[time-nuts] Weird HP 5334B problem

[Didier Juges]

My HP 5334B has been doing the following since I got it off eBay this 
summer. It's not a major problem, but it is irritating and I would like 
to fix it, but I don't know where to start...

At power up, when being fed two signals for TIC measurements, the 
instrument passes self test without error but is stupid: the trigger 
LEDs blink wildly without relation to the signals being fed, and the TIC 
measurement is stable but wrong (for instance, it would display 99 uS 
when fed two signals with 20 nS delay).

As soon as I press the "READ LEVEL" button, the trigger LEDs start 
blinking in phase with the signal (assuming the trigger levels are set 
right, of course), and if I then press the TI button, it all works as it 
should. So I know now each time I power up the instrument, I press READ 
LEVELS once before doing anything else, and from there on, the 
instrument works as it should.

It seems as if the trigger level DACs are not being initialized properly 
at power up, and pressing the READ LEVELS button fixes it.

I did the READ LEVELS tests in the service manual and it passes.

My counter is serial number 2804A01295, the service manual I have (from 
the HP web site) applies to instruments with prefix 2937A, even though 
it does list some changes that apply to early 2704A series. 
Particularly, there were some changes to the reset circuit with the 
early 2704A series, I do not know if these changes are in my instrument.

Does anyone knows the differences between the 2937A series and the 2804A 
series? Anyone has a service manual for the 2804A series?

Thanks for any help.

Didier