pnp darlington pairs

[humptydumpty]

i have some low gain transistors, and was gonna attempt using the darlington method to increase the gain

if im to believe what i read, i take the gain from the first tranny and multiply times the second tranny, and id have my gain?

so if i had one 23 and one 13, id have a gain of 299?

please talk me through this if im not right, and also sorry if this question is redundant, ive had a long week

[R.G.]

i have some low gain transistors, and was gonna attempt using the darlington method to increase the gain

if im to believe what i read, i take the gain from the first tranny and multiply times the second tranny, and id have my gain?

so if i had one 23 and one 13, id have a gain of 299?

please talk me through this if im not right, and also sorry if this question is redundant, ive had a long week

Yes - and no.

Yes, the gain is the product of the two gains.

The "no" part is more subtle. Current gain is not a fixed number, it's variable with temperature and operating conditions. In particular, it varies with collector current itself. This is why when a manufacturer specifies gain, they specify the current at which it's measured. It turns out that current gain is smaller at smaller currents, increasing with current up to some maximum, then starting to fall off again. This is on top of it changing with temperature.

When you operate two transistors as a darlington, the gains do multiply, but the first transistor is working at a collector current that's the gain of the second transistor smaller than the second transistor is, so the first transistor is running at a much smaller current. So it's gain is lower. Integrated darlingtons take this effect into account and have better low current gains overall because the first transistor is designed for lower currents as much as it can be.

And if you're doing PNP darlingtons out of germanium, you will run into the bane of germanium - leakage. Germanium leaks about 1000 times as much as silicon. In a darlington, the second transistor multiplies the first transistor's leakage current by its own current gain. So the darlington's leakage will be a lot bigger than either one of the underlying transistors' leakage.

Still, it may work OK in non-critical applications. Try it out. Nothing really to lose.

[Brymus]

These are the ones I sent you right?
That I bought from another forum member.The gains are supposed to be higher than they read due to an internal resistor.
IIRC the leakage on those was extremely low,those are the GT313 IIRC,all the Russian PNPs I have tested have had extremely low leakage.
Let me know how they sound as I havent even used the ones(any PNPs GE) I have yet.
It was suggested to me to use a Rangemaster to get a better idea of the gain,you might pop them in one then listen and see if they sound low or higher gain.

[Brymus]

BTW thats a good idea  ,that I didnt think of...
For using these low gain PNPs I was sent by mistake, (a different batch I bought on Ebay-GT402's)
So definetly post your results.
Bryan

[humptydumpty]

Yes, they were the ones you sent me.

I will be messing around with different ideas shortly, and I would be glad to post my results!