Agilent 8935 E6380A test set repair SOLVED

[Howardlong]

Folks,

I received one of these units last week but it was DOA, the seller had not run all the tests.

The 8935 is a comms test set including RF sweep generator, spectrum analyser, and analogue transmitter and receiver tester, pretty much the Swiss Army Knife of RF with coverage from 400kHz to 1GHz. It also has cellular stuff but I'm not interested in that.

The module level service manual's available but I couldn't find any schematics.

First problem was to fix the low battery warning, six T15 screws two AA batteries later all good.

However there was no RF out or in on any port irrespective of mode.

I ran through the many integrated tests and after twenty minutes or somit was clear that the 40V supply wasn't happy.

So I took it apaaaart, just the front and back plastic covers, side covers and handle, and power supply. 30 or so T15 screws later (all the same screws, beauty) I was in like Flynn.

I measured the supplies, and the 40V one was about 3.5V, not a happy camper. So I took the two power supply boards out to check for visible damage but there wasn't any.

The 40V comes from the A3A2 board which is no touchy due to mains voltages all over the board, and it's proper full fat 240V in these parts too. I visually traced the 40V to a TO220 device (U2) with a clip on heat sink undeneath a load of the rubbery gunge stuff they put on PSU boards to stop the large caps flapping around in the breeze. After removing some of the gunge and the clip on heatsink, imagine my delight when I read the immortal marking "LM317"! Sweet!

I desolder it, set up a little test and sure enough it's dead. So I replace it with one from stock, reassemble the unit and hey presto she's fixed. All the tests run bar some of the tests on the 1.7-2GHz cell band, but I'm not fussed about those.

It took about a couple of hours to take apart, diagnose and fix. Hopefully if someone else encounters similar symptoms this little story will help them too.

[TheSteve]

I love it when the bad part happens to be something common. That is a nice test set and will come in very handy for all kinds of radio related testing.

[Howardlong]

I love it when the bad part happens to be something common. That is a nice test set and will come in very handy for all kinds of radio related testing.

Yeah, to go with the 2955 and the two Anritsus I already have :-) 

[HighVoltage]

I had to look the 8935 instrument up.
Does it look like the one in the attached picture?

You got very lucky with the repair on such a complicated instrument, congratulations.

[Howardlong]

That's the one. It's a bit of a boat anchor and I'm still learning how to drive it. I've had the spec an working, swept a filter, used a return loss bridge with it, tested a transmitter, but I am still figuring out how to get a receiver SINAD test to work, it is showing very low, 8dB or so, on a full quieting signal, or indeed a 1kHz sine direct from a signal generator! Hopefully it's just operator error, but this is the main test I intended using it for to test receiver sensitivity on a production line. The UI is wuite a bit more complex than other test sets I've used, with several screens to configure.

[Howardlong]

Update: I figured out the SINAD problem, it was operator error, you need to ground the Audio In Lo input in the AF Analyser screen.

If in doubt, RTFM and refer to the comprehensive testing instructions and diagnostics!

I am trying to figure out if it's possible to get the SINAD reading to be large text as that's its production use. The workaround at the moment is an external monitor, it has a composite video output.

Another nice feature is that you can set it up to switch on in your favourite setup, and setups can be saved and recalled locally on the device rather than having to find PCMCIA flash cards.