About This "Simple Machine"

[shaan]

I recently assembled this amp and it is running good.

For the place of the CCS load resistor (R12) I have tried both the configs shown in the two pictures and both simulate and work fine with the distortion figures being identical at 1kHz@10V P-P into 8ohm.

My questions are:-

What are the advantages/disadvantages of the two CCS configs?

How can I reduce THD without adding more gain stages?

Thanks.


.

[apexaudio]

A-class

[event horizon]

Hi shaan, i can't help you with the CCs

I have a possible suggestion to get the THD lower though Try substituting R6 (10k) on the collector of the input transistor & put a simple constant current in it's place. This should increase open loop gain & thus with feedback lower the overall THD..

I reckon

[Bigun]

I see Nelson Pass Zen in this - you could do worse than review those designs.

I don't like that you have the small output bypass capacitor on a different side of the resistor than the main output capacitor - easthetically speaking.

The first circuit is not a CCS, but closer to SRPP with Q2 monitoring current flow through R12 that is flowing through output device Q4 and therefore is modulated by the signal. Look at Zen version 4 for an example.

The second circuit is a CCS load since Q2 monitors current flowing through R12 which sees only the current through output device Q3. Look at Zen version 1 to learn about this one.

Why do you want to reduce THD further - these designs (Zen amps) are known to sound wonderful. It usually helps to push more current through the output devices but you already have 2 Amps bias current.

[Bigun]

I nearly missed it - add a bootstrap capacitor from the output to the junctions of R10 and R11, this will provide a much better current supply for Q2. You might like to reduce R10 to around 1k.

[shaan]

@bigun

I am sorry it was a mistake. Correct schematic here.

[shaan]

Originally Posted by event horizon

Hi shaan, i can't help you with the CCs

I have a possible suggestion to get the THD lower though Try substituting R6 (10k) on the collector of the input transistor & put a simple constant current in it's place. This should increase open loop gain & thus with feedback lower the overall THD..

I reckon

Thanks for the suggestion.

[event horizon]

Originally Posted by Bigun

I nearly missed it - add a bootstrap capacitor from the output to the junctions of R10 and R11, this will provide a much better current supply for Q2. You might like to reduce R10 to around 1k.

Are you sure about this? I'm only asking as the upper mosfet gate won't be driven linearly. It's gate voltage is controlled by the transistor that's effectively turning it into a constant current. Bootstrap to that & the CC is no longer a CC but a very modulated one.

You did give me a good idea on how to make the amp swing right to the positive rail though (or damn near). Split R10 into two resistors & then use a bootstrap from the output & connect it to the junction of the two new resistors replacing R10. If you don't do this you'll not get enough gate voltage on the upper mosfet to get right close to the rail.

[shaan]

thanks all for the replies.

[event horizon]

You could try something else to lower THD as well. Whether you adopt a constant current instead of R6 on the collector of the input transistor or not, there would be a distinct advantage to placing a simple emitter follower where the lower power mosfet is connected.

Right now the input transistor is dealing with the current accross R6 & the current needed to charge the gate of the mosfet. If you can get rid of the gate charging from the input transistor (whilst increasing the current available to charge the gate) the whole thing will speed up & again you should see a lowering of THD.

It's a nice simple circuit, thought about the same sort of thing quite often. Seeing this makes me want to experiment

Have fun