HP 3310A vintage function generator: bringing it back to perfection
Вставка
- Опубліковано 30 вер 2024
- A run through the adjustment of the vintage HP 3310A oscillator that we demonstrated in a previous video.
Previous video on the HP oscillators demonstrating the HP 3310A:
• Vintage HP Function Ge...
Previous video on the repair of the related HP 3300A oscillator:
• HP 3300A Vintage Funct...
Our sponsors
- PCBWay: fast turn PCBs, www.pcbway.com
- Keysight: test instruments: www.keysight.com
- Samtec: connectors: www.samtec.com
Support the team on Patreon: / curiousmarc
Buy our merch on Teespring: teespring.com/...
Get more info on the companion site: www.curiousmar...
Contact info: www.youtube.co...
"You can do most of this with what you already have" Yes, this is surprisingly possible a lot of the time, even calibrating my 8660 was possible with what I had around. Sometimes you see stuff like sampling scope with 10ps resolution though, which I don't have(yet), but the modern equivalent would be a 100Gs/s+ scope, which is not really in my price range either. And usually it's the instruments you wouldn't expect to need such gear as well.
Now I'm down another rabbit hole :)
You've got so much fine test equipment and I really enjoy the restoration/calibration videos!
Given how spot on everything else was with regards to calibration.. maybe it was the main knob just offset/moved ?
Possibly. I was thinking the same.
Yeah but going through it from start to finish gives you the utmost confidence in your equipment rather than just relying some past mystery.
Laughing out loud because of that final adjustment. That is a very interesting one.
I have trouble believing it was intentionally designed that way, as opposed to kludged into the calibration procedure after the fact with a straight face.
@@frankwales I think I agree.
You should point out that calibration should follow a suitable warm-up period... typically about 1 hour. I suspect that the previous calibration was performed using a poor time base -- maybe an analog scope that swept too fast or slow.
Power supplies on HP gear are rather binary in my experience(very limited compared to yours probably), they're either spot on or faulty, there's no in between. I've had a lot of non HP gear and the rails were always off. I'd love to hear your take on this!
(uninformed/unresearched opinion) They are probably overengineered to the point that if there's overvoltage, a crowbar circuit will insta-kill the supply and if there's undervoltage, it just shuts off and the limits are probably set to
I guess poking a capacitor with your finger makes the adjustment 'digital'...
Regarding what the adjustment does, what's the mechanism behind the change we see when the position of the cap is changed?
I think this would be known as capacitive coupling
Not 100% sure, but I think it couples capacitively to the heatsink of the transistor, which is connected to its collector.
Just don't try it with HV caps in tube circuits :)
@@KeritechElectronics if you’re following along “you’re doing so at your own risk” 🤣
@@Mark_C1 true, it's essential to know what you're doing :)
From my sparkling well-equipped lab in a Big Aerospace Company, I always felt as though I should have taken the reliable crew down in the calibration lab a Christmas gift. Like a good bottle of Scotch. Never did, however. Probably just as well :-)
Thanks! the eBay 3310 and 3312 were just delivered. Guess what I'm doing this weekend :-)
Very satisfying to watch you calibrating this generator. Thank you!
In this world of growing tensions I find here a calm space to watch old gear being loved and made better again - if only the voted-for folk could just understand that it’s better to fix than destroy
“..and it’s : ‘pot’ on….” 😁👍, thank you, love watching this stuff.
You need an hp orchestra with all the lovely sounds and beeps from the lovely hp gear from the golden decades of hp goodness
Did they recommend a specific interval for recalibrating it?
No. They say only if it fails the performance checks (they have a whole long chapter on performance checking), or if you have changed a component.
CuriousMarc attempts to the world's largest symphasizer made entirely of Lab test equipment. 🤣😂🤣 Don't tempt him he'll do it.
Sam Battle has done that already though :)
@@KeritechElectronics Ha ha ha ha. Furbies and gameboys ain't lab equipment (well maybe these days it might be so). But Sam does some excellent work and am looking forward to visiting him this year.
@@MrMaxeemum I mean a hell of a lot of old tube generators. That collection could rival Mr Carlson's collection of Tek scopes :)
I'm working on a vst emulation of this!
This is a very useful video for me, thanks!
Excellent idea! Good luck!
Stay tuned for the next episode? It looks like that instrument will have no trouble staying tuned that long!
Excellent. Thanks for doing this. Glad you stuck your hand in there first.
Watched the last half of this video on my phone and a little XB12 Bluetooth speaker at the bench. The final frequency sweep had the speaker unexpectedly levitate and sprint for the table edge like a rocket.
I really need to get a digital oscilloscope. I don't have need for one super often but when I DO need one not having it can be absolutely maddening.
"This looks pretty straightforward actually"... I think our definitions of straightforward are wildly different. :-)
Great video series, I really enjoyed it.
But, but... Why would you want to use an analog WFG in 2022 when the digital ones are so amazing? Smaller, better, cheaper, faster, way more accurate, no scope needed: just set the numbers. Plus computer control. I started using modern WFGs 15 years ago and never looked back.
I had a 3310A that somebody had attempted to repair, but they ended up zapping several transistors in the power supply, output stage, and the signal source. I managed to repair the power supply and output stage, but it was a pain to work on because all those boards interconnect. The wires started breaking off of their connectors to the front panel controls, and I had a broken trace within one of the circuit boards. Finally had to admit defeat on that one. FYI, the manual on Keysight's site is a poor scan and is missing some parts of the schematic. I found a better scan that's complete if it would be useful to you.
Oooh I think I have the 3310A myself, not sure if it's working though at the moment.
Just purchased one on eBay for $40.00. SCORE!
Did he just say "top board ,bottom board and cyborg"lol .It must be his cool accent
Bending a cap... weird!!! What is mechanically happening here that is causing this to change? Bizarre.
I think it is capacitive coupling between the cap and the collector of the transistor, which is connected to its heatsink.
@@CuriousMarc This is a variation on a "gimmick" capacitor, very common in ham radio gear. Typically though a gimmick is just a piece of wire that you bend closer or farther to another wire or component.
A video of "having your tongue at the right angle".
Sounds like the BBC radiophonic workshop, totally ace.
I'm going on 79 yrs old
Still have my 4262A LCR
METER in my garage been with me 35 yrs or so,started my business with it making transformers, and test equipment. It's a fabulous
Tester. I kept it because
It has so many memory's
Of the people that used it
And when we shipped our
Magnetics out we knew
They were run thru our 4262A
I loved there so test feature
So much I built selt test in to my toroid turns tester 3500
Yup but it needs a good home ?
Got an blue finish 3310A some weeks ago. So your video definitely helps a lot. Appreciated! Thx for taking the friction wheel near to the camera, as that one is missing completely on my generator. Will have to 3d-print something. The frequency pot is working in princple, but shows some on/off behavior. Any hint how to 'repair' this little fully encapsulated guy?
Your video has inspired me to try my first calibration of my HP 3311A Function Generator. I hope I don’t fluff it up.
Excellent, thanks Marc. Had a good chuckle about the 75MV (that is Mega, 5-153) adjustment as printed in the manual. I guess you were using an attenuator 😉
Can you explain the electronic theory behind bending it? Is it changing the inductive field on the transistor?
It's adjusting capacitance. There's a parasitic capacitor with one plate being the transistor's case and the other plate being one side of the capacitor being bent. Bending the capacitor changes the spacing between the plates of this parasitic capacitor, which changes its capacitance.
Things like this were sometimes semi-intentional by the analog gurus. Bob Pease wrote an article about how he discovered a discrete op amp worked only because a few femtofarads of capacitance between its two circuit boards provided positive feedback necessary to make it function. When competitors tried to reverse engineer the design to copy it, they failed because they didn't find that tiny parasitic capacitance the designer hid in plain sight.
“Just for the fun of it, I added a zero.”
I do the same, just because I can.
12:19 mmm, crunchy calibration! :) I like it. Oh, and I love your channel, the test equipment, and apollo equipment, I'm working through it, sir. Thank you for your detail.
Excellent content, Marc, thank you for making this!
I have the B model and now I know I definitely need a solid digital oscilloscope. The only properly working things I have at the moment is a Chinese multimeter and a Chinese 60 MHz signal generator/counter. But I'd like the oscilloscope to be a proper established brand (although I fully expect it to be made in China). Does anybody have any recommendations? My budget is about 600 USD / EUR. Thanks to anyone for any suggestions.
I bought a little 4 channel Rigol for the Computer History Museum restoration team for less than that, and it was very capable, including Fourrier transform for watching the spectra.
Made in China is not a bad thing so long as you know it passed the brands quality standards. I used to work in the Automotive industry and worked all over China for many many car companies (we supplied equipment to them) I once got talking with someone high up in VW and they admitted it was more expensive to manufacture in China just because of the rejects. Brands earn their respect through quality inspection. VW Chinese partners (FAW / SAIC / JAC) used the exact same equipment, the exact same production lines and produced absolute garbage. Do not buy Chinese brands unless you are willing to accept the possibility of a bad product. They are getting better but the ethics are not there.
@@CuriousMarc Thanks for the suggestion, Marc, I'll have a look at some Rigols.
If you like tedious lineups...grab yourself an old Link 110 TV camera... :) And for advanced level masochists, the Link 120 was even 'better'.
Surely an EMI would be more 'fun'!
EMI2001 of course...
@@kiltrash1 It’s a long time ago….but ISTR that you could get a better picture out if the 2001!
Another HP test gear droolfest... 🙂
Australia here too...love Mr Fancy Pants videos
Hi since you like vintage equipment, do you have any datasheet for AWG710B arbitrary waveform gen 4.2GS/s ?? i cant get anything from google :(
its Tektronix
HP should pay you for showing us how to maintain vintage gear like this. This is brilliant.
Must have really made for slow mfg for these. Each needs a lot of technician cal etc
Probably not that bad. Except for tweaking the two caps you could probably do the calibration in about 10 minutes with practice and the right test setup, which HP would have had. Maybe a half hour for the whole calibration, which was likely a lot faster than these came off the production line.
Nice callibration video, but bending the cap? I don't understand here why this works. Can you show in the diagram what you did here?
Capacitance varies with the distance between two conductors. You typically adjust rise time and overshoot by adjusting capacitance in some way. In this case, there is a small capacitance between the outside foil on that capacitor and the heatsink on the collector of the output transistor that is amplifying the square wave. By moving the cap closer or farther from the heatsink you change this capacitance, which might be in a feedback circuit (I haven't looked a the schematic).
@@lwilton Thank you. that makes it clear to me. I'd never realize this could happen. But it sounds very logical.
I just broke mine by accident by putting 550V into the input
No worry. It will just buff out.
@@CuriousMarc Well i got further with it it seems like it blew Q14 (one of the output transistors) and something on the power supply and also the 47Ohm (39ohm in my version) resistor got a big crack through it and turned into a 2MΩ one, i hope that prevented damaging the frequency generator brain board A1.
The 25V (+ and -) rail are both current limiting, without the frequency generator part, so only the final amplfiier as the load.
No transistors seem to have shorted, but i havent verified the diode drop voltages on most.
most passives allign with the parts list except for a few cases of being a later revision.
I hope that the frequency signal shaping board A1 is fine and just the power supply and the amplfiier board died.
@@CuriousMarc Update: figured out what is wrong with it, the signal shaping board A1 is fine, just the output stage got blown, i ordered replacements and we will see if it does work after what i did.
@@janno288 Excellent news! Good luck on your repair!
@@CuriousMarc Thanks a lot!
I would love to watch a detailed explanation for how a diode array creates such a pristine sine wave. Probably like watching paint dry for most but I have no clue how that works as well as I it does.
Paul Falstad has a visualizing circuit simulator with lots of example circuits, one of which is a triangle to sine converter.
@@malloryworlton6359 I was intrigued by its accuracy as well and tried it in LTSpice, which worked nicely as well. It doesn't look as fun as Paul's simulation though.
Спасибо, отличное качество видео и подача материала.
I used to work as an administrator for avionics installation engineers, part of which involved sending gear for calibration. I used to wonder whether equipment was still "in calibration" after being flown to distant jobs in challenging climates.
If it isn't "tortoises all the way down" "calibration all the way down", might be a better option.
Idle curiosity is that pencil a pentel ss475, you need a pentel Kerry to go with the 1970s theme of your vintage hp gear!
The sounds you were making were reminiscent of the soundtrack from the 1956 classic "Forbidden Planet".
In that case they used a musical instrument called a Theremin. Those were very popular in Hollywood for SF movies in the 1950s.
@@lwilton Actually I ended up Googling Forbidden Planet after that. While FP was the first movie to use an all-electronic soundtrack, a theremin was not apparently used for it. Instead it was apparently some sort of home-brew synthesizer. I'm familiar with theremins. I built a small one from a kit some years ago. If I ever wish to induce my gf to leave me, all I'd need to do is play the theremin for an extended period of time. lol.
While have to do mine once I find a power cord