Prev: Limiter + Opamp + Reverse phase on negative swing
Next: Powering isolated RS485 transceivers over cable
From: Bitrex on 15 Jul 2010 04:18
Your earlier posts got me experimenting with some blocking oscillator
circuits. Here's something a little bit different using an
autotransformer - I found this circuit in a cute 1965 book titled
"Industrial Transistor Circuits."

The coil I used was 15 turns of 16 gauge for the top winding and 5 turns
of 24 gauge magnet wire for the bottom winding on a 1" diameter air
core, with a KSC1173 3A power transistor. The output is about 28V p2p
(pretty close to the transistor's VCEO), and while the top of the output
wave looks nice the bottom has a lot of ringing on the upswing - I'm
sure my homebrew transformer has a lot of leakage inductance. Maybe a
snubber somewhere would help? The frequency of oscillation was about 100kHz.



Out 0.33 uF
o----------------o--o--+.
--- | )|
VCC --- | )|
o------o--)--+'
15V | )|
| )|
.----o-----------)---'
| | 510 1/2W |
| .-. |
--- | | |/
--- | | ---|
| '-' | |>
| | | |
'----o------' |
|
0.1 uF ===
GND
(created by AACircuit v1.28.6 beta 04/19/05 www.tech-chat.de)
From: Tim Williams on 15 Jul 2010 05:34
Could be the base ringing when reverse biased. I usually add a small resistor (maybe 22 ohms in this size) to dampen it and soften forward base current.

With the tank, you'll be looking at something more "class C self excited", too.

You should be able to get a coathanger to glow red inside that coil.

Tim

--
Deep Friar: a very philosophical monk.
Website: http://webpages.charter.net/dawill/tmoranwms

"Bitrex" <bitrex(a)de.lete.earthlink.net> wrote in message news:7JSdnddP5-qaW6PRnZ2dnUVZ_qednZ2d(a)earthlink.com...
> Your earlier posts got me experimenting with some blocking oscillator
> circuits. Here's something a little bit different using an
> autotransformer - I found this circuit in a cute 1965 book titled
> "Industrial Transistor Circuits."
>
> The coil I used was 15 turns of 16 gauge for the top winding and 5 turns
> of 24 gauge magnet wire for the bottom winding on a 1" diameter air
> core, with a KSC1173 3A power transistor. The output is about 28V p2p
> (pretty close to the transistor's VCEO), and while the top of the output
> wave looks nice the bottom has a lot of ringing on the upswing - I'm
> sure my homebrew transformer has a lot of leakage inductance. Maybe a
> snubber somewhere would help? The frequency of oscillation was about 100kHz.
>
>
>
> Out 0.33 uF
> o----------------o--o--+.
> --- | )|
> VCC --- | )|
> o------o--)--+'
> 15V | )|
> | )|
> .----o-----------)---'
> | | 510 1/2W |
> | .-. |
> --- | | |/
> --- | | ---|
> | '-' | |>
> | | | |
> '----o------' |
> |
> 0.1 uF ===
> GND
> (created by AACircuit v1.28.6 beta 04/19/05 www.tech-chat.de)
From: Bitrex on 15 Jul 2010 07:03
Tim Williams wrote:
> Could be the base ringing when reverse biased. I usually add a small resistor (maybe 22 ohms in this size) to dampen it and soften forward base current.

I'll give that a try - the ringing seems to happen right when the
transistor goes back into cutoff after the current pulse so that seems
plausible. I think I'd better add a protection diode as the reverse
bias is coming pretty close to the base-emitter junction limit.

>
> With the tank, you'll be looking at something more "class C self excited", too.

I misunderstood the operation of the circuit somewhat, but after
setting it up in LTSpice and looking carefully at the waveforms plotted
together it makes sense now. The transistor is in cutoff most of the
time, and only comes up out of cutoff briefly, when the collector has
swung all the way down, to deliver a current pulse into the tank. It
then goes back into cutoff again as the tank swings back up.

>
> You should be able to get a coathanger to glow red inside that coil.

Nice! I'll give it a try and report back.

>
> Tim
>
From: Tim Williams on 15 Jul 2010 10:53
"Bitrex" <bitrex(a)de.lete.earthlink.net> wrote in message news:A4KdnZKxtb85caPRnZ2dnUVZ_sCdnZ2d(a)earthlink.com...
> I'll give that a try - the ringing seems to happen right when the
> transistor goes back into cutoff after the current pulse so that seems
> plausible. I think I'd better add a protection diode as the reverse
> bias is coming pretty close to the base-emitter junction limit.

Be careful when you do that -- the base ordinarily consumes current, which is good because it sets bias. With a reverse diode, bias goes both ways, and the transistor stays about half on... phut, bang.

One thing you can do is add a series diode, so it only turns on. But with nothing to remove charge, the transistor stays on. So you can add a capacitor to bypass the diode and suck some charge out. Or you can add a current-limiting transistor, so it stays on until the current reaches the limit, then shuts off. This would put operation closer to class E, where you get a big flat spot while the transistor charges the inductor.

Tim

--
Deep Friar: a very philosophical monk.
Website: http://webpages.charter.net/dawill/tmoranwms
 | 
Pages: 1
Prev: Limiter + Opamp + Reverse phase on negative swing
Next: Powering isolated RS485 transceivers over cable