HTML-код
- Опубликовано: 26 июн 2024
- In this episode Shahriar dives into the world of high-voltage circuit design. In order to appreciate the engineering involved in the design of a 40kV amplifier, various design techniques for power supplies and their applications are presented. This tutorial is organized as follows:
00:00 - Introductions
00:29 - Amplifier design specifications and challenges
03:18 - Typical high-voltage generation using transformers and its limitations
06:41 - Open-loop transformer high-voltage power supply for discharge tube use-cases
10:18 - Inaccurate HV power supply making precision isolation measurements with Fluke 1587 FC
12:39 - Stanford Research PS350 high-voltage DC power supply design and measurements
21:44 - Vitrek V74 HiPot safety tester design, applications and example breakdown measurement
27:01 - Architecture of a high-voltage amplifier circuit, circuit diagram and theory of operation
32:43 - Teardown & analysis of the Trek Model 20/20B +/- 20kV amplifier instrument
42:02 - Linearity and waveform measurements of the high-voltage amplifier instrument
46:26 - Making ultra-high voltage time-domain measurements using a capacitive divider
48:45 - Discharge tube response as a function of AC frequency & high-power sparks
50:12 - Concluding remarks
Please consider supporting The Signal Path:
www.TheSignalPath.com/Donate
/ thesignalpath
Other links:
www.TheSignalPath.com
/ thesignalpath 
Наука
Cool! Yeah that 10M input droping would be the MOV(s) ever so slightly starting to leak. I'm a bit surprised that thing didn't have potted HV sections.
Don't ever apologize for a long video ever again, young man. More TSP more better!
He's probably one of the few RUclipsrs who can convince the audience to watch an hour long video without skipping.
And what a story this one was. Best ever!
Woohoo, this is a long one, awesome :) Missed opportunity at the intro to mention 25 kg weight of this "opamp" and need of a forklift to install it :) With amplifier like this you very quickly learn how to NOT make mistakes and triple check entered values, unless there are sacrificial oscilloscopes and probes! BTW interesting note as on CPEM2024 conference there is research presented for HV metrology with 4000 kV (yes, 4 Megavolts) for lightning impulse simulator generator. TSP now need one of these to lightning-test some RF equipment ;)
Previous job was in the electronics manufacturing industry. Our Scanning Electron Microscope (SEM) field service engineer covered the SEMs and TEMs at oak ridge national labs (a few hours away from where I was). Some of the high end models have to maintain close to 1 MV with micro volt stability. He said servicing them was fun. Everything in them developed a permanent charge so dust would settle on them quickly. Once you cleaned it it was a race to get it assembled and if a spec of dust made it on some parts you’d quickly find out when it arced out.
A very wise friend and colleague of of mine used to say "If you're not afraid, you don't understand!" It was in an entirely different context, but it surely applies quite nicely here. My existing healthy respect for "high" voltages certainly makes me very afraid and anxious just watching this! Thanks for taking the time working up to the actual device testing at the end.
That is a terrifying piece of equipment!
The usual GAS from looking at a TSP video was not kicking in this time, I would not want to go near that beast.
Super interesting video as usual!
Amazing amplifier, 50 minutes that feel like 10, as always spectacular detail, clear and precise explanations, i love every second, thank you!.
High-voltage design has for so long been a foggy mystery to me, and there is so little information on it out there. Thank you so much for making this one! I learned a ton!
Excellent video, thank you!
I worked with design of HV switchmode power supplies for electrostatic precipitators. 35kW and 100kV. Nasty stuff capable of creating x-rays, I was really scared when testing them.
That was ridiculously exciting and fascinating at the same time. I built high voltage circuits out of old TVs when I was 14 or so, just to draw arcs across things - glass becomes conductive when it melts, who jbew? How I survived... who knows. Amazing to think that nearly every household had a a 20 kV+ generator in their living rooms in the days pf colour CRT TVs, and before HV semiconductor rectifiers the thermionic diodes had to be shielded to limit X-ray exposure. And then Granma puts a plant in a pot on top of the set and waters it. Wild.
I understand exactly what you are talking about. But obviously I somehow survived my youth in the 70s 😂
BTW, even my latest „line contact“ (240V) is now 10y back. I now have to make sure my (now 3yo) grandson also learns about the power of electricity.
I grew up around my grandpa's electronics workbench and i learned electronics by helping him fix a lot of CRT TV sets (tube and solid state). Therefore i developed a very healthy respect for high voltage. That thing is scary! Thank you for showing it to us.
Scary, well slightly. You should try older style radar systems, nasty nasty HV pulse transformers driving the mag. 50KV to 100KV or so off load which happened when the mag went open circuit.. Very impressive sparks when things went wrong., but that's a long story or two.
@@andye2005 well, actually my grandpa used to be a radio and radar electronics tech in the Navy in from the '60s until the '90s, so I got to hear those stories also. 🙂
Best channel on RUclips, imo. Thank you for such a thoughtful and informative vid. Your tutorials never disappoint!
Standing arc in flash light tube...beautifull 😍
Very informative lesson with most important details. Great job!
Thanks for explaining and showing this unique high voltage amp, super cool device! 21:45 looks like Pooch has been doing some hipot safety testing already :)
I really liked this! I’m working my way up to understanding 5% of what you’re talking about, im super glad this channel exists and is active
Thanks for putting out a new long tutorial video! Excellent subject to learn during the holiday time!
A superb video - a very rare insight into high voltage electronics. I can’t imagine what the engineering challenges are like in +\-800kV DC power line distribution systems they’re using in China where they use inverters to convert back to AC.
I think your static arc may be related to the acoustic cavity resonance of the tube, judging from the behaviour of only stopping at very precise frequencies.
If anyone knows more than me about the acoustics of plasma arcs, do chime in!
Thank you for all the effort you put in these videos!
Fantastic, one hour of great science thanks !!!.......cheers.
ممنون شهریار
Huge huge amount of useful and hard to obtain information. Thanks
Very informative video. I learned a lot. The design of the 40kV amplifier is just so clever.
My palms got a little sweaty during the demos, thou.
I will second what Ziferten said, the time flew by as you delved into this very rarely addressed dangerous and challenging design of High-Voltage amplification.
I hope Pooch will appreciate the amount of attention he got in this video.
He had to be locked away due to the exposed lines during the experiments.
@@Thesignalpath But you compensated it with his name on the device panel.
This is one of the more terrifying instruments I've ever seen. And to think that you're not even pushing it to its limits. Be careful!
Just the amplifier I was looking for my upcoming Karaoke night!
Even the equipment is off, I still get nervous when I see you using that metal pointer on circuits 😂
Finally some power/high voltage related too from signal path😭
Another brilliant and exciting video, about a subject of which I know nothing. However I think that I will stick to +/- 12V
Great engineer, great explainer !
Loved it. All that information and then destruction at the end. Reminds me of the Photonic Induction channel where I think you got that one image of the microwave xfmr. You find the coolest things to show here!
Absolutely superb. I’ve always loved high voltage power supplies and I’ve designed a few, but this took my info to a new level.
We used such amplifiers in the accelerator lab for an electric beam deflector. Back in the 7ties we worked with tubes.
We use those HV bench multis for adjusting and calibrating hipot testers.
We use a bunch of Vitrek power supplies at work for testing high voltage capacitors. We even have a few 957i(s) that go up to 15kV.
Glad you and that Tektronx scope survived.
I got the same feeling watching the end of this video as watching a high wire act not using a safety net !
Fascinating. A beautiful instrument. I would love to see the schematic for the stacked high voltage stages to see how the biasing is achieved and protecting the devices from overvoltage. I have an equally lethal vintage power supply. 17kV DC from a mains transformer and what looks like stacked selenium rectifiers then a large "Visconol" smoothing capacitor. It has no problem pumping out 30mA on a short circuit. To be treated with great respect.
Where I got my education, we did some HV lab tests and some more were just shown demonstrations. The “volt meter” was a couple of polished 2 ft diameter metal spheres as a spark gap. besides the air gap, we needed air prssure and humidity data for correction of our result. I still recall a couple of memorable equipment in the lab. One was a MECHANICAL rectifier, an about 6 ft diameter disk rotated by a synchronous motor. There were contacts at the disk perimeter, and other contacts near the disk, outside, maybe leaving a 1/8” gap. If I recall, we got about 200 kV DC with it. Mighty sparking in operation! Another device was a little more “modern” device, hybrid Greinacher-Villard voltage multiplier that had a number of series connected ceramic capacitors and matching number of spark gaps. Finally, in actually in a different lab I saw a voltage divider built with a 20 or 30 ft long polyethylene tube, filled with ion exchange purified water. I was told it could be used to over 500 kV as the upper resistor of the voltage divider.
At work we have a soft start that is feeding a 3400HP motor powered by 4160Vac. The SCR stack has much the same sort of arangement with individual boards driven optically by the trigger board. This thing is on a whole other level though. Easy to understand why the amp cost $70,000.00 USD. It is clearly not a trivial mater to engineer and make such an instrument. Great video as usual!
What a wonderful video, Shahriar. Thank you! I am sure there are many unusual electrical and electronic 'features' that can only be explored with HV and AC. This amplifier opens up a whole avenue for interesting studies and experiments. I'll mention a few that I'd love to see and that you may consider interesting: corona discharge, dielectric barrier discharge (applications in ozone generation and plasma treatment), inductive heating, shielding effectiveness. Oh, and I've noticed that the plastic tip or your hook clip started to melt during the last experiment. There is probably a short thin vertical "gap" in them. I did exactly the same thing by testing high currents through a pencil lead. All for science :)
Very interesting, thanks for another great video!
Thank you! I really enjoyed the structure of the lesson, going through successive definitions of HV and the design of a supply at each level.
It would be very cool to see this directly driving an electrostatic loudspeaker.
You can start your own radio station with this thing.
Just awesome engineering, and an awesome tutorial!
Gash, our company, my supervisor, and me, designed several variants of these, for a military classified project, in 1971. Later, these same techniques were used all over the place, for CRT drivers. I worked for IBM using direct drive CRTs, for video displays in 1966, where a video pen was used for drawing on the screen. Later we designed some nuclear weapon sensing and recording instrumentation with a slew-rate of about 300V-per-microsecond. By the 1990s we were up to about 1000V-per-nanosecond. Newer laser technology allows femto-second risetimes, or less.
This video is awesome! Thank you.
That's one scary piece of gear. At those voltage levels it would not take much leakage current to wipe out a bunch of test equipment connected or not. We had someone bring a Van de Graaff generator into the lab, wiped out several scopes sitting unconnected on an adjacent bench.
Other than some conformance testing I wonder what the primary use for those voltages, some kind of particle physics research. In the HV theme would love to see a schematic breakdown of a 1MEV DC to AC power transmission line converter.
Great video.
Great video, thanks for making it!
Great video, thank you.
Probably one of your coolest videos
Fantastic, I only wish we could see a teardown of one of these units.
Very cool video. Thanks for posting this.
Extremely cool! I'm working with a particular TREK-built power supply for a particular application, and they are very neat and tidy inside, my eyes lit up when you showed the back of the unit. My supply is only a mere +/-2kv, but was still very impressed with the construction, very similar to this. They got the entire PSU into a 2U rack unit. Awesome!
Didn't mention that mine is six phase output tho ;)
i love HV,. u got me at high voltage,.lol, nice vide,. great information!!THX
Just wow!
This is a nice episode, thank you. Please do more like this
I've always had this hankering to build a high power RF amplifier with vacuum tubes. I have all the parts including the super high voltage transformer. For some reason building the thing never floats to the top of the project priority list. This video reminds me why that is. I can get the full legal 1.5KW with RF mosfets at 65V so dabbling with 1000's volts and giant glowing glass tubes makes my tummy wobbly. I like life.
This is absolutely amazing！！！
That was a Top Video, thank you! Cool devices, good explanation, good video and editing choices
Really interesting video.
fantastic video many thanks...
BADASS!
i really liked it. The only thing which would make me nervous being in your position is the pointing tool you used. Any plastic is better than your metal with a sharp tip.
We see the wire “repair” on your videos then we click view! 😂😂😂
This was such an interesting video
Barely see any lab setting high voltage on youtube
what a beast of a machine
Wow - we where looking for such a device some time ago and I remember that the cost is also astonishing!
With those little calibration tubes I found a small HV dc power supply on amazon and picked off before the rectification and it drive the lamps nicely off a battery which makes it a lot more portable for calibrating spectrometers.
I've seen spectrum tubes before but not ones as small or as fully contained as those.
Should start marketing this product to high end audiophiles, lol.
Very cool video for such a beast. Nit-picking@ 44:44, RMS value after adding a DC offset should be 2.121kV (3000/sqrt(2)) theoretically.
Nice Tru at a joke but the distortionnis way too high 😂
I’d like to see the companion electrostatic speaker this thing could drive! Morbidly fascinating, scary stuff, even on video.
What a monster!
That is crazy
I have 2 of the smaller and older TREK610C amplifiers. They use the same style voltage dividers as Yours, I have wondered how that is built, to expensive to take apart though... One trade name of these was coratrol, I do not know which part it refers to. These were very expensive back then, over 10000$ for the 610C if I remember correctly, still work fine. These use vacuum tubes, the somewhat later 601D was solid state, same specs. TREK s really good quality.
The Oriel "supply" in the beginning @ 6:35 should be called a "ballast". These use a lot of thin wires on the secondary, where the high DCR of the windings allows the "loaded" output to decrease in voltage, limiting the output current, essentially becoming a series wire wound resistor. These are common in all discharge tubes used in lighting, from fluorescent tubes to street lamps to movie projector arc lamps. What is "discharge"? Arcing which is a temporary very high current short.
Most of the current limiting comes from magnetic shunts, so that the magnetic field in the coil can be controlled, so that it can generate a high voltage, but as you attempt to draw power the shunts limit the current that can flow. Microwave ovens use these, and if you remove the shunts you will almost instantly destroy the magnetron and diode, as now the current is no longer being regulated by the shunt laminations between the primary and the 2 secondary windings.
@@SeanBZA No. See the items listed in the examples and the load being arc lamps. Magnetrons are not arc lamps and the power supply is a transformer, not a ballast. Perhaps you should de-construct a readily available fluorescent lamp ballast to see lots, and lots of fine wire. You should measure the resistance and then apply Ohm's law or simply research Electrical AC Ballasts.
@@shazam6274 Simple inductive ballasts use reactance, not resistance, they use thin wire simply to keep cost down, they can make it low DC resistance, but it really does not do more than dissipate heat, making the ballast less efficient. they are calculated for a specific inductance, to provide a certain current flow when the lamp is in circuit, and the resistance and core losses make them much less efficient, but also make them a lot smaller to fit the fitting. You can look at 90 year old ballasts, which were made with larger cores and thicker wire, but the WWII shortage of copper and iron made them make them as thin and light as possible, running hotter, as a cost cutting measure, so the manufacturers kept them that way, because the customer was paying for the wasted power, and they did not want to make a more efficient ballast that would be triple the cost.
Magnetrons are essentially the same as an arc lamp, current draw is non linear when filaments are hot, and they need to have a roughly current limited supply to operate, thus the shunts to do so cheaply.
The only thing I'd use a MOT for outside of a microwave oven is to build a high current transformer for a spot welder. It involves removing the secondary coil and rewinding it with a few turns of very heavy copper wire.
That's more or less all they're really good for, everything about them is marginal except the potential to kill in the factory secondary, just good enough to run the mag with forced air cooling, for only a few minutes at a time in the super cheap dorm room specials. They are not the easily reworked general purpose power transformers they appear to be.
👍👍👍👌👌👌....
This was very good. High voltage I'm pretty rusty on.
Yessssss
My electrostatic speakers operate at about magnitude lower high voltage and use a wide bandwidth power audio transformer to dramatically step-up the voltage swing to into the mostly capacitive electrostatic transducers. Create like there’s thousands of volts behind your voice. 🤪
I work on high power PA audio amplifiers. They work with another magnitude lower voltage but are able to drive up to +-250V into 8 and 4Ohm loads. The biggest 2 and 4 channel models can put out 7.5kW and 5kW of rated power per channel and have a powersupply capable of 3.5kW average power with a truly scary ammount of output capacitors.
Well you are a braver man than I. That machine is downright scary. Duck.
cool.....
Very cool and quite scary instrument
Great video. You are so brave! I have a tesla coil that can make 6 inch sparks. I only run it in a room with none of my other equipment - no oscilloscopes, computers, etc. I keep at least 3 feet away from it - the thing scares me!
What are the practical applications for this instrument?
More like what aren't the practical application for this beast :P
@@great__successsrsly what
maybe feeding the deflection or the prime acceleration on particles accelerators, with modulated signal, but not sure
Back in the old days CRT TVs ran at 25kV find an old one on garbage day you have the bits to make a HV DC power supply.
+1 for more high voltage/power topics
I'm really curious what sort of semiconductors are used for the final output stages. Any way you could snake an endoscope in there to get a better view?
Also surprised at the construction of those "light pipes". I'm guessing that's another good reason the lid is interlocked to the electronics. Any ambient light getting inside might completely throw off that interface. Wonder why they didn't cover them in black goop.
I'm surprised it wasn't a tube amp, tubes do well for high voltages.
What kind of tube would handle 20kV for something like this? The ones I've seen with that kind of voltage ratings are massive with plate dissipation rated in hundred of kilowatts.
@@rocketman221projects A custom made one, unlike custom silicon, tubes are fairly easy to manufacture and even amateurs have made some that work well.
PRAM- These devices are also used to measure high voltage power lines
Can You do a X-Ray of the output divider tube in the HV amplifier?
That would be interesting indeed. I'd expect to see a looong chain of high-voltage resistors and capacitors in the potting compound.
I have no clue how to use any of the equipment (some I don't even know what they are for) in the back. But I just think it all looks cool.
6:16 is that our old (and much missed) friend Photonicinduction?
I had the same thought
Very very interesting video! Thanks!! If you still have the gear it would be interesting to see if you can deflect that stable Xenon plasma with a magnet - on the end of a long insulating stick!
Regarding non-10MOhm input resistance: It might be current sinking by multimeter for measurement? Or resistance non-linearity at high voltage?
nice
How do they design such high-voltage capacitors? I mean it must not spark through anywhere. So this constrains the width between the electrodes and choice of dielectic medium. But this would also constrain the capacitive value. And ofcourse the construction of the component, connecting leads, housing etc., it must not have any irregularity that would cause it to spark too soon. High voltage is quite interesting. At school (long time ago) i had to experiment with high voltage across a domestic plastic screw terminal block to determine the spark voltage, i believe it was 2+ kV when it started to arc, also audibly 😁
You may have violated the breakdown voltage rating on that proto board 😱
I wonder if there would be visible corona between the rows in the dark. 🤔