coil discharge

The energy in the capacitor has to be dumped directly to the coil. The resistance needs to be low as possible and the switch needs to be fast. Your circuit will just bleed the capacitor through the resistor and the transistor, that is if your transistor can withstand the surge. If you want to save money you can knife switch the capacitor into the circuit when you want to fire. Tom

I hope your not driving the base of your transistor through a 4.7K resistor like their using to drive SCR gate in the circuit you linked to. You will need the lowest possible gate resistance to drive the transistor completely into saturation and very quickly. A 4.7K resistor will have the transistor operating in the linear region and it will have to absorb the brunt of the power instead of the coil. Try using a 10 ohm resistor for the base-drive limiting resistor.

If your still killing transistor you can try paralleling transistors but you'll need a low-value stabilizing resistor in series with each emitter which only needs to be a few tenths or hundreth of an ohm.

Dorian

The transistor shouldn't be at risk if it is driven into saturation. What is the failure mode? short, open, from which elements? That gives you an idea on how it died and what is wrong.

A low gain, high speed, high voltage, part like the 2SC3979 should be handle what you want to do if properly biased and protected from transients. It would be a great part for making an electronic ignition - switching 12 volts into an auto induction coil for instance.

I wouldn't use it for a capacitive discharge ignition (essentially what you, or "the coil gun kid," are doing Perhaps if you gave us some idea of your intention/goal is we could better advise . . . The site you link to was done by someone with a barely rudimentary (mis)understanding of electronic theory.

SCR's are the part of choice for dumping large currents fast into inductors in the 1-1000 volt range - - - and surviving the aftermath.

An SCR can be made out of two transistors (probably not practical for what you are doing) see

Wire the pnp/npn pair in the configuration shown and it behaves "just like an SCR." I've used it once or twice when I didn't have an SCR on hand. I've never had cause to try it in a high voltage application though . . .

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

The #1 Newsgroup Service in the World! 120,000+ Newsgroups

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

Spec sheet for the transistor you are using

www.ortodoxism.ro/datasheets/panasonic/SJD00122BED.pdf

In addition to the steady-state (strictly DC switching) applications there's something called "the safe operating area" or SOA for bipolar and FET transistors - the rules change when you drive inductive loads with single or repetitive pulses.

Page three shows a TUT (transistor under test) with bias and clamps to protect it from the inductive kick (reverse EMF) generated by the inductor in the collector circuit.

The places where they show variable batteries and diodes - those would be zener diodes in the real world . . .

To safely apply the part, you need to drive it into conduction - not just tickle it a little into the linear range - turn it on hard. Then you need to deal with the transients and ringing that develop in the inductor when you turn it off . . . with diodes to clamp the voltages and keep them from harming (breaking down the insulation inherent in the junctions) the transistor.

That digital scope you showed in one of your earlier posts does not show an accurate picture of what is going on transient wise. A good

100 MHZ or better analog scope will show things with much better resolution - there's voltages and ringing there that you don't see now.

This stuff takes a better understanding than "I've got a transistor and I want to switch 300 volts into a coil."

It is doable but takes a few more parts to keep the transistor from failing.

Don't get discouraged. Learning this stuff takes some time and effort

- you are doing it the best way (in my opinion) with real parts and real circuits - but it does take some "book learning" too.

I remember my first power transistor - they cost a lot back when I was in high school. Mine was a donation/gift from my father. Germanium PNP in the new TO3 case - thick solid copper with a brass nipple where they back fill them with dry nitrogen to protect the wire bonds from corrosion. I made a circuit to switch my bicycle headlights from generator to battery as the RPM dropped too low to get usable light from the generator. Later it became the driver for a blocking oscillator that worked a horn type loud speaker in front of my bike. We didn't just buy new parts, we had to reuse the ones from earlier projects - a small germanium transistor was $5. them were the good old days

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

The #1 Newsgroup Service in the World! 120,000+ Newsgroups

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

If you have a complimentary pair (PNP & NPN) of power transistors that can both handle the voltage & current you want, then you can make your own SCR. Connect the base of each transistor to the collector of the other, the NPN emitter acts as the cathode and the PNP emitter acts as the anode - each base/collector pair acts as a gate, so you have +ve trigger & -ve trigger like an SCS.

It is goign to work for this project , also I was looking on doing a mini tesla coil, wich requires basically the same circuit but having a frequency. (after reading) Using a SCR, I will have to discharge the capacitor completely before the next impulse. I'd like the capacitor to discharge to about half its potential (or around) otherwise I will not be able to have the frequencies I want.

Ken

Hi, Ken. First, the battery in the illustration is backwards -- the gate has to be *positive* in relation to the cathode of the SCR in order for the SCR to fire.

But, from your description, you're not interested in using an SCR. You want to fire this thing with a transistor, and interrupt the discharge at about 50% of Vcc.

Your basic problem is going to be peak discharge current. In order get useful advice (choosing a transistor and driver circuit), you'll have to provide more information about the coil. Can you find out the inductance and resistance of the coil?

Thanks Chris

How do you calculate inductance with a multimeter ?. I dont feel like unwinding it to get its lenght. Anybody knows magnet wire density for gauge

23? Also , is there equations to determine the amplitude of spike voltage induced into a coil? I was wandering, if a dc motor is used, isn't there also spikes occuring in the coils of those motor. big spike could cause damage, but wouls smaller spikes help the motor turn faster (or with more torque) ? I mean if your using a small dc motor, like the one in my radio controll car (9v) the spikes behing, let's say 20v. These spikes would actually help the coil repulsion to the permanent magnet on the stator no ?

Ken

One way would be to get a 6 or 12V ac 60Hz output from a transformer as a voltage source. Connect it to the coil through a series resistor. Read the voltage across coil and the resistor, and change the resistor until you get equal readings. When the readings are equal, you can calculate the inductance by L = R / (2pi *f), or L (Henrys) = R (ohms) / 377. That's assuming that the coil will have a much lower resistance than the resistor; the formula is a little more complicated if it isn't: L = sqrt ( Rexternal ^2 - Rcoil ^2 ) / 377. * Before you start, you should measure the resistance of the coil and make sure that the external resistor never gets so low that the current rating of the transformer is exceeded. You may find that this procedure won't work at 60Hz if the inductance of the coil is too low. Oh, and don't electrocute yourself. It would probably hurt.

Yes- the voltage induced in a coil by current change is V = L*di/dt, or inductance times the rate of change of current. In the case you are talking about, though, you are applying an external voltage, and getting a resulting rate of change of current ( the formula is the same).

There will be spikes from the interruption of the current (same formula as above) as the motor rotates, but they don't make the motor run faster- they're incidental.

-- john

put it in series with a known resistance and hook it up to an AC supply. measure voltages.

Bye. Jasen