Is there any advantage to using mosfets over bipolar in a class AB low voltage (+/-40V) design?
Haarder to bias and significant headroom limitation.
Haarder to bias and significant headroom limitation.
I suppose there are the more desirable MOSFETs -- especially having a lower threshold.
I got one designed in simulation but headroom limitation in truly sad and took some real effort to get a good bias technique going well
I got one designed in simulation but headroom limitation in truly sad and took some real effort to get a good bias technique going well
Read my article on Class AB Biasing.
From the point of view of distortion in class AB, BJTs are better. This is because the Vbe of a BJT changes much less with Ic than a MOSFETs' Vgs changes with Id. In each case, the relationship is non-linear.
IMO, at 40V, both BJTs and MOSFETs benefit from having higher-voltage rails for the drivers.
Ed
From the point of view of distortion in class AB, BJTs are better. This is because the Vbe of a BJT changes much less with Ic than a MOSFETs' Vgs changes with Id. In each case, the relationship is non-linear.
IMO, at 40V, both BJTs and MOSFETs benefit from having higher-voltage rails for the drivers.
Ed
Good article
I am using current sensing on one side when it is off the adjust bias for mosfet and works.
If power stage is open loop then i suppose you get more tube-like clipping as a possible benefit for guitar
I am using current sensing on one side when it is off the adjust bias for mosfet and works.
If power stage is open loop then i suppose you get more tube-like clipping as a possible benefit for guitar
Thanks!
This board has a number of threads on auto-bias. On class AB, fast auto-bias works but slow auto-bias is nearly impossible (i.e. no-one has gotten satisfactory results as far as I can tell).
Ed
This board has a number of threads on auto-bias. On class AB, fast auto-bias works but slow auto-bias is nearly impossible (i.e. no-one has gotten satisfactory results as far as I can tell).
Ed
When you use them in one very specific topology. There is no law stating that output transistors have to be connected as complementary emitter followers or complementary source followers and that they have to be driven from a low (open-loop) impedance.
That waveform looks interesting for guitar. Could you explain roughly how the circuit works? I am not familiar with this kind of circuit, but it makes me curious.
My variant is "Audio power with a new loop", Electronics World February 1996, pages 140...143, https://worldradiohistory.com/UK/Wireless-World/90s/Electronics-World-1996-02-S-OCR.pdf
Of course, but emitter followers and source followers are the most commonly-used topologies for the output stage.
Making the feedback loop fast enough for audio frequencies will result in the output transistors running in class A.
Ed
Indeed. If you had written 'For a class-AB complementary emitter or source follower topology driven from a low open-loop impedance, BJTs are better from the point of view of distortion', I would have agreed, but the way you phased it in post #4, it could be mistaken for a topology-independent conclusion - if someone would read it without looking at the link, that is.
I doubt if that is even possible with my amplifier. It was never meant to be a guitar amplifier.
I think we are nit-picking. The context is the OP's circuit, a source follower.
Okay, I see why: the sensing is shut off during positive halves. That assumes the load is not reactive.
Ed