Tektronix Sampling Head SD-26 EEPROM Failure and Repair

[Andie]

Hello everyone!

Recently I acquired a Tektronix CSA803C, along with two SD-26 sampling heads. I was quite happy with it, until one of the sampling heads stopped being recognized by the scope. 

Due to the fact that just before the failure I was playing around with the scope-built-in-calibration of sampling heads and with the help of the scope's extended diagnostics, I was quite sure that there was a problem with the sampling head's EEPROM. Examination of the unsoldered EEPROM showed that it could be read, but writing to it failed. The simplest repair seemed to be cloning the EEPROM of my second, still working, SD-26. The factory calibration of the defective device would be lost, but after a successful user calibration this wouldn't be a problem.

Solely the repaired sampling head would then carry the same serial number as the donating device ...

The sampling head's serial number can easily be found in the hexdump, but changing it requires updating the checksum which is stored in the upper two byte of the EEPROM. What checksum algorithm does the scope use? Unfortunately I was too lazy to try out.

It turned out that it wasn't necessary to know the algorithm. Comparing the hexdump of the EEPROM with the error message from the oscilloscope revealed that the error message reported the checksum stored in the EEPROM as well as another value, which appeared to be the computed checksum, based on the actual content.



Extended Diagnostics of CSA803C


So the steps would be:

1. Clone the EEPROM of the working SD-26 and at the same time change the serial number.
2. Put together the sampling head and let the scope throw an error, due to invalid checksum.
3. Check the extended diagnostics in order to learn the new checksum.
4. Write the new checksum to the EEPROM.
5. Put together the sampling head again. Done.

This requires connecting the EEPROM to the board twice. Not knowing how often I had to alter the EEPROM content, I put a socket on the board with the intention to leave it there forever. A flat socket with the pins shortened. Modified and shortened pins on the EEPROM. The EEPROM casing was ground down to lose 1mm from the top.


In the background SD-26 with replaced EEPROM, in the foreground SD-26 with replaced EEPROM, socketed

Regards,
Andreas

[rhb]

Thanks for posting this.  I just bought an 11801, a pair of SD-22 heads and a 1 m extension cable.  I don't have any experience with these.  It will be an adventure.

The unit I bought has an E5622 error which is the NVRAM.  I found those at Digikey, but I'd like to replace the other two lithium batteries, but I cannot find out what the battery type is.  The manual just says a BT130 and BT150.

[dcarr]

In my notes I found this:
BT130 on A14 I/O board - RTC, 2.7V threshold
BT150 on A18 memory board - 2.7V threshold
Lithium Poly-Carbon monoflouride, BR 2/3 A
Panasonic BR-2/3AE2P or BR-2/3AE2SP