Crapacitors

Tants across supply rails are mini-detonators. Current through the cap, C di/dt, can ignite them. Use aluminums.

It will make a damned fine temperature sensor, too.

Not just baseband -- scouting Digi-Key, you can get Y5V down to 10nF. There's surely something this would be really good for, but in ten years I haven't thought of it yet. (Plenty of ideas for VCOs, resonant tanks, etc, but no killer apps.)

It might be handy for tuning the tanks of one of these resonant wireless energy schemes--crapacitors + inductors makes a handy tunable, low-loss energy storage.

Interesting gadget. Just about what I asked for some time ago.

Cheers! Jeroen Belleman

Yes, very impressively bad. Looks like fodder for a green grant--it could be sooo much better with just a little invest-mint(tm).

they are pretty ungraceful about failure. I'm sort of surprised nobody makes internally fused tantalums.

No one said it was useful for energy conversion.

It's interesting, because normally non-linearity will only produce harmonics, multiples, of the input frequency. This one also produces sub-harmonics, at half the frequency in this case, just using a single passive non-linear element.

There is an optical domain equivalent gadget that fascinates the quantum theory crowd since quite some time already. It also has an impressively poor conversion efficiency: A BBO crystal. I've been looking for a simple passive circuit with similar behaviour. This appears to qualify.

Jeroen Belleman

he

Which is exactly why it's ripe for green (or green for ripening), so to speak.

Replacing John's schottky with a Y5V capacitor gives the same results, for similar reasons. You can try it with a varactor in LTSpice. I just made a /3.

The resonant energy reference wasn't entirely frivolous: this circuit points out pretty well the exact challenges of wireless resonant transfer in pragmatic detail. With ideal parts, the weak transfer- coupling isn't a problem since all the un-transferred energy just sloshes around un-wasted in a perfect tank circuit--there aren't any ohmic losses. In reality, there are ohmic losses, and that kills it.

The world of optics and atoms is a different matter entirely; there you have all sorts of nearly ideal components.

e.g.https://groups.google.com/group/sci.electronics.design/msg/1257d50b73...

I used the diode because I didn't have an LT Spice model for a nonlinear cap. I think the conversion efficiency could be less ludicrous if the thing was futzed with some. I just proved that it was possible!

There's a sort of circular reasoning, that the bigger the subharmonic voltage, the bigger the subharmonic pumping energy is. So there should be room for improvement.

In theory, one can create an arbitrarily humongous magnetic field by feeding a small amount of power into a resonant tank. But the Q will getcha in real life. Real life kills so many great ideas...

Oops, that's C dv/dt

I've seen internally fused tants. But if it shorts, it doesn't catch fire, but it's still a failure.

Polymer tantalums, niobium caps, and polymer aluminums don't detonate. I really like polymer alums.

e.g.https://groups.google.com/group/sci.electronics.design/msg/1257d50b73...

Hmmm, maybe a step-recovery diode or even a PIN diode could do subharmonic generation.

You forgot to mention that they could be used in an artificial transmission line to speed up the leading edge of a pulse ..

Sure. I know John--he's not the monster you think. The FLIR stuff in the SMD heatsink thread would be closely-guarded competitive info at other places, not cheerfully shared (with photos) on the web.

Not really. What harm does it do me to share this information?

Restaurants publish their recipies and cookbooks. Doctors have seminars where they teach one another their techniques. Unions have apprenticeship programs. EEs are paranoid; most won't even show their products, much less discuss the innards.

You're a moron, and a useless moron at that. I care a lot about thermal design: it dominates a lot of high-speed and precision electronics.

And what is that "accolades and tears" "time on the cross" nonsense about? I think it's fun to discover things, measure things, and share them and argue about them; it makes the electronics better.

And I think you are a useless, crabby old hen who has almost nothing to say about electronics design, thermal design, component behavior.

Do some real electronics and quit cluck-clucking about personalities. It might make you less crabby.

How's that Variac experiment coming along? Gonna share it with us soon?

the

in

That's your attitude, but not others'. The fact is it's a pretty cool trick that lets you make cutting-edge stuff better than the other guys. I just finished plowing through a load of references on SMD heatsinking that emphasized how important thermocouple placement was to making reliable measurements.

here

ams.

Service industries sell their services. Companies sell products, and not ideas. Some companies ' livelihoods are based on just a few ideas, techniques, or trade secrets. Like getting max thermal performance, maybe.

the

It's not always paranoia. When working for others I don't divulge their secrets, whatever they think those might be, however small they might seem to me. They're not mine to give. That's just good manners.

the

Oh that's just silly. I've known John for over thirty years.