If anyone is interested, see attached FW image for 663(19/21)(B/D) version 3.01. This is the firmware which supports datalogging as per some other threads, e.g.
https://www.eevblog.com/forum/testgear/66319b-firmware-image/.This came from my 66309D, which was sold "as-is" as a 66319D, with the 66319D firmware, front label and FAKE rear serial number sticker (the real one saying 66309D was still underneath). The price however wasn't high even for what I got vs what it was sold as (about 90 GBP shipped) and it wasn't obviously broken other than lying about the model name so I didn't make a fuss. Some observations about changing firmware and other oddities on my unit:
- The hardware clearly isn't that of a 66319D, as it is missing the extra PCB with analogue multipliers etc used for the output resistance control function.
- Running 66319D FW on a 66309D seems to work for the most part - the 1A range (unique to this model) is broken and resistance setting does nothing. I didn't test the fast datalogging capability but I suspect it will work given there is no difference in the logic/ADC board. This may be of use to those who want the fast datalogging on a model which doesn't support it otherwise, and are happy with losing current autoranging capability (manual range set is still OK if you avoid 1A).
- The EEPROM holding calibration etc is not necessarily compatible between the different FW types (in my case trying 66309 and 66319) - I had to put either a 66309D EEPROM image (thanks "TheSteve" for helping with this) or blank it totally to run the correct 66309D FW, otherwise I was seeing issues like the fan not running.
- After changing the FW or EEPROM image you need to do the initialisation procedure and calibration, and for a blank EEPROM also set the SN as noted here:
https://www.eevblog.com/forum/repair/hp-66309d-issues- My unit had a zero ohm link fitted at R427 (with flux everywhere - clearly not factory), which broke the 20mA range, DVM and 2nd output measurements by stopping the analogue mux which switched these into the ADC input from selecting the right channel. Major WTF as to why this was done, removing the 0R link fixed all this.
- It also had U327 obviously removed - this is a LM311 used for overvoltage detection; replacing it restored this function (another WTF)