Curious oscillator circiut

3.3K at the base, beta of over 100, means maybe most of an amp through the power supply, right? I'm thinking it's an interaction between the PS's overcurrent shutdown (notice the nice "flat" spot in the curve) and the load. He didn't put the PS voltage on the scope too, that's the first thing I'd look at!

Tim.

I've built basically this same circuit so I assume yours is doing the same thing. In my case it was a high voltage NPN transistor that did it.

The collector load needs to have a high Q for it to happen. When the collector junction goes reversed, it stores charge in the base. This is what causes the flat spot you see just as it comes out of the reverse situation. You are seeing the storage time of the transistor when operated in this funny way. This storage time is what is adding energy to the tuned circuit.

You can see if this is true by putting a diode in series with the transistor to prevent the reverse current. This will completely kill the oscillation if it is the same effect as I saw. If getting the coil to ring down is what you want, a Schottky diode in series may be the final answer.

Since you are running at a low voltage, you could add a small dual Schottky as a "Baker clamp". This will clip the positive peaks.

Vcc ! --- Vcc ^ ! ! !/ e ---/\\/\\/\\-++--! ! !\\ --- ! ^ ! ! ! ----+

--
--
kensmith@rahul.net   forging knowledge

There was a 47uf tantalum cap across the PSU, so not that. Remember the transistor is biased off DC-wise after the initial kick to start it.

In my case the fix was to tri-state the MCU pin that was driving the transistor instead of driving it high.

It appears to be a blocking oscillator....... But I think there is more going on at the base than the scope trace is showing.

Don

[... funny oscillator ...]] Vcc ! !/ e ---/\\/\\/---! !\\ ! To LC

I'd worry about that in production. When the collector goes above Vcc, it will pull the base above as well. A current will flow through the protection diodes on the PIC.

--
--
kensmith@rahul.net   forging knowledge

Mike,

It might be just due to the collector to base capacitance.

Bypass the base to the emittter.

Mark

But, but, but, nobody's got the point yet, that the whole circuit is STOOPID, but works anyway!

So much for simulation! ;-P

Cheers! Rich

--
-----BEGIN GEEK CODE BLOCK-----
Version: 3.1
GAT(E P) dpu s: a++ C++@ P+ L++>+ !E W+ N++ o? K? w-- !O !M !V PS+++ 
PE Y+ PGP- t 5+++)-; X- R- tv+ b+ DI++++>+ D-? G e+$ h+ r-- z+ 
------END GEEK CODE BLOCK------

behave a little

Once you get the tank circuit oscillating, and the transistor's turned off, there's no load to 'sap' power and stop the oscillation.

Try adding an R in series with the L.

Graham

In message , Ken Smith writes

I think that is exactly it. There's a very similar explanation (base charge) for the behaviour of an old (1957) push pull dc-dc converter..... centre-

-tapped collector-collector saturating transformer, cross-coupled base resistors, etc. At switchover, there is a spike at the collector of the OFF-going transistor, which is transferred by the transformer into a reverse-voltage spike onto the collector of the OFF (but ON-coming) transistor. This is supposed to pre-charge it's base so that it comes ON much faster.

--
Tony Williams