"As Bill and Dave intended". "No resonant cavities were harmed during the filming" Both absolute gold. I have HP equipment dating as far back as the 400C voltmeter. Love seeing your restorations / repairs.

The "Do not tamper" warning makes total sense for the time period. HP had to carefully calibrate the units, and to assume that even government labs having the necessary equipment to carefully retune them would have been foolish. But I am sure they were disassembled and cleaned any way, just that HP took no responsibility for their accuracy after.

With how HP or HPE operate today it's refreshing to see a piece of their hardware you can use without 30 licenses attached to it costing you an arm and a leg.

The "perspex" material is actually called "Rexolite" and is more like solid polystyrene. Relatively low dielectric constant (usually around 2) and very low loss (tan-delta). It also has very low moisture absorbtion, commonly used in microwave applications.

I'm on the other side of the world (of CA that is) and Marc always manages to post his latest video during my lunch brake. Yay!!

I could not help to notice, as a tech (retired), that many pieces of tech equipment you've shown I used on telecom equipment. Brings back memories,....

Every time I watch your videos I instantly go to eBay even though I have modern test equipment these videos make me want old tech

3:30 I am not an EE, but I do think I "get something" out of these lessons on RF and even just basic Physics Eletro-magnetism fundamentals. 👍We don't need no teardowns of Apple Vision Pro.! No way man. Gimme a teardown of this sweet ol' analog HP 5245L freq. counter!

I think that heterodyning is my favourite RF trick, and I've never seen a more beautiful implementation than these plug-in modules. Very cool!

It's a very satisfying process taking a mechanism apart, cleaning it, lubricating it and then putting it back together again. I always marvel at how something which didn't work properly or at all now functions smoothly, everything rotating and clicking. It's like magic. Great job, well done. Thanks for the video, Marc!

I absolutely loved this video. I am always so impressed at the technology of that period. And especially how both HP and Collins Radio implemented it so skillfully. Thank you very much for such a well done video.

Can not imagine what it took to manufacture and calibrate in the 60s though I often hear that it can be good if not better than today's workmanship. What I do know is that aerospace funding had a massive boost in the late 1950s and early 60s. I wonder why lol. Great video as always. Off topic, I have a 10590A which is not in use. Planned to get one of these oscillators to test it in. Came with my 5345A that runs a 5355A compatible with its 5356D attached. What I love about the 5345A isn't the fact that I can use my 11970W mixer to it but rather that it actually has someone's name (I'm assuming they did QC) inked on the display module when I repaired it. Simple faulty SCR on the a power module but decided to fully tear it down to rid of the dust bunnies. I took care not to remove any markings of any kind. These instruments were evidently made with care after all. To see the names and markings adds personality imho.

The only thing wrong with these episodes is there is no mystery. Marc ALWAYS figures it out and fixes the thing. Fabulous!

A cavity search always makes for an interesting KZfaq video! 🤪😅👍

Yay! I have NO idea what any of these instruments do but I am fascinated by the video documentation of their repair!

[slaps top of retro HP equipment] 'this baby can fit so much physics'

Very cool repair. HP sure knew what they were doing.

Wonderful RF engineering where shape and form are function at these frequencies.

Ah yes, the SRD. The tool of choice for generating Gigahertz frequencies, in the 1960s and for at least two or three more decades. I have a rubidium clock source made by Datum Efratom in the late 1990s and the synthesizer generating the 6.8-odd GHz representing the hyperfine transition that is the atomic reference for the output frequency is basically using the same principle, a SRD comb generator and a resonant cavity to pick out the correct harmonic.

These levels of RF just boggle the mind. When you start dealing with 'cans' as coaxial lines, it just amazes me. And that you were able to disassemble and repair this is amazing. (not to mention you just 'happen to have' the signal generators to test this stuff :) )

I used one of these in the 1980s in a HF transmitting station and used/rebuilt tunable cavity resonators on 100-500MHz landmobile kit during the same period (the Q on those cavities was so high that their bandwidth was less than 10kHz) Reassembly of the cavity is highly sensitive to electrical continuity of all mating surfaces - there are large RF currents flowing on the internal cavity surfaces around the top end - right where your cap is. It matters more when hanging these after transmitters (heat losses) but it hugely affects Q if there is corrosion or poor conductivity across the surfaces. Those tarnished silver faces need cleaning (immersion in saltwater along with an aluminium strip will work best as it's a non-abrasive method - clean & dry thoroughly afterwards) The robust design and warning are because these were intended to be used in military environments and army/airforce bois have a tendency to be rough on their equipment. Reassembly with the dial markings being correctly aligned is "difficult" (one of my ham-fisted cow-orkers pulled our one apart. It never worked properly afterwards)

I loved working at HP, and I enjoy watching these videos almost as much! Simply wonderful.

RF Magic. Thank you for this beautiful Teardown.

I think you and Ken should try and design a K-band plugin- 18-26.5GHz, using some more modern components.

What a great job you are doing Marc! You explain what you do and how the equipment/apparatus you work with in an excellent way. I get very excited every time a new @CuriousMarc video is published. Why this channel doesn't have even more subscribers is a mystery to me. Thank you so much!

“CAUTION: DO NOT ATTEMPT DISASSEMBLY OF CAVITY" - challenge accepted. 🥸👍🏻

Ah yes, the black arts of RF and resonate cavities! Another excellent repair Marc! *(...and you didn't even have to break out the Smith charts and protractors!)

Later, frequency counters were equipped with a YIG-tuned filter that filters the incoming signal instead of the local oscillator (LO). Then, the filtered signal is sampled/mixed with a pulse of approximately several hundred megahertz. Also, manual power metering is done using the microcontroller to search for signals within the 1-18 GHz range. Very good video.

Could you have used a lathe and dial test indicator to true the female part and then put the dial test indicator to achieve the same concentricity? All in all it's great that it worked out for you.

Fantastic! RF is a magic I will never fully comprehend. And it blows my mind even more they were able to master it to that level almost 70 years ago.

you just saved me a million hrs figuring out how to repair mine, just scored the same counter 5245L and the same 500MHz plugin 5253B mine is also a bit stuck, I was going to figure it all out and explain it make a video about it, but now no need to add more bytes to the world of internet, since this video explain it all :-) the funny thing is the 5245L main counter unit exist in several versions looking a bit different on the outside, now i got both versions, I also got one with 5267A plugin a time interval unit, that one is on my YT in case you like to see it.

Did I just see something impossible proven to be possible? 😮 The accuracy of this 60’s magic is hard to believe. I’ll watch it a couple more times. And a warning to the other stuck plug-ins: Resistance is futile !!

OMG I don't understand much of anything about all the technical stuff yet am hooked on this channel and all the stuff they fix 😁😁😁

As someone who arguably watches too many automotive repair channels - next time you are faced with gears coated in old grease and stuck screws, try recruiting your favorite penetrating oil (or at least WD40) and a quick trip to the oven (or perhaps just the outside, at this time of year in California). I imagine that should make disassembly a whole lot less stressful in such cases. BTW, instead of just generating a frequency comb, you can also use such a pulse train to feed a sampling mixer that can than serve as the phase detector in a PLL. This way you can lock a continuously variable oscillator to a fixed frequency grid, you just need to limit the PLL VCO pulling range so that it can only do fine-tuning. This concept even briefly made an appearance in the consumer space by way of a few high-end FM tuners of the late '70s / early '80s. The point was combining the good frontend selectivity, linearity and low phase noise of a mechanically-tuned affair (not to mention lovely knob feel) with the accuracy and lack of drift of a PLL job. These concepts became redundant as people figured out how to make lower-noise PLL VCOs and gain distribution improved. The first instances of oscillator grid-locking I could find even predate these HP units. The Siemens E311 communications receiver (1960) did this for its 1st LO, using funky electrically permeability-tuned inductors for fine-tuning no less - the varicap diode had yet to be invented, so tweaking inductance via core saturation was the next best thing. I presume HP went with the "filtered comb" approach in order to keep size and complexity of this compact module down, alongside the large relative tuning range required.

Another 30 mins of fascination thank you!

As always a pleasure 😀

I’m always amazed by these fix ‘em videos. 👍 Plus, I feel like I’m learning something. Merci, Marc!

Wonderful video of both the breakdown and explanation of function. Well done. I appreciate your efforts.

Amazing technology from the year I was born! Great work, Marc, on restoring this masterpiece.

As always wonderfully explanation thank you

EIGHTEEN gigahertz? What kind of black magic grey beard wizardry did they use to build that in the 60's!? I bet it even involved virgin sacrifices.

Wonderful stuff!

Such beautiful engineering. A work of art.

Bloody amazing.

Really interesting! Learned alot again! Thnx Guys!

It's amazing how clever and advanced engineers in the 60s already were! With all this witchcraft and rocket science they could have built spacecraft and flown to the moon... Oh, wait - they did... 😂😊

Thanks, awesome. Luckily my 5253B works, and thanks to your video I don't need to take it apart to see what Bill and Dave are hiding from me.

RF gadget with a boiler in the middle of it! LMAO! Good one, Marc! Oooo, transmission line calculations... break out the Smith Charts!

Looking forward to the 8663 repair...love these things. I have 7 in the lab.

- Do not attempt to open! - Open or do not open, there is no attempt! We open!

Those nixie tube displays have such an awesome look.

My HP-8640B has one of these adjustable cavities. Except… it has plastic gears in the knobs and they haven’t exactly aged well. Great video as always!

Marc I'd watch you fix a toaster

Pretty Cool

I have one of these and also the plug in too. Mine was working last time I checked it. Must get it out and test it.

I wanted to call it magical first but actually those engineers were just geniuses.

Looking at the characteristics of step recovery diodes, I have to wonder how this effect was even noticed in 1960. You would have to be looking at the switching characteristics of diodes with equipment that had a very high frequency response.

"This set is equipped with a safety interlock. Do not attempt to defeat its purpose." I remember ignoring that all the time!

HP came out of a shed. And obviously, there were a lot of very sharp tools in that shed.

I thought for sure that disassembly would be a one way journey, but no, how could i doubt your skills! Splendid and fascinating work and a great video. Thank you

Superb series! Takes me back to an earlier life servicing equipment like this. Incidentally, I was watching a North Korea documentary and spotted an HP 5245 on the USS Pueblo (4:52 mark in this video: W_Z1oSINw-M) but it's missing the plugin module, nixies and knobs in other photos of it.

Wow! And amazing. I must say I get nervous and anxious when you take these marvelous instruments apart. We can only admire and be humbled by the engineering genius it took to deal with these high frequencies 60 years ago. Wow!

I suppose you could make a cone shim so the taper can seat on the bore, and that should locate it accurately before the tightening of the locking plate. If the gap is under 0.002" a paper cone might do.

Tnx,super hp Warsaw Poland.

The black magic of _The Cavity_ would make for a good legendary item in Diablo IV ...

I have one question left unanswered : What is the purpose of rotating the diode and the antenna assemblies ?

RF black magic, but with gears, and hand-cranked.

I remember having to deal with the same solidified grease while servicing HP 200 signal generators in the 1980s

HP was awesome in the day.

Looking at the disassembly and how you did it you must have owned some boxes of Meccano as a child 😂

I've repaired a good few of these but never dared go inside the cavity. Now I'm wondering if the 8640A/B signal generator cavity looks the same inside (with the addition of the little fine tuning flag)... At least these don't have plastic gears, wait for the 5254A/B/C; when they become sticky they strip the gear teeth. The gear with the dial attached has 61 teeth which is a most annoying prime number when it comes to cutting new ones. I seem to remember I had to get out the oxy-acetylene cutting torch and make a new diving plate!

Kind of have no choice but to disassemble the cavity to fix mechanical issues. You will want to take lots of pictures and keep a fine tip sharpie handy 😉

that's magic

wow, 18gHz (!!) fun facts: at 18gHz the time between 2 steps is only 56 pikoseconds. in this short time, light travels only 1.6 centimeters (or 0.66 inch)

Great video. How do the rf wizards account for the stub resonator properties also changing significantly over the operating range? At 50 MHz you’re quite far from l/4. Does the q-factor drop for these frequencies?

“How do they get a counter to work on higher frequencies just by plugging in a gadget?” Me, who has seen lots of SDR tutorials on frequency division or multiplication circuits to get the entire spectrum within the limited range of the hardware: oh I know this for once! Edit: well, I was still surprised by the _implementation,_ if not the theory! It's good to show how cost reduction from earlier optimal designs isn't a recent phenomenon at all. Thankfully the drawback didn't affect end use in this case, but it always starts by making repairs more perilous! I could easily imagine someone taking off those final screws back in the day and getting in trouble, especially if they'd seen earlier units with centring guides and weren't taking their time double-checking everything.

The process of taking 2 frequencies and making 2 new frequencies, sum and difference is called HETERODYNING.

If you form extremely short pulses with that SRB diode of the 10 MHz: - are the short pulses like a gate or do they add high frequency portions to the 10 MHz ? - do you only get multiples (harmonics) of the 10 MHz with the short pulses and no other frequencies? - if yes: on digital signals: do (extremely) sharp/quick signal flanks "only" create multiples / harmonics of the frequencies that are already in the base signal?

"FOR THE LOVE OF GOD AND ALL THAT IS HOLY IN THE UNIVERSE, PLEASE DO NOT TRY TO SERVICE THIS UNIT." -handwritten sign on a big beige cabinet in the IT department of my college, c. 1994. It wasn't my department so I didn't ask.

Nice 👍 21:04

Marc, Regarding fabricating an extender: I went quickly through all the relevant plug-in schematics I have (nine or ten), and I could only find but 10 (out of 50) unused pins on the connector. The TI and Prescaler plug-ins used quite a few pins. Have you ever seen a universal HP extender? I wonder where HP put the coaxial leads? Thanks for the video.

That (then) greenish grease is also horrible in 1930s--1960s cameras, the focusing thread is almost always stuck and you have to disassemble it using gentle heat, clean it, reassemble with fresh grease and then you have to find the infinity stop again ..

These must have been SO expensive back in the day.

Thank You Mark, great video as usual, one question Sir if you please, why the mixer assembly with the Comb gen. assembly rotates while we tune the cavity?

the black art of rf voodoo

As always a great video and I admire your RF knowledge and the way you film these adventures. They are very much enjoyable and educational as well. Thank you Marc!

Does it have an ad block plugin at 666Mhz?

Does anybody know why the drive and readout antennas are made to rotate as the cavity is tuned? That's really interesting. I wonder how accurate that part has to be. I imagine it is something like changing the length from the antenna to the nearest wall, if you draw a line perpendicular to the antennas ?

Not very relevant to this video, but: yesterday I met a 12-year-old girl named Nixie. I asked her if that were her full name or a shortened form of something else, and she said no, that's her full name. I said I had never met anyone named Nixie, but that there was a vacuum tube with that name, and she said "I know." Somebody out there is raising their daughter right...

The yellow glass block looks exactly like zerodur ceramic.

Is this down-conversion similar to wadley loop?

* I just picked up an HP 3439A Voltmeter, then ordered the original service manual! I'll see if I can restore the beast...

I just recently acquired a 5245L made in 1967. Works great, but some of the plug-ins I have could use some adjustments. Would you have the wiring diagram for your custom 50 pin plug-in extension cable? I’ve tried a few old 50 pin SCSI extension cables and while they physically fit, they do not pass the signals correctly. I must be missing something critical.

Wizardry on all levels

Why is there a video amplifier in there? Surely video amplifier is some sort of terminology I don’t understand.

Amazing witchcraft!

can't wait to see the other plugins insights videos 😊

Could you get away with soaking the whole bottom in IPA or something? to get it somewhat unstuck

@11:40 - In all fairness to HP and the vintage electronics, I would be averse to having a cavity search too. Just saying.