Plasma, Flybacks, RF Burns and Exploding Transisters

yes, take your pic from any major electronics stockist's catlogue, or IR website etc.

Colin =^.^=

WuzhengNA a écrit :

Sure. That'll be four, two of which payable in advance. Another one at half the project, the last at delivery. Oh, Euros of course.

Yep. And with a fair amount of knowhow one can make mosfets melt as well as BJTs.

So will I :-)

Seriously, if it's really worth something and you and the money, then subcontract this, and carefully choose your subcontractor. I'm afraid this is pretty well above your head.

If you grind the top off its 'hat' then you might find the resulting gobs of plasma are more accessible to your external system assuming you can work out how to get them in there.

I find no fault in your logic. If you replace the 2N3055 with a MOSFET then the 2N3055 should stop melting.

This might be a similar process to thought transference but the problem might be that the problem is transferred but remains local to yourself....... bummer.

Oh, ignore Eeyore..... He moved to Darlington....

and thinks it's cool to be a Duddy Fuddy.

DNA

There's two/three ways to deal with the problem.

You use a transistor suitable for "horizontal output" service on TV's (what normally pumps a deflection flyback transformer in a TV to generate 10-40KV) they usually have a breakdown voltage over 400 volts (versus the 80-100 the 3055 has)

You clamp the voltage using something that reacts fast enough to protect the transistor - like a power high speed damping diode - reverse biased across the primary of the coil. Slick and easy - but some times you can't afford dump the spikes - you want them for generating high voltages - and you need to keep the spike - sounds like what you are about.

Some times you can just use a ferrite bead in one leg or another of the circuit - they kill the fast rise times that can kill transistors with something called "storage time." Storage Time is just the high frequency response of a transistor - you push square waves through a transistor - turn on and turn off don't happen instantaneously - there's a delay (2N3055's are cheap and antiques - they are low voltage and they aren't very fast). You can get a faster transistor or use a ferrite bead to kill the fast transients . . .

It really helps to put a scope on it - sacrifice a transistor and watch what is happening. or tell us how the 3055's died.

The Tesla coil folks should be able to help with their forums - should you want to join a forum.

You want a lot of hot plasma in an enclosed space - piece of cake. I fooled with telsa and induction coils - reliving my youth - it is very easy to get what you want, but there may be better ways to do it than a flyback HV generator.

With Tesla coils - plasma is usually a problem - generates ozone and that's harmful to lungs.

I have a 1KW induction coil that takes 6 horizontal deflection coils in parallel to run - the transistors have a 1500 volt breakdown and cost $8-9 each, That's on a par with using 12 volts at 120 amps to produce HV. I use 30 VDC and ~400 Hz - the coil produces 100Khz ringing and ~40 - 50 KV. - can make X-rays . . .

----== Posted via Newsfeeds.Com - Unlimited-Unrestricted-Secure Usenet News==----

----= East and West-Coast Server Farms - Total Privacy via Encryption =----

Where is Mike Harrison when we need him? Hopfully preparing somthing extra lethal for the Dorkbot London christmas special but thats anther story.

I'm assuming that the can is closed, thermally insulated to a practical level and reasonably gas impermeable.

I'm assuming that you need to make the plasma quickly enough to not melt a steel part inside the can that is being treated.

I suspect 1440 watts is not enough.

If my assumptions are incorrect and you want an open ended can of white hot plasma then it is obviously not enough. You don't get plasma from a 3Kw electric fire element.

Lets consider some numbers to get a vague idea. This is mostly guesswork with formulas and values via google. Others may have a better knowledge of the maths physics and phsical values.

Aprox 2200 cubic inches of air aprox 14000 Kelvin

Google calculator says thats 0.037 cubic meters,

Closing the can at 20degC you have about 0.05Kg of air.

Specific heat capacity of air is aprox 1 joule per gram per degree kelvin acording to wikipedia.

To increase the temperature of the air to 14000K therefore requires about 685000 joules.

The can, the workpiece and the catalyst are going to absorb some energy. For a rough guess heating

Obviously you don't want thermal equilibrium and melted steel.

Heat loss? Can surface area aprox 1000 square centimeters. Actually thats hard to estimate.

Lets just say 5 megajoules.

I'm guessing 45% efficency in your DC to RF conversion and that's probably generous considering you mentioned

10 megajoules.

Bob

I would assume (right or wrong) that he wants to goose some platinum or palladium coated ceramic beads to get them going before the catalyst is hot enough to work on its own?

Generally speaking . . . a car catalyst is used to burn off carbon monoxide. CO is flammable, just not at low temperatures and not without oxygen.

The air pump was the first solution to reducing CO emissions - just give it the oxygen it requires and it will burn off - and make the exhaust system glow red hot, and settle for an incomplete burn..

Plan B was to let the unburned hydrocarbons into the catalytic converter and they (in the presence of platinum) would combine with the oxygen in the exhaust and burn off - at a high temperature.

I'm sort of assuming a lot here . . .

But, take the residual molecules of oxygen in an exhaust and make them more reactive by making ozone and that just might make for a really high performance catalytic reactor. So use a high frequency source of HV to bathe the little catalyst beads in an ozone plasma

Hey, what do I know?

From what I've seen with 1KW plasma - a catalytic converter may be moot. But I'm assuming that's not his goal - just get a conventional system going earlier or make something that requires less platinum? Platinum costs money - even in the flash coating they use in ceramic beads for catalytic converter.

CO is fuel - any engine that can't burn it, and produce work, should be redesigned IMHO.

----== Posted via Newsfeeds.Com - Unlimited-Unrestricted-Secure Usenet News==----

----= East and West-Coast Server Farms - Total Privacy via Encryption =----

why not put the catalytic converter inside the cylinder where any energy released can do useful work ?

Colin =^.^=

You all are great - here is EXACTLY what I am trying to accomplish, since you have guessed generally -

This is NOT an emission control device for a gasoline engine (which is very easy, due to so much hydrocarbon left in the exhaust, just inject some atmosphere, hit the catalyst, and you are good) This is NOT to pre-heat the catalyst, nor is it to burn off residuals.

This is a system for diesel engines. Diesels run very lean. Exhaust has gobs of Oxygen left in the exhaust, which makes Cat Converters somewhat useless.

Diesel exhaust has PM (particulate matter) and NOx, which are the big problems. The common HC, CO, are easy. NOx and PM are not.

I need the plasma / corona / magic genie to be active in a running exhaust stream to create radicals of N and O, to form up with the NO to form NO2, O2 and N2 with the help of the catalyst.

I have theorized a clear N2 run thru a plasma jet, creating radical N to recombine making N2 and O2, but don't want to carry a tank of nitrogen on the truck.

The 6 diameter by 20 long "can" is not closed - it is what the exhaust will flow thru.

I would love to have someone design this part - I have a desk full of stuff, parts, but I know my limitations.

Aha I see, but I think your gona need a whole lot more power, have you ever seen a plasma cutting torch ? the one I worked on would just about run off a 13amp 240v socket without tripping the main circuit breaker too often, but the amount of plasma isnt very volumous, very fun to play with however.

Colin =^.^=

You have clarified what you are trying to do. I thought you wanted to do vapour deposition of the catalyst or somthing.

One idea for the particulate matter is to use a cyclone system similar to a bagless vacumn cleaner to collect it. James Dyson, inventor of the bagless vacum looked into it and IIRC found enviromental regulations made disposal of the collected more troublesome than spewing it into the air. I can envisage some sort of scheme to collect the particulate matter then burn it off with a heating element when you have a big lump but making it work reliably in automotive conditions would be another matter.

I'm still not convinced you have sufficient power to do anything but make the exhaust gas a bit hotter. Anyone got any estimates of temperature required, input gas temperature and flow rate?

I suspect the extra fuel burned to power your idea would make it a net loss for the enviroment although it might be a way of somehow "gamign" emissions regulations.

Bob

Again, great input. I have figured out the diesel Particulate Matter - Catalyst for a gas engine is simply an open honeycomb, the holes thru the honeycomb go clear thru.

This diesel catalyst looks similar, but every other "hole" is plugged on opposite ends of the honeycomb. Lights off at 200C.

I'm trying to get radicals in the exhaust stream, from the exhaust stream. Assumed (heh heh) that plasma is the way to go. Pulsed corona? How about a multi-layered OAUGDP? Don't care what it LOOKS like, we need the effect.

Use a microwave oven, build your own waveguide as a section of tailpipe. Then use a wire mesh like those used on microwave oven doors to block in the plasma, and allow the exhaust to pass. Place a sparkplug in it. The sparkplug will ignite any gases that haven't burned. There is enough power if converted to 120Volts you can have

Then the microwave will sustain the plasma in the section of the exhaust you've chosen. Then follow with more pipe and the muffler.

And what is a OAUGDP?

My telsa coils seemed to be great ozone producers. Had to keep the window open and fan running. The most ozone was produced when working at lower voltages with high power - thick long blue streamers from the top of the coil instead of forked lightening or big white arcs.

Can't really judge ozone volume by sense of smell. A tiny amount is perceptible, but a large amount has only a slight increase in perceived intensity. Stretched latex rubber bands in the airstream between the coil and window were a crude indicator.

Haven't messed with that stuff in years, but I'm sure the Telsa forums are still going strong. They had several mosfet driven coils.

The Amateur Scientist article in Scientific American from the 50's had a nice compact Tesla coil for a home made X-ray machine. They use a "kicker" coil to power it - a mosfet or transistor switch could probably be used instead of a mechanical interrupter.

Check out

----== Posted via Newsfeeds.Com - Unlimited-Unrestricted-Secure Usenet News==----

----= East and West-Coast Server Farms - Total Privacy via Encryption =----

On a sunny day (Sat, 02 Dec 2006 09:45:26 -0500) it happened default wrote in :

The old toobe TVs used 25kV and a parallel stabilizer PD500, or PD510:

It may be possible to get usable X-rays from the lowly #47 lamp. Back in the day I put all manner of things atop the coils - those miniature bayonet lamps would fluoresce green (just the glass envelope - not the inside)

----== Posted via Newsfeeds.Com - Unlimited-Unrestricted-Secure Usenet News==----

----= East and West-Coast Server Farms - Total Privacy via Encryption =----

I'm no plasma scientist, but isnt this going to be, like, impossible?

Plasma is really, really, really hot. Hotter than Britney in a leather bikini. We're talking hot.

Now I assume you need to mix the exhaust gases with the plasma, otherwise they're not going to have a chance to react.

And just ballparking it, let's assume a 400 hp diesel engine is putting out one cubic (large) tomato can of gases every second.

And let's assume each atom of plasma can snuggle up to TEN atoms of exhaust before it cools off too much to be useful.

So you need about one tenth of a tomato can (approximately one tic/tac box) of plasma per second.

Now that's going to take roughly 74 flubillion joules per second, assuming 100% electric to plasma efficiency.

Which is about 3.3 x 10^5 times more power than the deisel puts out in total.

I could be off by a digit or two,

Real physicists help me out here.

The bottom line is: have you calculated the power required?

I followed the noah link backwards and found an interesting tidbit on X-ray radiographs. He discovered that Polaroid film is sensitive to X rays and not visible light as it is developing. He takes a picture of nothing then has a 60 second window to expose it to X-rays

----== Posted via Newsfeeds.Com - Unlimited-Unrestricted-Secure Usenet News==----

----= East and West-Coast Server Farms - Total Privacy via Encryption =----

Uh, no.

A car exhaust has very little unburned hydrocarbons in it, if you look at the sticker under your car hood, it's going to say something like

Do a little math, my guess is a microwave oven is a couple powers of ten too weak to make that enough plasma.

Atmosphere Uniform Glow Discharge Plasma AUGDP

goggle is my friend

----== Posted via Newsfeeds.Com - Unlimited-Unrestricted-Secure Usenet News==----

----= East and West-Coast Server Farms - Total Privacy via Encryption =----