3458A broken GPIB?

[essele]

In a moment of utter madness I've bought another 3458A ... I don't really know why, I guess I'm addicted!

Anyway, it was a faulty unit with a high memory error, which I've already fixed with my previously used FRAM boards for memory and cal data.

It also had a fan that sounded more like a circular saw - now replaced, and the front terminals had been pushed back and broken the plastic screw fixings off of the front panel ... now fixed with a 3d printed part and some epoxy.

But ... the GPIB interface isn't working, nothing detected at all ... I spend bloody ages trying to determine this since my natural assumption was it was a failure with my own-built GPIB interface, but I have now confirmed it works on another 3458A, and another (not of my making) interface also shows the same behaviour.

The connector looks ok and checks out with a continuity tester ... so my assumption is that the GPIB line drivers are faulty ... is that a reasonable place to start?

I haven't started troubleshooting them yet (still waiting for the epoxy to dry!) ... I'm also assuming that it's reasonably easy to see if they are working. I can setup a regular activity and trace the signals through the drivers.

Has anyone else seen these fail?

This unit has a DS75161AN and a SN75ALS160N.

Thanks.

[IanJ]

Hi,

Yes, I bought a faulty 3458A a couple years ago and GPIB was faulty, same issue I.e. absolutely nothing.
I replaced both driver IC'S and it was working again.

Ian.

[wolfy007]

Did they show a HPIB error on startup?

[essele]

Now fixed.

I was probing the driver IC's with my scope whilst running some repeating comms and everything looked fine .. I then retested and it was fine ... disconnected my scope and it failed again.

The GPIB connector didn't appear to be grounded properly, although the schematic does indeed show a different GND (DCOM2) .. however this did remind me that I'd seen a service note to do with GPIB comms (3458A-13A) so thanks to xdevs I managed to find it and it was indeed the problem.

https://xdevs.com/doc/HP_Agilent_Keysight/3458A/service/3458A-13A.pdf

So all fixed.

[CalMachine]

Now fixed.

I was probing the driver IC's with my scope whilst running some repeating comms and everything looked fine .. I then retested and it was fine ... disconnected my scope and it failed again.

The GPIB connector didn't appear to be grounded properly, although the schematic does indeed show a different GND (DCOM2) .. however this did remind me that I'd seen a service note to do with GPIB comms (3458A-13A) so thanks to xdevs I managed to find it and it was indeed the problem.

https://xdevs.com/doc/HP_Agilent_Keysight/3458A/service/3458A-13A.pdf

So all fixed.

I've had this same issue on one of my 3458As before.  Following this Appnote 13A fixed my issue.  Glad to see it has fixed others!   

[bingo600]

fixed with my previously used FRAM boards for memory and cal data.

@essele

Where did you get the FRAM boards from ?
I'm looking to use FRAM on my '58 , but would love to get a PCB layout for those boards.
Instead of DIY'ing them

/Bingo

[essele]

Hi Bingo,

They are my own creation and manufactured by JLCPCB. I had a couple left over from my other meter, so it was easy.

The schematics, layout, and gerbers (and a photo) are all attached on this thread:

https://www.eevblog.com/forum/metrology/3458a-upgrade-questions/new/?topicseen#new

I was lucky enough to be able to program the CAL FRAM using an SOIC adapter before soldering it to the board. I haven't tried programming it once on the adapter ... obviously it should work, but I've not tested it.

Lee.