Single or dual differential?

[capslock]

Originally posted by lumanauw
...

This design has single differential, and pushpull VAS. The levelshifter is done by folded cascode. (leading to low openloop gain) The output of right differential is converted to drive lower VAS transistor with current mirrors. This is from AD817 opamp.

Can anyone help, if I want to make this design an audio power amp, what is the bias current for Differential, Folded cascode, and final transistor? The supply is about +/-50V.

Where are the VAS transistors in this design? This is a folded cascode (all the gain is in this one stage!), followed by a complementary buffer. I don't think a complementary buffer will work too well as an output stage in a power amp.

[Ultima Thule]

Frank,

How can a cascoded JFET have higher gain then a non cascoded JFET?

Or am I missing something...

[lukio]

Hi, lumanaw...

You asked for AD 817 version for power amp application. I've built that and tested succesfully. It based on differential folded cascode with local feedback. It has about 100 W (8ohm). The result make me happy. If you interest you may contact me..


cheeers,

Lukio

[Jocko Homo]

Cascode increases gain????????????? Really????

You need a remedial course, proctored by someone as mean as me. [joke]


Jocko

[john curl]

Folks, this is getting you nowhere. Speculation as to how circuits work is not that same as understanding how circuits work.
First, what point is there to go to extremes to reduce the noise of the power amp? It is the last chain in the link and the vast bulk of the noise in your audio system will come from the preamp and earlier sources.
Second, using a very large, expensive, fet on the input is great for MC inputs, but the inherently high non-linear capacitance in the large area input fets will give you more high frequency harmonic distortion than a more optimum geormetry like the 2SK389. Noise difference? Almost unmeasurable in a power amp design.
Cascoding would reduce the distortion, but not remove it completely.
Paralleling FET's is what I do for MC inputs, but it only makes the situation worse for distortion with a power amp, or even a line preamp.
Think it through, look it up on the internet, or give thought of what audio designers have already stated on this thread and others.

[Fred Dieckmann]

"First, what point is there to go to extremes to reduce the noise of the power amp? It is the last chain in the link and the vast bulk of the noise in your audio system will come from the preamp and earlier sources."

This from the guy who lectured me on the noise contributions of gate resistors in line level circuits? (Humor attempt)

Actually I agree, people tend focus on parameters that are relatively unimportant at the expense of those that are very important in a particular design. Unfortunately, knowing which is which comes with a lot of experience. That is why its nice to have Mr. Curl around for the insight picked up in a long and successful career as audio designer.

Thanks John.

[fdegrove]

Hi,

How can a cascoded JFET have higher gain then a non cascoded JFET?

Sorry about the confusion....
A cascode with a triode was what was meant.

You need a remedial course, proctored by someone as mean as me. [joke]

I'll take the Gamble....

[jam]

Frank,

I guess you just got spanked..........................

Regards,
Jam

P.S. This message has been brought to you by the SOLID STATE LOVERS SOCIETY

[john curl]

Folks, I hope we can still keep this thread going. I am not trying to stop anyone from asking questions or giving new input. However, if an experienced audio designer warns you about something, and this is not contradicted by another experienced audio designer, then, usually, the warning is valid and useful.

[lukio]

Hi every body...

Here is the schematic of my folded cascode with local feedback..

The specifications are :

Bandwidth – 3 db = 3 Hz -750 kHz
Bandwidth – 3 db / filter = 300 kHz
THD+N = 0.003 % (3-32 kHz)
Slew Rate = 170 V/us
Rise time step response = 1.3 us
Settling time (0.01%) = 5 us
Step response = Perfect w/o overshoot and ringing
PSRR = +-- 70 dB (10-10 kHz)
Equivalent input noise = 4.3 uV/sqrt Hz
Power = 96 Watt 8 ohm
160 Watt 4 ohm
Gain = 27 X ( 28 dB)
S/N ratio = >100 dB
Input sensitivity = 1 V
Input impedance = +- 100 kOhm
Output impedance = 0.15 ohm
Max.output current = 4 A
Supply voltage = + - 40 Volt unregulated
DC offset = Max. -35 mV (Adjustable)