Yet another N-Channel Amplifier

[Pelle]

Hi again!

By changing C10 to 22pF and increasing R37 to 330 ohm the oscillation wanished. Now I can do some real testing on the amplifier. I'll get back to you as soon as I have some measurements.

Can someone explain why reducing C10 help prevent oscillation?
Sounds a bit backwards to me...

-Pelle

[tschrama]

Since you changed two things, what did stop the oscilation? Increasing the value of R37, or lowering c10? My guess is that increasing R37 is enough, maybe use 470 Ohm, and that c10 can be 47pF.


Regards,
Thijs

[Pelle]

I first did increase R37, but that did not help alone. Then I changed C10 to 22pF and the oscillations was gone.
Sorry if I wasn't clear enough.

-Pelle

[tschrama]

No Problem, but maybe I was wrong and just decreasing C10 to 22pF is enough? 470 Ohm is a common value for gatesnoopers thouhg..

[Pelle]

I did some initial testing with a speaker and it revealed that the amplifier sounds quite good even though I don't have enough heatsink to use more than 25mA of bias.
When not loaded at +-37V it clips asymmetrical at +33.6V and -30V as expected.

Still no one that can tell me why the amp got stable by lowering the value of C10?

-Pelle

[MikeB]

Originally posted by Pelle

Still no one that can tell me why the amp got stable by lowering the value of C10?

-Pelle

Phew, feedback-caps ? I still haven't figured out all about this feedbackcap,
but understanding all about polefreq and phaseshifting is not too
easy. But proper adjusting of these caps is one of the mainkeys to
a good sounding amp. (And also stable)
This cap has more than one purpose, 1st it does phaseshifting on
the feedbacked signal, 2nd it lowers gain for higher freqs.
(Please tell me anyone if this is not correct)
So this cap needs to be adjusted, that for freqs where phasehift
at output get larger than 90°, gain is already below unitygain,
because at this point your negative feedback turns into positive
feedback, giving a nice oscillator.
But, as you don't have cdoms in your amps, this feedbackcap needs
to be small.

Mike

[efren sabio]

Hi,

Is this amp working? Because this is the 1st time I've seen a
device, Q13, is connected this way. Moreover, Q3 doesn't
have the usual compensation cap to avoid hi-freq oscillation.
But if you could actually hear a good sound coming off your
speaker then....congratulations!
I might try this topology some day. On the other hand, I
think a servo to zero the output is not always neccessary.
And what's this 'cross-conduction' about? Doesn't in class-A
this is accepted?

Peace.

[MikeB]

hi efren !

"Cross conduction" is something evil...
It is related to speed of transistors, if one transistor opens faster than
the complement closes, you suddenly get a big current through both
transistors, as both transistor are open at the same time.
This is mainly a classB (AB) problem, as transistors have to close/open.
Switch on time can be shorter than switch off time.
This results in very high currents for high frequencies and can also
create crossoverdistortions. These currents can reach values high
enough to burn up an amplifier.

Mike

[Pelle]

efren,

the amp is working even though Q13 might seem a bit odd. If you look at the output-stage you will notice that it uses only N-channel mosfets, that's why Q13 is needed. Q13 is used for inverting and biasing the signal to M19. Hope this answer your questions.
I have C10 for compensation and it seems to work so far. I've not done enough tests to verify 'total' stability of the amp.

-Pelle

[efren sabio]

Pelle,

You're right, Q13 does that! Thanks.
Now, where's the cap at the ground end of R10?

Efren