Help repairing vector signal generator with low output

[K1JOS]

I acquired an R&S SM300 signal generator has some output issue only between certain level ranges independent of selected frequency.  The self-test program passes all internal testing.  The following is a description of the symptoms and perhaps someone can advise if I should be focusing on the attenuator or power amplifier?  On my recently calibrated HP 438A, the SM300 output seems linearly accurate from -100dBm up to -26dBM but at all levels tested it is exactly 10dBm lower than as set on the front panel controls, i.e., at -26 dBM my power meter reads -36dBM.  As soon as I set the front panel controls up 1 dBm to -25dBm, all signal is lost (or it is less than I can measure ~-120 dBM).  Signal remains lost until I reach a setting of 0 dBM where the signal returns but my power meter only shows -36 dBM and increasing higher than 0 dBm doesn’t do anything – signal still there but at -36dBM.  The SM300 should go up to +13 dBm.

The R&S service manual is unhelpful other than saying replace the board… no schematics and nothing on component level.  I would like to get some idea of how to trace through the problem and where to start - attenuator or the wide-band amplifier section.  I have contacted several repair services – most say they don’t have enough technical info from R&S to repair – while 2 others want 2x the current asking price for a functional used SM300.

Any advice appreciated.

Jerry NY2KW 

[K1JOS]

I believe the issue is in the attenuator - either control or the attenuator(s).  Any advice how to test/troubleshoot??

[nonlinear]

Could not find easily whether this thing has mechanical or electronic output attenuator.
Suspect that it has an electronic one as symptom is not typical for mechanical attenuators.

General approach - go from the RF output connector to the boards or blocks and check the levels.
You can use low frequency (1 MHz or something) and use oscilloscope to check for signal at different stages.

Post photos of internals, PCBs here, they will allow to advice something more specific.

[analogRF]

I believe the issue is in the attenuator - either control or the attenuator(s).  Any advice how to test/troubleshoot??

Did you finally fix the instrument?
Do you have any photos of the RF PCB sections where the attenuators are?

[K1JOS]

My apologies for not getting back here to report.  Its been a tough year with lots of personal distractions making me put projects on the back burner.  Several months ago I found a short exchange between a few owners of the R&S SM300 who had identical problem as me.  Both commented :   "I replaced the gain block AGB3303 by an BGA6589 (also a SOT89).  The MMIC is located at 16cm from the N-connector."

So although I dont have a schematic or parts diagram, my plan was to trace back from the N connector and verify that the gain block is where the problem lies.  I found a BGA6589 part on Ebay and hope thgat it is correct and original.  Hope to get to this in the coming weeks and will report back

thanks again

Jerry