all pins of the RAM, set the LA to one-shot each time and ran the multimeter through its different modes (VDC, VAC, OHMS, you get the picture). then put the data into excel, kept only the data where CS and OE were active and sorted the RAM output by address values. tedious, but it worked.

[uncle_bob]

Hi

A lot depends on exactly what happens to the device that normally *drives* the RAM when the RAM is running on battery. It's probably not a good thing to read the RAM and blow out the gizmo it's attached to. Most of the time this involves powering the whole thing up and holding the MCU (or whatever) in reset while you read the RAM.

Bob

[Le_Bassiste]

Hi

A lot depends on exactly what happens to the device that normally *drives* the RAM when the RAM is running on battery. It's probably not a good thing to read the RAM and blow out the gizmo it's attached to. Most of the time this involves powering the whole thing up and holding the MCU (or whatever) in reset while you read the RAM.

Bob

you probably misunderstood my statement. i didn't "drive" any lines in the unit. it wasn't even necessary to hold the MCU in reset. i simply "sniffed" all the pins of the RAM while exercising the unit in its different operating modes.

[Kleinstein]

The first choice would be getting the calibration data through the normal interface (GPIB,RS232,..).

Otherwise it's up to keeping the CPU in Reset state and hope one can takeover the bus. Or just sniff the normal operations.

[uncle_bob]

Hi

The limitation to the "normal operation" as compared to the "takeover" approach is the assumption that you can catch it reading everything in a finite amount of time. With some of my code, you would have to wait a couple of days to get it all in normal operation.

Bob

[Le_Bassiste]

 i was lucky not to come across any of your code in my HP3468A

[uncle_bob]

Hi

All of my stuff is buried in some pretty obscure places. There's a lot of it out there (pieces wise). You just are highly unlikely to ever run into it.

Bob