That is the PNP commmon-base circuit I suggested some days ago. But an R-C into the emitter would make it more practical, so the biased fet is usable as an amplifier or something.
A PNP diffamp would be an improvement; less loading on the fet circuit, no Vbe error. Still benefits from the RC.
John
Okay, now you can produce a schematic if someone asks.
John S
Right, some people get very upset if you expect them to visualize circuits all on their own.
I did have in mind an RC in the emitter, which is likely the same thing Ian suggested. But the PNP diffamp is appealing; discrete transistor diffamps turn out to be handy in a lot of situations.
John
A diffamp is unnecessary and only adds complexity.
John S
RC in the emitter? What does that do?
I would put an R-C-R in the collector. Are we talking about the same circuit?
John S
We don't know the actual requirements. But the diffamp eliminates the Vbe error and allows one to use a high impedance, long time constant RC in the drain voltage pickoff, which will keep things linear if one assumes that the circuit actually has a reason for existing, namely to center the bias of an AC amplifier.
Your circuit, as drawn, servoes the drain voltage but would get very weird if any signal were present.
If the only requirement is to generate a wimpy, static Vcc/2, there are much more rational ways to do it.
John
think.
1) What Vbe error? It meets all of the OP's specifications even over temperature. 2) The long time constant can appear in the collector.Yes, but that will be fixed with item 2) above. You are an advocate of giving minimal information, so you should not find this hard to accept.
I'm willing to critique your approaches as you have mine. Please post them from the LTSpice list.
The PNP Spice model you posted.
If there's any signal swing at the drain, the b-e junction of the PNP gets all nonlinear, not to mention being damaged by zenering. An R-C between the fet drain and the emitter averages out the drain voltage for the servo. Adding filtering to the collector circuit is too late; the nonlinearities and zenering aren't fixed.
The diffamp, fed from an RC, is better.
John
think.
Okay. I looked at the circuit (using LTSpice) with what I had in mind and I don't like it AC-wise.
Please modify my/your circuit to what you think is better and post it. I'd like to see what you have in mind and how it works.
Thanks, John S
think.
As I said, R-C filter the drain voltage into the PNP emitter. Better yet, RC filter into a PNP diffamp; that avoids some ugly numbers.
Ooh, even better, use pmos fets in the diffamp.
John
think.
Ooh, even better, modify the LTSpice schematic you got from me and post it. Do you know how to do that? If not, I'll try to help you.
John S
Too busy downloading stuff to waste time troll baiting.
Some retards should be publicly stoned to death by their peers.
You are one such retard.
You should probably be able to spell a word you intend to use in your petty attempt at an insult.
A cuckoo has far more on the ball than a dung beetle level twit like you, and John Larkin.
If he had any balls, he'd post a schematic. I mean what is it... five components!
It is easy to tell who the pussy is here. Ian made the claim. Ian needs to post the proof of the claim or IAN is the asswipe he is sporting himself as currently.
The Ian boy in a man's world wants to continually provide us with an excuse for his lackings, but also wants to trivialize the character flaw itself as well.
POST a schematic, not a 'description', idiot. Your inane 'description' doesn't cover it, boy. PROVE IT. POST the netlist, pussy.
Aha! - now we see which birds flock together.
Another one too thick to understand a simple description!
Here pussy, pussy, pussy.
Have something to add? Share your thoughts — no account required.
Ask the community — no account required