Class-A-Burn: an amp with capacitive feedback divider around MOSFET

[Wavebourn]

+36V powered, 4 Ohm loaded class A amp.

All OPAMPs are with JFET inputs.

Anyone to simulate?

Should be very fast and linear.

[atiq19]

is R21 connected right?

do you care to explain your schematic?

[Wavebourn]

Originally Posted by atiq19

is R21 connected right?

do you care to explain your schematic?

R21 provides current for shunt-regulated power source for opamps (18V) and for bias reference (6.2V).

The explanation is, MOSFETS have big enough capacitances that are non-linear. In any case MOSFET stage will have a parasitic input capacitance. What I did, I shunted this capacitances intentionally, by a capacitive voltage divider, turning MOSFET into a kind of vacuum tube triode. However, it has big input capacitance, but it is well known, very linear, and under our control.

I loaded this "Vacuumless Triode" on a counter-modulated current source and got a single-ended triode amplifier, active-loaded (that is better than reactive-loaded in terms of distortions and frequency response).

Yes, I used OPAMPS. But all of them have JFET inputs, and are used in inverting mode, with no common mode errors that are one of weak points of opamps.

[janneman]

Originally Posted by Wavebourn

R21 provides current for shunt-regulated power source for opamps (18V) and for bias reference (6.2V).

The explanation is, MOSFETS have big enough capacitances that are non-linear. In any case MOSFET stage will have a parasitic input capacitance. What I did, I shunted this capacitances intentionally, by a capacitive voltage divider, turning MOSFET into a kind of vacuum tube triode. However, it has big input capacitance, but it is well known, very linear, and under our control.

I loaded this "Vacuumless Triode" on a counter-modulated current source and got a single-ended triode amplifier, active-loaded (that is better that reactive-loaded in terms of distortions and frequency response).

Yes, I used OPAMPS. But all of them have JFET inputs, and are used in inverting mode, with no common mode errors that are one of weak points of opamps.

My friend used some of these vacuumless-triodes. Until he turned the bias up too high...

jd

[Wavebourn]

Originally Posted by janneman

My friend used some of these vacuumless-triodes. Until he turned the bias up too high...

Sure, class A needs proper cooling! Water cooling like on your picture is one of options!

[atiq19]

Hi Wavebourn,

Thanks for the explanation. So this is a triode 'indisguise' buffer? How much power do you get, given the class-A function?

I thought R21 is feedeing other ps lid (+Ve) of opamp A3?

Regards

Atiq

[Wavebourn]

It is a single ended vacuum-less triode amp. As drawn, you can get about 30W on 4 Ohm load, or 60W on 2 Ohm load. With no huge vacuum tubes, no B+ kilovolts, no filament heating. However, one of opamps must be capable of driving 160 Ohm load impedance, and a good heatsink will be needed. Source and emitter resistors mean you may parallel output devices spreading heat dissipation among them, so resistors must have higher resistance, number of devices times. Scott Wurcer gave me last year some samples of AD opamps that I may try in this amp.

[atiq19]

this can be a good buddy to drive this solid state triode buffer:

http://www.next-tube.com/articles/lineamp/LinAmp-I.pdf

[atiq19]

Wave,

what is a counter modulated CS? Could i use Nelson Pass over active CS?

atiq

[atiq19]

another companion for wavebourn ss triode buffer..