quiescent current in audio MOSFET class AB output stage

[Mac]

What is the best way to set the quiescent current in a class AB MOSFET
output stage? I plan to use an LM833 for the gain stage with feedback from
the output back to the LM833, so I am not worried about offset drift or
anything like that. I am a little worried about the quiescent current
getting out of hand from a heat dissipation perspective.

I plan to use +/-15 volt rails and am shooting for 10 watts into 8 ohms.

Is it safe to just use a trim pot in the biasing circuit, to set the
quiescent current or is it liable to drift a lot if I do that?

How about putting small source resistors in combination with the hand
trimming approach?

What I would really like is a way to get a reasonably bounded quiescent
current by design without trimming. Again, I don't care if the current
drifts a bit, as long as the power consumption doesn't get out of hand.
I'm thinking that 200 mA is a reasonable upper limit.

Best regards,

Mac

 

[Ban]

Mac wrote:
|| What is the best way to set the quiescent current in a class AB
|| MOSFET output stage? I plan to use an LM833 for the gain stage with
|| feedback from the output back to the LM833, so I am not worried
|| about offset drift or anything like that. I am a little worried
|| about the quiescent current getting out of hand from a heat
|| dissipation perspective.
||
|| I plan to use +/-15 volt rails and am shooting for 10 watts into 8
|| ohms.
||
|| Is it safe to just use a trim pot in the biasing circuit, to set the
|| quiescent current or is it liable to drift a lot if I do that?
||
|| How about putting small source resistors in combination with the hand
|| trimming approach?
||
|| What I would really like is a way to get a reasonably bounded
|| quiescent current by design without trimming. Again, I don't care if
|| the current drifts a bit, as long as the power consumption doesn't
|| get out of hand. I'm thinking that 200 mA is a reasonable upper
|| limit.
||
|| Best regards,
||
|| Mac

Mac,
200mA is way too much, you will create more heat than with a loud music
program. I would also advise you to take ready made amplifier ICs, which
have more chance of functioning than your selfmade circuits. ;-o
In this power range you should consider class-D chips, which considerably
cut down on power supply and heat sinking requirements.
Another thing, don't use regulated supply rails, it is an unnessesary
wastage of power and doesn't increase music fidelity. (This may not be true
for certain digital amplifier ICs).
The LM833 is utterly unsuitable to drive a MOS output stage, with a 2k load
the O/P voltage swing is only +/-10V. You should study the data sheet a bit
more and read some books about designing power amplifiers. The board layout
is another art that comes with experience and is absolutely essential for a
good sounding amp.
So better start with an integrated circuit, it is already challenging.

 

[default]

What is the best way to set the quiescent current in a class AB MOSFET
output stage? I plan to use an LM833 for the gain stage with feedback from
the output back to the LM833, so I am not worried about offset drift or
anything like that. I am a little worried about the quiescent current
getting out of hand from a heat dissipation perspective.

I plan to use +/-15 volt rails and am shooting for 10 watts into 8 ohms.

Is it safe to just use a trim pot in the biasing circuit, to set the
quiescent current or is it liable to drift a lot if I do that?

How about putting small source resistors in combination with the hand
trimming approach?

What I would really like is a way to get a reasonably bounded quiescent
current by design without trimming. Again, I don't care if the current
drifts a bit, as long as the power consumption doesn't get out of hand.
I'm thinking that 200 mA is a reasonable upper limit.

Best regards,

Mac

Don't sweat the drift with mosfets, their current gain goes down as
heat goes up (backwards from bipolar transistors).

If you have any load instability problems in a mosfet output amp they
will most often occur when the amp is cold and gain is high. (the
reason some mosfets sound better after they run for a time)

6 watts is a little high but should sound better for it - If you have
the heat sink and power, go for it. Or set it cold and see what it is
when heated.

 

[Winfield Hill]

default wrote...

Don't sweat the drift with mosfets, their current gain goes
down as heat goes up (backwards from bipolar transistors).

When used in linear circuits the implication is exactly
wrong. In fact the drain current of a typical power MOSFET
with fixed gate voltage goes up with temperature, creating
high-temperature degradation or possible thermal runaway,
just like bipolar transistors. An exception to this is the
special lateral power MOSFETs made by Hitachi / Renesas

Thanks,
- Win

whill_at_picovolt-dot-com

 

[John Woodgate]

I read in sci.electronics.design that default <R75/[email protected]>

Don't sweat the drift with mosfets, their current gain goes down as heat
goes up (backwards from bipolar transistors).

NO! Only SOME of them do that. MOST don't.

 

[John Fields]

default wrote...

When used in linear circuits the implication is exactly
wrong. In fact the drain current of a typical power MOSFET
with fixed gate voltage goes up with temperature, creating
high-temperature degradation or possible thermal runaway,
just like bipolar transistors. An exception to this is the
special lateral power MOSFETs made by Hitachi / Renesas

 

[default]

I read in sci.electronics.design that default <R75/[email protected]>


NO! Only SOME of them do that. MOST don't.

Really? Thanks. That's news to me.

No source or drain swamping resistors used in multiple parallel
transistors used in audio out stages - wouldn't that shoot down the
idea that they suffer from thermal runaway?

I'm going to do some looking around before I buy the thermal runaway
theory.

 

[Walter Harley]

No source or drain swamping resistors used in multiple parallel
transistors used in audio out stages - wouldn't that shoot down the
idea that they suffer from thermal runaway?

Not if you've repaired a bunch of amps that were designed that way, it
wouldn't.

 

[default]

I read in sci.electronics.design that default <R75/[email protected]>


NO! Only SOME of them do that. MOST don't.

I stand corrected.

"Having a negative temperature coefficient the threshold voltage falls
about 10% for each 45 degrees C rise in junction temperature."

Motorola "Power Mosfet Transistor Data"

OK you win.

During the periods outside of the switching transitions, the
current in a parallel group of HEXFET Power MOSFETs
will distribute itself in the individual devices in inverse
proportion to their ON resistance. The device with the
lowest ON resistance will carry the highest current. This
will, to an extent, be self-compensating, because the power
loss in this device will be the highest. It will run hottest,
and the increase in ON resistance due to heating will be
more than that of the other devices, which will tend to
equalize the current.

The reason why they can be run in parallel while eliminating emitter
swamping resistors.

I learned two things -

 

[Walter Harley]

What is the best way to set the quiescent current in a class AB MOSFET
output stage?

Douglas Self's book "Audio Power Amplifier Design Handbook" goes into
biasing arrangements at considerable length; however, he doesn't spend much
time on MOSFETs since he considers them to be pointless for audio. (He also
goes into the reasons for that, debunking the usual rationale behind using
MOSFETs in power amps.)

Most amps that I have worked on (MOSFET or BJT) use the same technique
that's already been recommended: a Vbe multiplier and a trim pot. There
just does not seem to be sufficient advantage to doing it any other way,
unless you expect to be making so many of these that the expense of setting
the trim pot properly in the factory is prohibitive. In other words:
sonically, the Vbe technique is quite good; its only disadvantage is that it
requires an adjustment whenever the transistors are replaced.

 

[default]

default wrote...

When used in linear circuits the implication is exactly
wrong. In fact the drain current of a typical power MOSFET
with fixed gate voltage goes up with temperature, creating
high-temperature degradation or possible thermal runaway,
just like bipolar transistors. An exception to this is the
special lateral power MOSFETs made by Hitachi / Renesas

Thanks,
- Win

whill_at_picovolt-dot-com

Well that explains what I'm seeing. I chose Hitachi output mosfets,
but only because they had a complementary pair in TO-3 cases.

 

[Winfield Hill]

John Fields wrote...

Would you post some references, please?

Well, I won't post an AoE reference, because we got it wrong!
OK, at least we were misleading, because the negative tempco
only occurs for high Vgs, i.e. for switching applications (so
it's OK to parallel switching FETs).

Study the Transfer Characteristics plots in the data sheet of
your favorite amplifier FET, and unless it's a lateral type,
you'll see what I'm talking about. You'll see crossing curves
at two or three temperatures. For Vgs below the crossover gate
voltage the tempco is positive, above it's negative. OK, now
notice the high drain current at crossover. Consider what the
power dissipation would be for a linear power-amp FET with say
50V operating at that current... Ouch! Impossible!

Thanks,
- Win

whill_at_picovolt-dot-com

 

[Winfield Hill]

default wrote...

Well that explains what I'm seeing. I chose Hitachi output mosfets,
but only because they had a complementary pair in TO-3 cases.

You probably made a good choice. What are the part numbers?

Thanks,
- Win

whill_at_picovolt-dot-com

 

[Winfield Hill]

During the periods outside of the switching transitions, the
This paragraph is true but BE CAREFUL, it only applies to FETs
used as switches with a high applied gate-to-source voltage.
It DOES NOT APPLY to ordinary power MOSFETs used in amplifiers!
See my answer to John Fields in this thread.

Thanks,
- Win

whill_at_picovolt-dot-com

 

[John Woodgate]

I read in sci.electronics.design that John Fields <jfields@austininstrum
ents.com> wrote (in <[email protected]>) about
'quiescent current in audio MOSFET class AB output stage', on Wed, 10
Dec 2003:

Would you post some references, please?

Is this what you want? (;-)

"George has served his term in a responsible position without actually
precipitating Armageddon. I have no hesitation in recommending him to
your advertised post of high-school janitor."

If so, I can contribute a few more.

 

[John Woodgate]

I read in sci.electronics.design that default <R75/[email protected]>

I'm going to do some looking around before I buy the thermal runaway
theory.

Just look at other articles in this thread. If it doesn't say 'Hitachi',
it may well run away. Even if it does, it might.

 

[Mac]

Mac wrote:
|| What is the best way to set the quiescent current in a class AB
|| MOSFET output stage? I plan to use an LM833 for the gain stage with
|| feedback from the output back to the LM833, so I am not worried
|| about offset drift or anything like that. I am a little worried
|| about the quiescent current getting out of hand from a heat
|| dissipation perspective.
||
|| I plan to use +/-15 volt rails and am shooting for 10 watts into 8
|| ohms.
||
|| Is it safe to just use a trim pot in the biasing circuit, to set the
|| quiescent current or is it liable to drift a lot if I do that?
||
|| How about putting small source resistors in combination with the hand
|| trimming approach?
||
|| What I would really like is a way to get a reasonably bounded
|| quiescent current by design without trimming. Again, I don't care if
|| the current drifts a bit, as long as the power consumption doesn't
|| get out of hand. I'm thinking that 200 mA is a reasonable upper
|| limit.
||
|| Best regards,
||
|| Mac

Mac,
200mA is way too much, you will create more heat than with a loud music
program.

0.2A * 15V = 3 watts per transistor. OK, we'll let the upper limit be 100
mA, or even 50.

I would also advise you to take ready made amplifier ICs, which
have more chance of functioning than your selfmade circuits. ;-o

Yeah, but my goal isn't to have an amplifier, my goal is to make one. ;-)

In this power range you should consider class-D chips, which considerably
cut down on power supply and heat sinking requirements.

Maybe some day I will build a class D amp. But that is not a goal right
now.

Another thing, don't use regulated supply rails, it is an unnessesary
wastage of power and doesn't increase music fidelity. (This may not be true
for certain digital amplifier ICs).

OK. I just need to make sure I don't violate any absolute maxima.

The LM833 is utterly unsuitable to drive a MOS output stage, with a 2k load
the O/P voltage swing is only +/-10V. You should study the data sheet a bit
more and read some books about designing power amplifiers. The board layout
is another art that comes with experience and is absolutely essential for a
good sounding amp.

Why is it utterly unsuitable? Just because of the voltage swing? In any
event, I can easily change my mind on the op-amp I use if I determine that
there is a real problem. It will certainly be hard to get 10 Watts out of
10 Volts of swing. ;-)

As for books, I'd be interested in recommedations. I only have general
books, including AoE and Sedra and Smith's "Microelectronic circuits." I
think it is the 4th edition. Neither one goes into great detail about
biasing power MOSFET's for class AB. In fact, AoE doesn't seem to address
it at all.

I should probably also mention that I am an electrical engineer and have
designed half a dozen or so real circuit boards. (This proposed design is
just for fun, at home on my own time.) The other boards I have done
include complex digital (with switching supplies) boards and a relatively
simple signal amplifier (op-amp) board for signals up to UHF. So I am not
totally unprepared to take on a board layout.

So better start with an integrated circuit, it is already challenging.

Thanks for the advice, but this holds no appeal for me. And, as I said,
I'm doing this for fun.

best regards,

Mac

 

[Mac]

Douglas Self's book "Audio Power Amplifier Design Handbook" goes into
biasing arrangements at considerable length; however, he doesn't spend much
time on MOSFETs since he considers them to be pointless for audio. (He also
goes into the reasons for that, debunking the usual rationale behind using
MOSFETs in power amps.)

My electronic design textbook doesn't go into class AB biasing for power
MOSFET's too much, either. It shows one or two examples, one using a Vbe
multiplier, and another using two Vbe multipliers.

Most amps that I have worked on (MOSFET or BJT) use the same technique
that's already been recommended: a Vbe multiplier and a trim pot. There
just does not seem to be sufficient advantage to doing it any other way,
unless you expect to be making so many of these that the expense of setting
the trim pot properly in the factory is prohibitive.

Not in my wildest dreams. ;-)

In other words:
sonically, the Vbe technique is quite good; its only disadvantage is that it
requires an adjustment whenever the transistors are replaced.

Thanks a lot Walter. This is fairly convincing. I will probably use this
technique. I'll re-post when I think I've got it right.

Naturally, I'll continue to think about ways that it could be done without
trimming.

regards,
Mac

 

[Walter Harley]

As for books, I'd be interested in recommedations. I only have general
books, including AoE and Sedra and Smith's "Microelectronic circuits." I
think it is the 4th edition. Neither one goes into great detail about
biasing power MOSFET's for class AB. In fact, AoE doesn't seem to address
it at all.

Douglas Self's "Handbook of Audio Power Amplifier Design". It will really
open your eyes to a lot of issues you might not have thought about (at
least, it did mine), and at the same time it is very concrete, readable, and
contains schematics that you can build yourself.

Although, as I said elsewhere, it doesn't talk much about MOSFETs except to
assert their inferiority to BJT's. But even if you're committed to building
with MOSFETs, I would recommend it; many of the issues are the same.

-w

 

[Mac]

Douglas Self's "Handbook of Audio Power Amplifier Design". It will really
open your eyes to a lot of issues you might not have thought about (at
least, it did mine), and at the same time it is very concrete, readable, and
contains schematics that you can build yourself.

Although, as I said elsewhere, it doesn't talk much about MOSFETs except to
assert their inferiority to BJT's. But even if you're committed to building
with MOSFETs, I would recommend it; many of the issues are the same.

-w

Thanks

Mac