Electronics > Repair

Rohde & Schwarz SMIQ03B with Error 110 Output unleveled; IQMOD: ALC failure

<< < (6/12) > >>

zmetzing:
Yep, that was the video that I watched before bidding on this unit. He replaces the final output device, which did not fix the issue on my unit. I'll be replacing the two drivers before that final device after I get the parts from Germany.

Final device is a SHF-0189, purchased from link in the i-hobby video shown in the previous message, but can also be found on eBay.
The two drivers in the chain before that are SHF-0186 devices, currently available on eBay from seller "rici61".

smgvbest:
Did you check out the repair that DaJMasta and I did.   there's allot of information in those posts
I also produced a spreadsheet to record all the TP's in the diag section to help zero in on the issues.
you would just run your's up to the 3Ghz instead of the 6Ghz we did

this is DaJMasta
https://www.eevblog.com/forum/repair/rohde-schwarz-smiq06b-repair-log-6ghz-board-alc-error/msg1193985/#msg1193985

this is mine posting
https://www.eevblog.com/forum/repair/rohde-schwarz-smiq06b-repair-6ghz-board-level-preset-error/msg1240640/#msg1240640

zmetzing:
I did check out those posts and they were very informative. I'm keeping my eyes open for a 6GHz extender board and the FM modulator board (although this could be done externally by feeding the proper I/Q signals).

I await the SHF-0186 devices from over the sea, and then I'll remove the 100 tiny screws from the IQ Modulator board again. :-)

zmetzing:
Well, the entire output stage has new amplifiers, but this was not the problem. So, I've stopped shotgunning it and started tracing the signal path.

I performed the CW leveling procedure IQCON 7.4.4.4 and found that the output from that module was ~2dB low. That has been rectified, but that wasn't the issue, either. Note: The SMIQ03 will run without the IQMOD module installed ("Press SELECT to simulate hardware"), so that R495 can be accessed when the front panel is removed and rotated out of the way. I have my SMIQ on it's side, which seems to work the best for keeping things out of the way but still accessible.

I then traced down the signal from the input connector X247 (FIQFIL) [schematic page 17] toward the FAM3300 off-page arrow. I see the amplifiers doing their thing along the path, with a selected frequency of 450.1 MHz and +13dBm selected. My analog o-scope only goes up to 470MHz, but the levels seem reasonable until I hit the Pi-attenuator R966/R120/R113. On the C593 side of things, the signal looks great (1.070V or 4dBm, just what we expect). On the C343 side of the path, the signal drops to something very low, and doesn't look like the 450.1 MHz signal I was tracing. I found R966 to be 51 Ohms, R120/R113 to be 100.0 Ohms, and there was a C590 piggy-backed across R966 which seems to be 1pF. This does not agree with the 1084.9800.01 schematic, but my board has a .08 revision, so maybe that's ok. This would be a 8.5dB attenuator, without the effects of the capacitor, which would allow a high-frequency bypass path for better leveling, perhaps?

I went ahead and replaced R966, but no joy. I can only think that something with the D_SWITCH circuitry is effectively grounding out the signal.

Next, I plan to lift C343 and see if the problem persists, to better isolate the problem.

zmetzing:

--- Quote from: eb4eqa on April 28, 2017, 01:19:42 pm ---I've repaired two SMIQs with that same error and the problem(s) were always caused by either RF amplifier(s) (MMIC or final FET) and RF switches (old Hittite parts). It seems R&S decided to drive the control voltajes a bit harder to improve isolation between ports, which has an impact on the part's life. I replaced them with something similar and modified the resistor networks to trim the control voltajes down to safer levels. I believe I have some of those parts still somewhere...

--- End quote ---

I seem to have the same issue with the switch S2 (SSW124). I removed C343 and observed a reduced amplitude, but still good, 450.1 MHz signal on the hot end of R113.

Replacing C343 and then breaking the signal path at C333, terminating that leg into a 47 ohm resistor, results in a garbage waveform of not much amplitude on R113. So, it looks like S2 has bit the dust. I'll be PM-ing you, EB4EQA, to find a suggest replacement part.

Postscript: Doing the math, I see that R&S is putting nearly -10V on those switch inputs, which are nominally -5V. I may increase the value of R36 substantially, perhaps to 4.7k, to reduce the voltage on those lines to something closer to -5V. Any thoughts?

Navigation

[0] Message Index

[#] Next page

[*] Previous page

There was an error while thanking
Thanking...
Go to full version
Powered by SMFPacks Advanced Attachments Uploader Mod