Thanks for all the super helpful responses. I now have a frame coming (thanks again Kurt)! While I have been waiting on pieces needed to reconstruct the RF section, I have been working with the IF section and I have it working pretty well and managed to get all the options enabled (except AY5 - see below). Here are a few things I learned that might help others:

1) The HP 89441A IF section is the same as the 89410A. If you enter service mode (the password is 1125) you can change which one it identifies as.
2) As far as I can tell, all of the options in the 89410A / 89441A IF section are software options except for UFG (4-MB Extended RAM and Additional I/O), AY5 (Precision Frequency Reference), AY9 (Extended Time Capture), and AY7 (second input channel).
3) There is a really nice "hack" to enable all the software options. Just Google for it.
4) You can convert the UFG options (4-MB RAM) to the UTH option (20-MB RAM) by adding a single 16MB RAM. Order Micron MT8D432M-6X from eBay etc. This is a really nice addition that costs about $10 to do so well worth it. You don't even need to change any of the software options - as long as you have UFG installed adding the additional SIMM will automatically cause UTH to show up as installed when the system is powered on.
5) If you have an 89441A system, you won't have option AY5 in the IF section, but that is OK - it is usually just installed in the RF section unless you have option AY4 which deletes the precision reference from the RF section.
6) There is a firmware update available from Keysight that, once installed, allows software options B7A, B73, B79, and 080 to be enabled (you still need to do the options hack to actually turn them on). However, these are all for testing obsolete cell phone standards, so I am not sure how much use this actually is... Get it here:

7) There is a piece of polyurethane foam between the main chassis and display section of the 89410A / 89441A. If you are servicing the unit anyway, I recommend removing this foam and cleaning up the residue as it turns to sticky conductive, corrosive tar. The foam isn't really needed for anything so the safest thing to do is just remove it. Lots of acetone will be needed to clean up the mess!


OK, with that contribution, I have another ask for the group. I am apparently missing the little metal clips that attach the front frames of the RF and IF sections together. The locking feet that attach the rear frames together came with the unit (and I have managed to use HeliCoils to repair the torn-out holes in the IF section frame where they mount), but the clips for the front were not included for some reason. The photo below shows the parts I am looking for. It looks like I need 4x of them along with the 8x metric screws needed to mount them. These same clips seem to have been used on the 8566 / 8568 spectrum analyzers and also appear to have been included in the locking feet kit (5062-3999) but I haven't managed to find an HP part number for the clips themselves.


The last piece I am looking for, and this is purely cosmetic, so I guess it isn't the most important, is the plastic trim for the top. I would be looking for the grey (not the older olive) color one. This was used on a lot of gear, but they are commonly missing too. I figure it is worth a try...



Thanks again all for the help.

-Matthew D'Asaro

Hi, I recently bought? HP 8753C with no option.? I would like to make a full upgrade with options? 002,010 and 006. (010) I have only an 85046A test set however I would like to use swept oscillator HP 8350B so hopefully, that will not caused any issue with soft or with the test set to future.
Anyway, I got the codes for HP 8753C but unfortunately, I struggling to found proper procedures on how to install these upgrades. An available method I found is only for 8753E and higher thus I would like to ask you if you can recommend where to find it or if its the same procedure as for 8753E. Thanks

If you have a spare 2 hours to watch this video, it may be helpful...or at least interesting.? However it does not address your exact situation.

Thank you Matt! This was indeed the missing piece of the puzzle in my head. I wondered if the mixer might do that and was hoping that the 8709A was "smart enough" to find the correct signal.
I have the manuals for both now - the mixer manual doesn't say too much and the 8709A manual goes deep into theory of operation. Good bed time reading!

regards,
Mark

Hi folks,
it's been quite a while since I started this thread. In the meantime I am retired, which to my surprise did not give me enough spare time to dig into the 3336A's weird flickering Display Problem in detail.
Fortunately due to the Corona Virus lockdown I had the chance last week to get the unit on my bench and check again for the root cause. I was fixing the Power Suppy already when? was trying to fix it several years ago.
To make a long stor short: The Problem turned out ot be a defective 2112 RAM chip (256 x 4Bit ;-)) A6U12 on the A6 board. One of the data output bits was very weak and most of the time unable to drive the data bus.
Luckily I had some 2114 chips (1k x 4Bit) from my first Rockwell AIM-65 singel board computer still lying in the junkbox. I took two of them and put them in with the help of simple 16pin to 18pin socket adapters.
The unit sprang immediately back to full operation, I also did the 3336A to 3336C conversion and a complete realignment as per service manual. Now it behaves like new!

vy 73 de Reinhard
OE3RDW in AT

Hi Mark,

The thing you might be missing is that the HP 934A is not just a regular mixer, it is a harmonic mixer.? Internal to the HP 934A, the reference input signal feeds a step recovery diode (SRD) circuit, which generates a comb of harmonic frequencies: 1 x Fref, 2 x Fref, 3 x Fref, 4 x Fref, and so on.? This comb of frequencies then mixes with the other input (i.e., the auxiliary output signal from your HP 83590A) in a regular mixer section, which is also inside the HP 934A.? The resulting harmonic mixer output, that arrives at your HP 8709A Synchronizer, is full of the sum and difference frequencies, of the pure sine wave from your 83590A and all the multiples of the reference frequency.? From all of these frequencies, an input filter on the HP 8709A Synchronizer only allows a 20 MHz signal to pass through -- It blocks all the other frequencies that will be present.



For example:
----------------
If your HP 8640A reference generator is tuned to 995 MHz, the 934A internal SRD device will generate a spectral comb of harmonic frequencies: 995 MHz, 1,990 MHz, 2,985 MHz, 3,980 MHz, 4,975 MHz, 5,970 MHz, 6,965 MHz, 7,960 MHz, and more.? As the frequencies get even higher, they will tend to roll-off in amplitude, until they finally become unusable.? So, even though the 8640A only goes up to 1024 MHz, you now have discrete frequencies through the microwave band, which are available to drive the mixer.

Now, assume that the 83590A is tuned to 4,000 MHz.? Your 995 MHz reference signal and all of its harmonics will be added and subtracted from this 4,000 MHz signal.? For simplicity, we will just look at the fundamental and first few harmonics.

reference fundamental frequency (N2 = 1, in the manual's equation)
4,000 MHz - 995 = 3,005 MHz.
4,000 MHz + 995 = 4,995 MHz.

reference 2nd harmonic (N2 = 2, in the manual's equation)
4,000 MHz - 1,990 = 2,010 MHz.
4,000 MHz + 1,990 = 5,990 MHz.

reference 3rd harmonic (N2 = 3, in the manual's equation)
4,000 MHz - 2,985 = 1,015 MHz.
4,000 MHz + 2,985 = 6,985 MHz.

reference 4th harmonic (N2 = 4, in the manual's equation)
4,000 MHz - 3,980 = 20 MHz.
4,000 MHz + 3,980 = 7,980 MHz.

The HP 934A mixer output, that is created by the subtraction of the 4,000 MHz signal and the 4th harmonic of the reference signal, is 20 MHz.? It is this 20 MHz signal that will drive the HP 8709A Synchronizer.? The synchronizer will filter out the other frequencies and not allow them to pass to its internal circuitry.

In the above example, you have just used your HP 8640A to synchronize a 4,000 MHz signal.? This is despite the fact that the 8640A only goes up to 1024 MHz.

As a further exercise: If we plug in the numbers to the (corrected) equation from the 83590A manual, we have:

(HP83590A aux output frequency - 20 MHz) / N2 = the required HP8640A reference frequency
(4,000 MHz - 20 MHz) / 4 = the required HP8640A reference frequency
3,980 MHz / 4 = the required HP8640A reference frequency
995 MHz = the required HP8640A reference frequency

The auxiliary RF output of the HP83590A only covers 2 through 7 GHz.? But, when this auxiliary RF output signal is phase-locked, the main RF output of the HP83590A will also be phase-locked.? The main output covers the full range of 2 through 20 GHz, which is just a multiple of the 2 through 7 GHz auxiliary output range.? The phase-lock example in the HP83590A manual calls this multiple: "N1", where "N1" can be equal to 1,2 or 3.

Sean,

It works again -- congratulations !!!!!

This is great news, indeed. Another good news is that now, you have a good knowledge of how this instrument works !
You may want to have a close look at surrounding soldered joints, especially the R's and C's in the Wien bridge and relay K1, because one of them was the culprit when the oscillator was running at 12 kHz. A defective or mis-soldered Q1 couldn't change the frequency.
The next step is calibration... of course very few of us are lucky enough to have all the required standard capacitors and inductors. But if you're willing to spend some time on the project, there's a number of things you can do to calibrate capacitors and inductors from your spare parts bin. If you'd like to share some ideas on this, just let me (let us) know.
And if you want to replace Q1, I have a few New Old Stock 1855-0082s (not a "close equivalent", they're the real thing from HP!) on hand. Just let me know.

Again, congratulations for a job well done. You just saved another HP instrument from the scrap heap !!!!

Joel

Greetings, a while ago 5-10 years ago maybe, Fluke offered free silicone test leads and there were some posts about this on Tekscopes and/or here on HP group. I know a bunch of ya'll got some. How are they? I got two pairs somehow, though I just ordered one pair. They all have intermittent connection from cable to probe end. I tend to treat all probes cables nicely as if they were scope probes, yet intermittent

My favorite sharp probes I have made from red and black pens with sewing needles extending maybe a quarter inch. Wonderfull for smd, getting into the wire end of connectors, even piercing wire insulation tests. Often the probe will touch or fall on my leg and giving a little prickle. A little pain often but no pain no gain as they say and I will keep using them. I call them my prickle probes.

Dave

开云体育

I’ve got ten according to my inventory(location 1989). Drop me an e-mail s.hanselman@...

I would take either one of you up on that...probably a good idea to have at least one spare on hand in case the one in circuit actually does have a mechanical fault.

Sean

Hi Don,

Thanks for the interest and feedback. I understand the intent but maybe I'm missing something here. Let's say I want to lock at 10 GHz. Wouldn't I need a reference that's 10GHz - 20 MHz?
If I have a sig gen that can do that, then I do not need to lock the 8350. (My goal is to have a STABLE source up to 20+ GHz) Maybe I can use harmonics somehow? Is that what you are saying?

Since my goal is to have stable 20+ GHz source, perhaps dump the 8350 and get a (high price) RF synth?

At any rate, it was fun restoring the plug ins (but they are a pain to work on!) and as always, learned a lot about this instrument. It may still have some uses in my lab. We'll see.

Best regards,
Mark

Also 8350B system AUX out is typically +8dBm at the 8350B tune freq.
Don Bitters

开云体育

My guess is that Q12 is shorted Collector to Base and CR15 is what is pulling TP4 down to -18V and caused R14 to overheat. Q12 is listed as an HP 1854-0087 (75MHz ?360mW NPN).

OK - You win, but the daisy chain cables are expensive. I'b be tempted to get a long 8502 - test set cable so you could "remote" the test set, which would make exchanging test sets easier.

If you don't already know about them, these guys can make just about any cable you need - not cheapo, but theey do good work -->


Have fun NOT watching TV :-)

Cheers!


Quoting "Reginald Beardsley via groups.io" <pulaskite@...>:

If you need that FET, I can send you one...

Dan in Chandler, AZ