Have you ever succeed in buiding a input stage with current mirror?

[thanh]

hi thanh !

I did some more checkings on the asymamp i posted here, it's kind
of useless. When clipping to + , it completely breaks down, seems
that currentmirror is not too easy. I could remove the buffer, but
this is bad for distortions. BTW, it's nearly identical to the blameless,
not so blameless at all... I would never build an amp showing these
symptoms !


So i will continue finetuning my symamp first.

Mike

mike! I don't understand "when clipping to +" . With 700mV pulse signal, output voltage only include "positive part" .
Yesterday, when I turned off the neonlight ,my amp also broke down . V- failured

[thanh]

mike! amp which I have just touch upon was modified with 3.3k collectorresistor in diffstage and increased current source .Orcad say thay open-loop gain is 55 ,not 55.000 . What fool orcad is

[MikeB]

hi lumanauw !
hmm, the output with my circuit is not inverting...
I am not sure how you want to use the currentmirror, can you show
some schematic ?
I use complementary diffamp, resistorloaded, and then symetrical
folded cascode. As i use negative side of diffamp, output is not
inverting. I can post schematic later.

hi thanh !
Not nice to hear you blew your amp... Did the PSU fail ?
What kind of amp was this ?
With clipping to +, i meant when reaching max positive voltageswing
where it should clamp, it breaks down to 0 volt until level drops again
below clippingthreshold.
I don't understand what you mean with openloopgain = 55 ?
Closedloop is 45 with my amp. So openloop should be much higher ?
Did you adjust DC for measuring with openloop ?

3.3k collectorresistor + increased currentsource ? That must be
a giant currentbias in vas ?

Mike

[MikeB]

Hi lumanauw !

As promised the schematic...

Mike

[lumanauw]

Hi, Mike,
I really forgot, that the signal can be taken from left or right leg of differetial.

Your design is 2 stages class AB! and it has low OL gain too. Should sound good. When will you finish the prototype?

What I intended to use is just like the "benny.jpg" in post #175. Current mirror from one leg of differential as VAS. What do you think?

[thanh]

hi thanh !
Not nice to hear you blew your amp... Did the PSU fail ?
What kind of amp was this ?
With clipping to +, i meant when reaching max positive voltageswing
where it should clamp, it breaks down to 0 volt until level drops again
below clippingthreshold.
I don't understand what you mean with openloopgain = 55 ?
Closedloop is 45 with my amp. So openloop should be much higher ?
Did you adjust DC for measuring with openloop ?

3.3k collectorresistor + increased currentsource ? That must be
a giant currentbias in vas ?

Mike

Thanks your explanation! I didn't adjust DC when I measure openloop gain because I think that adjusting or without adjusting is not a problem. OPen-loop gain is 55 and close-loop gain is 33

[thanh]

mikeb! my topo is similar your topo ,cfp inputstage +darlington VAs +EF output

[MikeB]

Originally posted by thanh

Thanks your explanation! I didn't adjust DC when I measure openloop gain because I think that adjusting or without adjusting is not a problem. OPen-loop gain is 55 and close-loop gain is 33

hi thanh !
I makes a big difference, i once forgot it, and suddenly had a OL-gain
of 3...
If the amp is stuck in clipping due to large dc-offset, it can't create a
signal !
Maybe forgot to put the open -input to gnd ?

Is your amp asym or sym ?

I had no reliability problems with my amp, the topology should be safe...

to lumanauw: I think i need a few days for the prototype.
I need to check your circuit, can't say much yet...

Mike

[lumanauw]

Hi, Mike,
I sketch the idea. The VAS (Level Shifter) has no current gain, just the same as one leg of differential.
Will it behave the same as if I use folded cascode, or it behave more like ordinary 3 stage power amp?

[MikeB]

hi !
I think, this circuit is a normal 3 stage, just one diode to reduce
gain in the diffamp. This should behave normal like a 3stage.
You have 2 active bjts in the vgainstages,the diode not counted.
You ensure with this circuit that current in vas is identical to
current in diffleg.
With the folded cascode i posted, you have one 1 active bjt in
the frontendloop. (per side, top/bottom)

Now i understand why i didn't understand you...
But its still a nice circuit !

Mike