R&S FSIQ 3 RF converter repair

[EricG]

I have Rohde & Schwarz FSIQ 3 signal/ spectrum analyzer that's been in storage for a while while I moved. When I powered it up this weekend the "OVLD" indicator remains on even with no input connected. running the self test indicates indicates RF converter failure. The almost useless "service manual" says that this is indicative that the first mixer has failed and the whole RF converter board needs replacement. I'm not happy with the prospect of finding a replacement since R&S p/n 1065.6768.03 seems to be almost unobtanium.

However, I noticed in the system messages menu that message 10130, "Err Reading I2C EEPROm on RF_MODULE board" was displayed. which brings me to my actual questions.

1. Is it possible that the EEPROM contains settings that allow the RF converter board to function properly and that just replacing the EEPROM might bring this board back to life? provided of course that those setting could be restored through calibration or some other means. If so, has anyone attempted this?

2. is there any USEFUL service information for this instrument (detailed schematics etc). the Service Manual isn't very detailed or helpful.

3. Is there a good source of R&S spares somewhere that has eluded me? searching Ebay and other sites does not drag up very much.

Any help would be very much appreciated

[EricG]

On the off chance that rohde and schwarz still had replacement parts for these I contacted their service dept. they had 2 modules in stock. a refurbished module and a new(!) one. R&S wants ~$10k for the new part and ~$2k for the refurb. Considering that working, units on ebay are running in the $2500 range, replacement parts seem a bit spendy.

one thing that I did notice when I removed the module from the chassis it that there was no cover on the front and back side of the module like there were on the other modules. this leads me to believe that the previous owner may have mucked about with the unit prior to selling it to me.

[ogden]

However, I noticed in the system messages menu that message 10130, "Err Reading I2C EEPROm on RF_MODULE board" was displayed. which brings me to my actual questions.

In case it is not dead EEPROM but whole I2C bus failure (disconnect or lockup) - that may lead to whole lot of other problems including loss of module functions. There could be other I2C slaves such as gpio extender chip(s) that often are used to control digital step attenuators, VGA's and so on. You shall check cabling/connectors first, perhaps look at I2C bus using scope.

[EricG]

I think that the I2C bus is common with all the modules, so if there was a bus failure, I would see other modules report similar errors, but as I said I'm lacking any reasonable service documentation to say anything educated.  All of the modules in the system connect to a common back plane for command/control/power, with coax jumping from module to module carrying the high frequency stuff so the number of cables to check is pretty small. I think next steps are to monitor the I2C bus to see what's actually happening as you suggest.

[ogden]

I think that the I2C bus is common with all the modules

I think that would be worst design decision one could make while building very complex and highly modular device. Better do not assume anything, verify everything

[EricG]

So I just found another service manual that's a but more complete. regarding the EEPROM it states:

"The correction and setting data for the boards are contained in an EEPROM on each board and are thus
replaced with board replacement."

so I guess that partially answers my original question #1, the question of what's on the EEPROM. It still doesn't answer the second part of that question, how do I replace the data on it if it is bad.

I'll hook up a scope to the eeprom and see what's going on there as a next step

[ogden]

so I guess that partially answers my original question #1, the question of what's on the EEPROM. It still doesn't answer the second part of that question, how do I replace the data on it if it is bad.

You replace EEPROM data same way it was initially written - by doing (factory) calibration of module/instrument. You may trick module & instrument back into operation by using EEPROM contents from another module/instrument - to check that module calibration data is only problem that have to be solved.

[Joni]

Hello!
I think the eprom data are also stored on the hard drive.
In the catalog D:\RUNTIME\CAL there should be a file named RF.DAT

[eb4eqa]

Hi all. I'm looking for the FSIQ Service Manual and I've been unable to find a copy. Someone willing to share it with me?

Thank you!
Roberto

[sbvr4]

FSIQ series is almost identical to the FSEA,B,M,K. Search for FSEA service manual, and you should find what you are looking for.

sbvr4

[Astralix]

I am also on the hunt for a true FSIQ Service Manual...

...but the FSE manuals are enough to fix that RF OVLD problem, when it comes with the message EEPROM RF Unit unreadable.
Follow the "service menu" instructions from the FSE manual and use the code to backup and restore the EEPROM of the RF unit.

I got a unit with the same issue and fixed the issue the following way:

Basic precausions:
- cloned the drive
- copied the whole c:\r_s directory, just because.
- Ran SELFTEST which failed with RF Unit EEPROM unreadable.
- Ran Full Calibration, which failed... guess why

Checking EEPROM of RF-Unit:
- Read RF unit EEPROM using service command 2.11.3.0 what worked without error.
- Using 2.11.3.1 to write back the copy of the RF.DAT without error.
- Restart of analyzer firmware still records an EEPROM error in the log.
- Result: I2C and EEPROM are fine, but EEPROM data is corrupt.

While testing around, I made some interesting find... there was another RF.DAT in the trashbin of the system...
- I restored RF.DAT from the trashbin (right-click --> restore)
- Service Code 2.11.3.1 wrote something back into the EEPROM of the RF Unit.
- Restarted the analyzer firmware, and yeeha! It works!
- Ran Selftest which now completes.
- Ran Full Calibration which worked fine.

I have no clue if someone tried to repair the fault or just tried to disable the unit...
But as the calibration also went through, I think the deleted RF.DAT was the exact one for this RF board.

Astralix