15W SE amplifier schematic

[SteveS]

I was wondering if some of the seasoned veterans around here could give me some comments on this design. What I am shooting for is about 10-15W class A and as simple as I could get it. Biasing the IRF540 pairs to 2A should get me there.

The power rail caps on the opamp were omitted for 'clarity'. Would I need to put filter networks on the bias and offset circuits even though I will be using regulated voltages (LM317/337)? Any comments or suggestions would be greatly appreciated.

Thank you,
SteveS

 

[SteveS]

I failed to mention that my preamp will make up for the minimal gain shown.

SteveS

 

[jam]

Steve,

If you take the feedback from the output you could probably dispense with the input coupling capacitor and improve your damping factor and distortion specs. The assumption here is that you will no longer need the offset adjustment portion of the circuit.

Jam

 

[tschrama]

Yep....

but I'm quite sure you will need a compensation cap if you take the feedback from the speaker output..


thijs

 

[tschrama]

There is a dual power supply? Why?

 

[SteveS]

jam:
I must admit most of the inspiration for this design is from Monarchy's SE-70 amplifier. It too does not include the OP stage in the feedback loop and from what I'v read this give a touch more openess to the sound. I'm not a big fan of using feedback, so I thought why not give it a try. The SE-70 however uses a push-pull OP stage biased to class A.


tschrama:
The reasons for the dual power supply are:

1. I wanted a clean voltage source for running the opamp and biasing (input stage and current source).

2. I have a trasformer with multiple taps to accomodate.

3. It will give me more voltage output from the opamp to compensate for Vgs.

However, if you (or anyone else) has a good idea on a circuit to provide 23V from the unregulated legs I'm all ears.

Thank you,
SteveS

 

[HarryHaller]

Class A

Untested for turn on thumps but should work. I will answer further questions that will be inevitable.

H.H.

 

[SteveS]

H.H.:

You were right about the questions.

The diodes on the Mosfet gates and the diode/resistor bias adjust I understand, but what do the 'new' components in the input and feedback circuits do?

Thanks,
SteveS

 

[HarryHaller]

ClassA

C1 1uF and R1 1Meg make the gain unity at DC and R4/R3 at audio frequencies

R6 and C3 filter RF out of input.

D1 and R9 set the bias current for output stage. Look at data sheet and or test for the gate to source voltage of mosfet at 2 amps. Zener voltage will equal Vgs+(2ampsXR9). Stay above 1 ohm for R9 for good thermal stability. Don't forget to test well before hooking up speaker. I would fuse 23 volts rails with 6 amp quick acting fuses.

H.H.

 

[SteveS]

H.H.

Thanks for the info!! You didn't happen to do any THD plots or anything, did you?

SteveS

 

[Nelson Pass]

R6 should be attached to the same - voltage as R9,
otherwise the AC difference between
the two negative rails will show up as AC current
through the current source, plus any drift between
-23 and -24 reg will cause major current value drift.

 

[PMA]

You can reduce +Vs for Q1, as there is always a voltage difference of some 4V between Gate and Source of the Q1. The circuit of current source Q2 might be a bit dangerous if done as shown. R6 bottom end should be connected to -Vs (-23V) of Q2. Then constant current will be: Ic = (Ubias - 4V)/R9.