Wiltron / Anritsu 69147A 20 GHz Sweep Generator - Error 113 YIG Loop Osc Failed

[John_ITIC]

Hello, I'm going over my test instruments to bring bad ones back to life. This Wiltron 69147A was purchased on Ebay in 2019. The unit had some lock error but that was repaired before I purchased the unit. I have used the unit only a few times over the years and it started (a couple of years ago) having issues with the YIG locking over approximately 11 GHz (Error 113). When taking it out of storage this week, I noticed that the issue happened over many other bands. After having had the units on some time for trouble-shooting, the issue seems now permanent, except the first 10 seconds after turning it on. I'm suspecting something gets hot and fails somehow. And whatever it is, seems to degrade over time - even when the unit is not in use.

The closest service manual I could find is "Anritsu_Wiltron_680XXB_681XXB_SYNTHESIZED_CWSWEEP_GENERATORS_Service_Manual-10370-10262.pdf", here: http://ftb.ko4bb.com/getsimple/index.php?id=manuals&dir=Anritsu-Wiltron

That manual describes the Error 113 and has trouble-shooting diagrams. But, I noticed that my card layout is not the same as the 68xxxB units. So, before starting the trouble-shooting, I was wondering if anyone happens to have the proper 69xxxA service manual?

Note that the A4 block has a different style than the other blocks. Is this correct? Or was the A4 swapped out by the seller before I purchased the unit?

Thanks,
/John.

[John_ITIC]

Attached is the ERR 113 self test result.

[John_ITIC]

I may have a clue what is going on. Per the attached images and PDFs, the A4 "course loop" module should output a frequency in the range 219.5 MHz to 245 MHz range, that varies with the configured output frequency. It does. But it also outputs many multiples of this frequency with such high amplitude that my frequency counter has a hard time locking onto the base frequency.

When looking through the maintenance manuals for similar units it is not quite clear whether these harmonics should be there.

The MG369xB maintenance manual states (see attached block diagram too):

One input to the YIG loop is the 219.5 to 245 MHz signal from the coarse loop. This signal is amplified to drive
the step-recovery diode (located on the A7 PCB). The step-recovery diode produces harmonics of the coarse loop
signal (≥ 1.9755 to > 20 GHz). These harmonics are used by the sampler.

The 680XXB/681XXB maintenance manual states:

The coarse-loop oscillator has a programming (tuning)
range of 439 to 490 MHz (219.5 to 245 MHz for
units with serial number 512001 and above) and a
resolution of 1 MHz. This provides harmonics from
1.9755 GHz to >20 GHz. This allows any YIG tuned
oscillator output frequency to be down converted
to a YIG IF signal of 21.5 to 40 MHz.

My unit is architecturally newer than the 68XXXB but older than the MG369xB. I have not been able to locate a maintenance manual for a 69xxxA unit.

I believe the wording of the newer document for the MG369xB  is closer to the truth. Both the attached block diagrams shows that the harmonics 1.9755 to 20 GHz are drawn after the SRD (Step Recovery Diode).

So, it looks like a low pass filter in my A4 Course Loop block is bad, such that all these harmonics are let through?

Thanks,
/John.

[John_ITIC]

Does anyone happen to know whether the signal paths in the YIG loop are supposed to carry spectrally pure sine waves or whether it is only the zero crossing interval that matters (i.e. distorted sine waveform is okay)? I measured the output from the A3 (10 MHz and 100 MHz references) and A4 (course loop). Only the 100 MHz signal has a clean sine wave, the other two are distorted sine waves. in particular, the A4J2 output resembles a square wave, which explains the many harmonics in the frequency domain.

Thanks,
/John.

 

[John_ITIC]

I managed to find the 69147A maintenance manual here. The link says user manual but it is the MM.
https://docs.alltest.net/manual/Alltest-Anritsu-69177A-UserManual-2266-.pdf

I was very happy to find the correct MM tonight as neither the 68XXB or MG369xB matches exactly my unit.

Thanks,
/John.

[metrologist]

Ah, I was going to post that link. It might just be the YIG driver - top of YIG, or a regulator circuit if voltages are not correct. YIG loop was always a little tricky to TS.

And there will be harmonics on the lower loop outputs. Anyway, report back what you find working through the procedures.

[Mkpirulo]

Hi John,

I was also about to send a link to that manual as well. 690xxB/691xxB Maintanance manual and the MG369xB will be your closest bet, at least it was for me.
I have a 69347B with options 21B and 2A. My unit has various problems one being -15V. This issue was fixed when I traced it down to a shorted feed thru. Now I still have the 113 error. I checked all frequencies and voltages. Only thing I found was the 28VT is around 26.5V but I don’t think that’s an issue. I’m focused on the YIG at this point. My unit losses signal around 15.5Ghz and comes and goes all the way to 20Ghz. I’m looking at the YIG output vs drive voltage via the switch modules sample port. My next step is to see if the drive to the YIG is correct around 15.5Ghz, through the drop out region.
If you have any pointers that would be great. Besides the 113 error it’s also strange that the 28V is low(26.5). I replaced that regulator but it did not change the voltage. Manufacturer’s recommendations are not to exceed 35V input voltage but it’s around 43V. This higher than normal drive might effect the output voltage. Wish I had another unit as comparison.

Thanks,
Michael

[John_ITIC]

Hello. Vacations, work and other work around the house have gotten in the way but I have now narrowed down my issue to the A7 YIG Loop board.

It turns out that the A7J1 input (from the A4 Coarse Loop) goes deaf for frequencies over 300 MHz or so.
At 4000 MHz RF out, A4 outputs 212.641MHz/-0.86 dBm - A7 locks OK.
At 4100 MHz RF out, A4 outputs 414.003MHz/-1.42 dBm - A7 fails locks. Same for higher A4 frequencies.
With A4 replaced by external signal generator 414.003MHz/+5 dBm - A7 locks OK.

The service manual states that the A4J2 output should be 0 dBm +/ 6dBm, so my A4 seems okay. The A7J1 input must be deaf.

Now, unfortunately, the only A7 YIG Loop board I have found for sale is way too expensive. That is half of what I paid for my 69147A.
https://www.ebay.com/itm/145987476741

So my next step is to take my A7 apart and try to trace out the schematics.
Then, try to identify what part possibly could act as a low-pass filter on the A7J1 input.
Per attached block diagram, could also be an issue with the SRD (Step Recovery Diode) or Sampler?

If anyone happen to have a A7 6800-D-40627 Rev C board for sale, please let me know.

Thanks,
/John.

[metrologist]

You'll need a card extender to TS the boards.

TP2: +5
TP3: +10
TP4: +15
TP 1,9: AGND

IIRC, the on-board TPs have an orange donut.

A7J1 goes to U8 MAR-3S -> U9 MAV-11S -> U10 MSA0520 -> J4 to the sampler. J2 and J3 are +/- IF

The IF goes to U5 -> U6, both MAR-6S -> J6 IF monitor, 21.5 to 40 MHz, -20 to -35 dBm (-26 dBm with sampler RF level at -10 dB)

U6 -> U7 MAR-3S -> Switched LPF CR5 to CR8 (85 MHz unlocked; 65 MHz locked w/FM; 50 MHz locked w/o FM) -> U13 MAR-6S -> IF limiter

TP5 is the IF Level, 3V typ RF ON, 0.4V typ RF OFF, TP7 is IF

IF -> U26 and U27 phase/frequency detector, pins 8 fine loop, pins 14 IF -> Q7 to Q10 frequency acquisition control

TP6 is lock monitor

TP8 is U39 and U39 FM fine loop divide by 128

[John_ITIC]

Thank you. I will try to correlate your information with my physical A7 board traces.

[EggertEnjoyer123]

I'd suspect the amplifier. Since everything is relatively low frequency (before the SRD), you can probably use a scope to check whether the amplitude is increasing after each amplifier.

What is probably happening is that one amplifier IC has failed, and the output of the sampler is too low for the loop to lock at some frequencies. The sensitivity varies a lot for the PLL, and generally some frequencies would lock and others wouldn't. I had a similar issue in one of my RF generators where some frequencies wouldn't lock and some would, and it turned out to be a dead RF amplifier.

[metrologist]

A7J1 goes to U8 MAR-3S -> U9 MAV-11S -> U10 MSA0520 -> J4 to the sampler. J2 and J3 are +/- IF

It should work with -4dBm input to U8 +10dBm out, U9 +17dBm, U10 +23dBm

Low noise supplies are +10, +10, +15V