开云体育

My HP 8660B/86602B (owned since 1995 - I've owned it longer than it's initial life in industry, I believe) failed to produce a stable output signal when powered up a few days ago (it has worked properly over the years, although it's use is not frequent.) I wish to tackle repair (at least as a learning experience.)

When tested today at some spot CW frequencies (1, 2, 5, 10, 20, 30, 50 and 100 MHz), output signal is below nominal by 200-300 kHz and is not stable (movement around carrier seen on spectrum analyzer display and listing on a receiver yields a noisy and unstable signal.) A swap-out of the 86602B RF plug-in (I have a spare) yielded no difference in behavior.

I have original operator/service manuals for the three key components (the 8660B, 86602B and 11661B frequency extension module inside the 8660B) and associated extension cables/boards (the 11672 service kit and 08660-60070 extender card set), so at least not flying totally blind.

My getting-started question is from this observation:

The 8660B manual's adjustment/test procedures don't mention whether the 11661B frequency extension module's behavior should be considered (as far as I can see, there is no mention of the 11661B in the 8660B manual at all.) It appears to be silent as to whether the 8660B should be evaluated with the 11661B installed (or not.) The 11661B manual focuses on that device and presumably assumes that the 8660B containing it is working normally.

Where (at the 8660B itself or the 11661B FEM) should I start troubleshooting?

Thanks for any suggestions (other than recycling the hardware and replacing it with a TinySA or similar)!

First experiment - swapped 11661B module with a supposedly-OK "just in case" spare I had picked up a long time ago (with understanding that the 11661B manual has an adjustment procedure to use when module swap/move occurs, which I didn't perform).

The output problem remains (fluctuating around 100-200 kHz below the set frequency), so taking this as a clue that the 11661B is not necessarily the problem, and starting into the fault finding tree in the 8660B manual.

Hi Ed, WB2UAQ ---

Thank you both for quick and helpful replies!

Ed - with your suggestion, I remembered that I do have a 11707A and it is now in the unit (I've never used it before as I've not had a 8660 failure, and I don't believe I've ever heard of a 11707A manual).

I also remembered that I have a 08660-90070 8660A/B/C service manual, which makes use of the 11707A test plug-in (an
online search for 11707A pointed me to that manual which makes use of it.)

Using the 08660-90070 manual, I am beginning tests:

The power supply voltage measurements on page first requiring those tests all show within spec.

The 350-450 MHz output is a couple of db below minimum spec (measured using 436A/8484A with 40 dB inline attenuator); all of the other RF level test points on the 11707A are above minimum spec.

I am now recharging TinySA Ultra (covers to 800 MHz) so I can look at the signal quality.

Thanks again, I'll update once I have any more specific results.

Dave

TinySA Ultra used to check outputs from the 11707A (except for the two 11661B-related outputs, as I forgot to enable the "Ultra" mode for coverage > 800 MHz.

The 500, 100, 20 and 350-450 MHz signals are stable and within spec except for the 350-450 MHz signal which is about 1 dB below minimum power spec.

The 20-30 MHz signal is unstable - while it has about -7 dBm output level (attached photo from the spectrum analyzer is down 10 dB as I have a 10 dB pad in the input path), the frequency is not stable, and the center frequency wanders (I'm not sure of the total extent of the frequency instability yet.) Spectrum analyzer display is attached.

The interesting thing at this point is that the spectrum display from the 20-30 MHz signal has the same deficiencies as the output from the generator itself. I do not know the expected frequency from the 20-30 MHz signal (the generator is in it's power-up frequency of 1.000 MHz), however, it is about 29.7 MHz, and measurements from the generator's output when I initially suspected a problem were routinely 250-300 kHz below the frequency entered on the keypad.

I've done no further testing (need get the TinySA Ultra into it's GHz-capable mode - this is no secret, just need find the appropriate instruction) so I can check the 2.75-4.05 GHz signals available on the 11707A.

Thanks again for the suggestions/help!

Dave

The 08660-90070 manual (HP 8660A/B/C service) has been helpful/friendly so far.

There are separate "big" operating/service manuals for 8660A, B and C; the 08660-90070 manual sends technicians to one of the "big" manuals at some point - I've not reached that yet.

The unstable 20-30 MHz output seen on HP 11707A test plug-in (typically been 200-300 kHz below specified frequency with considerable "wandering around" in that range) led me to another section in the 08660-90070 manual, which describes that frequency being driven by three synthesizers (called N1, N2 and N3). The three synthesizer outputs can be measured from the bottom side of the motherboard at pins shown in the same manual. After a little learning curve and oscilloscope adjustment, it turns out that those three signals appear to be correct (and stable) - the match tables in Appendix A.

N1, N2 and N3 feed two more synthesizers called SL1 and SL2 (Sl1 appears to be presented at the 11707A output.)

That's the next step.

I found this while searching for HP 11707A information:



If there is need and interest, perhaps upload it to the files section here?

I've found it to be very useful (more "cookbook style" than the single-model service manuals for 8660A, 8660B and 8660C.)

Peter this would be really helpful - in my searching, I've never found a manual (and I think that's a pretty common problem because while the operation is straightforward, no one seems to know the utility/meaning of the LED indicators, and of course being able to repair the thing would be invaluable!

Thanks,

Dave(s)

(thanks again, Peter!!!!)

... and now back to our regularly scheduled programming...

The SL2 loop oscillator output test point is frequency-stable. As SL2 output moves from 30 MHz (setting x.x00 000 Hz) to 20 MHz (setting x.x99 999Hz), it's amplitude drops by about 50% but at each test point, SL2 output frequency matches the expected values in table A-5 and does not vary.

The test instruction paragraph 5-13 does not indicate a minimum signal level for SL2 oscillator output. The "wandering" problem at the 11707A 20-30 MHz output doesn't appear to be correlated to SL2 output level (that is, it wanders regardless of the bottom 5 digits.) Also, the wandering is wider in range (up to 300 kHz from what I can tell) than SL2's available tuning range.

SL2 oscillator output is available at two different test points - A2TP6 (callout 27 in figure C3, and requested in the test instruction paragraph 5-13), and XA15-2 pin 14 (callout 38 in figure C3). I found XA15-2 easier to access (A2TP6 is buried under two horizontally-mounted plug-in boards behind the front panel DCU assembly); XA15-2 is a pin of one of the card edge connectors (not particularly easy to reach, but certainly easier to see.)

Next stop will be SL1 output (which from what I can tell sends signal to the 11707A 20-30 MHz test connector, so hopefully the problem will then be somewhat isolated.

Arrived at SL1 oscillator output - it is unstable (wandering over a few hundred kHz.)

Re-confirmed stability of N1, N2, N3 and SL2 outputs using a second instrument.

Attached is photo from SL1 phase error test point.

Next stop is the 8660B main service manual, as the 00070 manual goes into no further detail at this point (there are three plug-in cards and possibly other components involved.)

Today is "no good deed goes unpunished"...

Yesterday assembled four 15-pin extender cards (had the PCBs made a long time ago, but never built the 15-pin versions til yesterday.)

Initial goal has been to carry out verification and attempted adjustment of SL1's synthesizer which appears to be the source of the wandering problem in the 20-30 MHz composite signal provided to the RF plug-in. The 8660B service manual section 5-33 (adjustments for SL1) specifies test points (A2TPxxx) that are on the motherboard's top side, and adjustments made via the bottom side (enclosed PCB area). The four test point signals are also accessible on PCBs A15 and A19 (bottom side), hence the extender board idea. Photo attached of the two boards on their (new) extenders.

(photo attached)

Restarted the synthesizer in this configuration (no smoke emerging), counter attached to the 20-30 MHz port of 11707A, the signal there is now stable (not wandering). Checked this across it's working range (settings xxx0000000 to xxx9999999 Hz) and the 11707A 20-30 MHz output frequency appears stable. No components have been replaced, and after the synthesizer started showing the wandering problem, the problem persisted until now.

At this point I'm going to put the A15 and A19 boards back into normal position and reattach their shield covers and see if the problem reappears.

Just a quick comment:

I once tried adjusting one of the oscillators on the 8660 on an extender. Once it was placed back in the slot with the cover on, it would not lock. I had to go back and re-adjust the oscillator in the slot with the cover on.

Gary Appel