nobody special said:The only thing to remember is that any resistance between the sources of the differential pair will mess up the symmetry of the outputs
As a concept, you cannot eliminate resistance between the
Sources, as the transconductance of the devices themselves
gives rise to an apparent resistance. Resistance placed
between the Sources lowers the open loop gain of the circuit.
Nelson Pass said:
As a concept, you cannot eliminate resistance between the
Sources, as the transconductance of the devices themselves
gives rise to an apparent resistance. Resistance placed
between the Sources lowers the open loop gain of the circuit.
Hi Mr.Pass,
You mean we must keep the resistors R3 and R4,right?
thanks
Leo
Attachments
R3 and R4 serve to degenerate the MOSFETs in the differential. Depending on your frame of mind, you can choose to regard this as feedback...or not...whichever suits your mood. The IRF MOSFETs have very high transconductance compared to something like the 2SK170 or the 2SK246. In principle, you can re-fiddle the circuit to offset that gain. One way to do so would be to lower the value of the load resistors--but then you have to increase the bias in order to assure something approaching equal voltage swing. That increased bias will mean more dissipation in the load resistors, which is in turn offset somewhat by the lower resistance. You may need more robust load resistors to handle the heat (I haven't done any calculations--this is all seat of my pants). The load on the power supply is also increased, which could lead to a bigger transformer. Then your parts cost increases. And so on and so on. In other words, it's all a balancing act. Nelson has chosen a balance that works well for him, but you could very well get crazy and use some of those aluminum body power resistors, put heavy heatsinks on the MOSFETs, throw a monster power supply into the mix, and end up with a very different flavor from the stock circuit.
Me? I kinda like the idea of lower resistance loads, which in turn gives you wider bandwidth, but nothing is free and you'll have to be careful about just throwing random parts into the circuit without watching the thing on a scope to make sure that you haven't shot yourself in the foot. Something I do every so often, just to keep myself humble.
(Humble? Oh dear Gussie, the boy is gonna get zorched by a lightning bolt for saying that.)
Okay, how about I say it keeps me from getting too far ahead of myself?
(I suddenly feel kinda prickly all over. Didn't I read somewhere that just before lightning strikes, you feel like you've got ants crawling all over y--
Me? I kinda like the idea of lower resistance loads, which in turn gives you wider bandwidth, but nothing is free and you'll have to be careful about just throwing random parts into the circuit without watching the thing on a scope to make sure that you haven't shot yourself in the foot. Something I do every so often, just to keep myself humble.
(Humble? Oh dear Gussie, the boy is gonna get zorched by a lightning bolt for saying that.)
Okay, how about I say it keeps me from getting too far ahead of myself?
(I suddenly feel kinda prickly all over. Didn't I read somewhere that just before lightning strikes, you feel like you've got ants crawling all over y--
You mean we must keep the resistors R3 and R4,right?
I am using a dual-pot (500 ohms) in conjunction with R3 and R4 (10 ohms each). It makes an effective gain adjustment for the modified BOSOZ circuit.
JJ
GRollins said:R3 and R4 serve to degenerate the MOSFETs in the differential. ....
Hi Grey,Another a stupid question.Could you tell me how to calculate the circuitry input/output impedance which in post#26.
thanks.
Leo
And the circuit....
Hi sianturi,
I am intersted to realize a Unbalanced to Balanced converter, and I think (and hope) that this is one of the best that I have seen. Can you gently confirm this choise?
And again, what are the values of R3L, R4L, R5L and R6L?
Many thanks for your help. Best regards.
Davide.
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.
- Home
- Amplifiers
- Pass Labs
- BoSoZ with CCS -- need sugestion...