I think Joerg gave me the idea for this:
It's a transformer, an inductor, and a pair of switching transistors. In theory you need a capacitor across one the windings on the transformer to create the resonant tank, but often the inter-winding capacitance is big enough to do the job on its own.
Typically you put a third winding on the transformer to drive the gates/bases of the transistors, and use a couple of resistors to get the DC bias right.
It really isn't a large number of components, and none of them are expensive.
Who knows? It isn't the kind of feature you'd boast about.
Jim Williams series of application notes AN45, AN49, AN51, AN55, AN61, AN65 do suggests that it was popular for one particular application, though cold-cathode back-light do now seem to have gone out of fashion.
John Larkin doesn't seem to like transformers and he certainly doesn't like designing special purpose transformers for particular jobs and getting them made in small volumes.
Ferrite cores for small transformers are still stocked by broad-line distributors, and you can now buy cores that aren't too lossy at a couple of MHz, and GaN switches to drive them.
Baxandall invented his resonant inverter in 1959 specifically for use with bipolar transistor switches.
It's a transformer, an inductor, and a pair of switching transistors. In theory you need a capacitor across one the windings on the transformer to create the resonant tank, but often the inter-winding capacitance is big enough to do the job on its own.
Typically you put a third winding on the transformer to drive the gates/bases of the transistors, and use a couple of resistors to get the DC bias right.
It really isn't a large number of components, and none of them are expensive.
Who knows? It isn't the kind of feature you'd boast about.
Jim Williams series of application notes AN45, AN49, AN51, AN55, AN61, AN65 do suggests that it was popular for one particular application, though cold-cathode back-light do now seem to have gone out of fashion.
John Larkin doesn't seem to like transformers and he certainly doesn't like designing special purpose transformers for particular jobs and getting them made in small volumes.
Ferrite cores for small transformers are still stocked by broad-line distributors, and you can now buy cores that aren't too lossy at a couple of MHz, and GaN switches to drive them.
Baxandall invented his resonant inverter in 1959 specifically for use with bipolar transistor switches.
What I'd wonder about is something like the idea of the thyrototron after the thyratron (gates, valves, the thyra-), then is for field effect devices since plasma as a state of matter yet doesn't count for a "moving part" and thusly is yet by definition solid-state the circuit, about variable inverters, converters, and transformers, and the continuously variable, basically about thyrototron-in thyrototron-out DC-DC continuously-variable converters.
What I'd wonder about is something like the idea of the thyrototron after the thyratron (gates, valves, the thyra-), then is for field effect devices since plasma as a state of matter yet doesn't count for a "moving part" and thusly is yet by definition solid-state the circuit, about variable inverters, converters, and transformers, and the continuously variable, basically about thyrototron-in thyrototron-out DC-DC continuously-variable converters.
The thyrototron is nebulous idea that nobody has built yet.
If you can't buy it off the shelf from a broad-line distributors it isn't interesting here.
John Larkin's real objection to the Baxandall class-D inverter is that it usually needs a special purpose transformer specifically designed for each particular job. All that has to change ion the transformer design is the turns ratio - you can buy off the shelf cores and coil formers that any little shop with a cheap coil-winding machine can turn into the transformer you need - but you do have to work out the number of turns and the size of the wire, and that kind of little shop tinkering is below John Larkin's dignity, even if most of the rest of us can live with it.
They are fine if they are available and affordable and, preferably, surface-mount. Coilcraft has some cool stuff. The DRQ-series parts are great, made by several sources.
I don't like custom transformers. They are expensive, hard to multi-source, and really slow down a design.
We rarely design in siicon bipolar transistors these days, and never for power. Logic-level mosfets cost a few cents. GaN is fabulous.
John Larkin Highland Tech Glen Canyon Design Center Lunatic Fringe Electronics
Thanks for writing.
(This "thyrototron" is a sort of "ideal commutator" device after ideas of the Faraday rotation atop a variable transformer that makes for being an "ideal commutator" of the "solid-state" variety.)
The idea that any circuit is an RLC circuit agreeably is very involved, then as with regards to the off-the-shelf and the lines-of-manufacturing, it's agreeable that synthesis is both practice and art.
For maybe 1000x the cost of a part you can get overnight from Mouser.
- but you do have to work out the number of turns
I design transformers when it makes sense. We did a bunch of pot-core transmission-line transformers last intern season, but I'm trying to design them out this summer.
Capacitors don't saturate. But transformers can invert.
John Larkin Highland Tech Glen Canyon Design Center Lunatic Fringe Electronics
What's a thyrototron?
John Larkin Highland Tech Glen Canyon Design Center Lunatic Fringe Electronics
Thanks for writing.
If you recall, last year when the "matter and energy" thread or what it was you spilled it over to sci.physics.relativity, then at some point we got talking about power conversion and then I mentioned this outline of the concept and factors of the design of an "ideal solid-state commutator" called the "thyrototron".
So, by definition pretty much it's any "ideal solid-state commutator", here then "the thyrototron" is basically a DC electrode pointing through a ring magnet with some grooves and lands mated over a variable transformer, DC in, AC out.
Then the idea of "DC-in DC-out" as "thyrototron-in thyrototron-out" is a power converter, that's vari-able.
I'd think you'd recall since you posted an outline of it to some electronics design site or if you search for "thyrototron".
Oh.
I don't recall seeing the "thyrototron" term before. It sounds goofy.
A google search has no hits.
John Larkin Highland Tech Glen Canyon Design Center Lunatic Fringe Electronics
I think the industry has overcome that limitation. Energizer is claiming a shelf life of 10-15 years for their lithium Energizer AAs. The overhead loading for the step-down should be about the same as for a step-up.
You don't multi-source custom transformers. You get them made by somebody you know and trust. That's what "custom-made" is all about.
1959 is 67 years ago. Not the "olden days" you thought that you were referring to, but things have changed since then.
Ignorance is expensive, and it can lead to implausible claims. If the part you can buy from Mouser won't do the job you need done, it is worthless.
Of course you are. Transmission line transformers are difficult to understand. and you don't like looking ignorant.
Capacitors don't "saturate". They break down, which is a different way of going non-linear.
Transformers can invert - it is a virtue, rather than some kind of unpredictable trap for the pig-ignorant.
Actually, high-value ceramic caps *do* saturate.
Jeroen Belleman
One of the chores of electronic design is keeping up with new parts. And having the restraint to not use too many of them.
John Larkin Highland Tech Glen Canyon Design Center Lunatic Fringe Electronics
Our pot core txline transformers limit our pulse widths, both single-shot volt-seconds and our ability to make multiple pulses per trigger.
Caps running out of charge just make long pulses droop.
Making 700 volt pulses is nasty. Transformers saturate and ceramic caps run out of C, and nobody makes wet electrolytics or polymers up there.
It's almost criminal that people sell a 10uF 50V cap that has 2 uF at
50V. And usually don't admit it on the data sheet.Most ceramic caps don't break down at 2x or even 5x rated voltage. They just quit being caps.
John Larkin Highland Tech Glen Canyon Design Center Lunatic Fringe Electronics
There's probably a 1000 chinese manufacturers of high voltage film capacitors. Most will custom manufacture low quantities (100's).
They will typically offer from 100n up to about 2uF, with voltage ratings up to about 30kV.
Dawncap comes to mind (since I have one of their business cards at hand) but if you search you will be spoilt for choice.
News to me. The odder very high-dielectric constant ceramic capacitors do do a lot of odd things, but we avoided them because of that.
Of course they do, and you can't be bothered to work out what's going on when you run into those limits. All electronic components have limits and its our job to know where they are and design around them,
The market for polypropylene dielectric film capacitors does seem to extend to voltages higher that 700V.
It's almost criminal that people sell a 10uF 50V cap that has 2 uF at
That's an exotic high dielectric constant ceramic part. You do need to read the data sheets carefully. If it doesn't specify how the capacitance changes with applied voltage you may need to be careful.
Not in my experience. But I worked in places where people did seem to know what they were doing, and didn't design in funny ceramic caps that claimed to offer huge capacitances in very small volumes.
Have something to add? Share your thoughts — no account required.
Ask the community — no account required