I am quite sure Patrick has the intelligence on how to properly bias up the circuit with diode in place.
Hopefully other people would also realise to adjust trimpots back to zero before turning on and slowly rebiasing the circuit.
Andrew,
Do you believe that the diode does indeed contribute to a higher Class potentital than what is set at stand still?
Do you believe that the diode does indeed contribute to a higher Class potentital than what is set at stand still?
> it is very quick and easy to wire a diode into your circuit.
How do you now I have not tried ?
🙂
Have you tried what I proposed ?
Patrick
How do you now I have not tried ?
🙂
Have you tried what I proposed ?
Patrick
no.
I think the ClassA output remains the same.
I think that the amp needs the biasing to be redesigned such that the diodes do not pass during the quiescent condition.
I believe that if the diodes start passing during the periods that output current is between Ibias current value and two times Ibias current value, that this will introduce distortions that are avoided by ordinary ClassA operation.
I believe the diodes should only pass when the output current exceeds the maximum ClassA current.
I think that ClassA operation will be compromised if the "diodes turn on" while the output current is still in the ClassA region.
I have no idea what that "diode turn on" current level ref. maximum ClassA current should be. Could it be @ 0.9*Ibias, or 1.7*Ibias, or 2.5*Ibias?
It is because of all these imagined concerns and my lack of knowledge that I keep asking the same question.
This implies that the diodes have already turned on at the normally recommended Ibias level of ~600mA, ie. zero output current level. Otherwise there would be no need to turn down the bias adjusters.
Last edited:
I have not tried anything as yet. I just got back into this after over a 1 year break.
It is easy enough for me to try for myself. I will do it eventually.
Just want to finish my speakers first.
I have too many ideas and if I don't focus on one of them I won't get anything done.
It is easy enough for me to try for myself. I will do it eventually.
Just want to finish my speakers first.
I have too many ideas and if I don't focus on one of them I won't get anything done.
There are so many variables Andrew, the only way is to just try for yourself.
You guys really know how to get me off track.
If you guys want me to do it in the lab I will do it with real performance measurements.
You're just proving how bad my memory is.
I don't recall any of those 5 posts.
February ! I was in Morocco. Did you lot explore this without my permission?
Can't be that, I posted on 10th and 17th Feb.
I don't recall any of those 5 posts.
February ! I was in Morocco. Did you lot explore this without my permission?
Can't be that, I posted on 10th and 17th Feb.
Last edited:
I guess one can say I have a prejudice against the diode. I do.
But it is not unfounded, as you can see from post #71, 72.
Real measurements ain't too different. 😉
Patrick
But it is not unfounded, as you can see from post #71, 72.
Real measurements ain't too different. 😉
Patrick
Last edited:
You guys need to think outside the box.
Instead of thinking I don't like the idea of this. Why not think how can I make this work.
There are already so many other non-linearities without the diodes.
I reckon it is a drop in the ocean.
Nelson is not going to suggest something that will make the amp sound worse.
It's just a puzzle to solve.
Instead of thinking I don't like the idea of this. Why not think how can I make this work.
There are already so many other non-linearities without the diodes.
I reckon it is a drop in the ocean.
Nelson is not going to suggest something that will make the amp sound worse.
It's just a puzzle to solve.
Please go ahead. I am not stopping you by any means.
I personally have no desire to take on the challenge of adding an exponential device to a quadratic one.
Patrick
I personally have no desire to take on the challenge of adding an exponential device to a quadratic one.
Patrick
Since we are on the topic, I have another suggestion for those who want challenges.
Forget about the diode and use a 22000uF cap to parallel the source resistor.
6.3V is probably enough voltage.
Takes you 10 minutes to try ?
And ask the tube guys what they think.
😉
Patrick
Forget about the diode and use a 22000uF cap to parallel the source resistor.
6.3V is probably enough voltage.
Takes you 10 minutes to try ?
And ask the tube guys what they think.
😉
Patrick
Last edited:
THanks. That confirms my thoughts about what is happening here. I am shocked you are willing to even speak of tubes. Dinosaur lovers.
OOPs
post484
It should read:
This implies that the diodes have already turned on at the normally recommended bias level of ~600mVrs, ie. zero output current level.
post484
Most will have realised that my 800mA should have read 600mVrs, the bias voltage across the source resistor.
It should read:
This implies that the diodes have already turned on at the normally recommended bias level of ~600mVrs, ie. zero output current level.
I think to view this properly, you have to consider what the degeneration resistor is doing, how what it is doing is affected by the AC signal through it, and what if the effect of this AC signal was being mitigated by either a diode, or a cap asa Patrick suggested. I ran headlong into this subject when i was trying to figure out Juma's preamp version of the F5. Got me looking into degeneration and what is actually happening.
Re #493, you can read all about the cap in Horowitz & Hill 2nd Edition Section 2.13 "Bypassed emitter resistor".
Or John Linsley Hood 1997 Valve & Transistor Amps Fig. 3.3c & associated text.
A cap has an advantage over the diode in that it only acts for AC.
So the DC bias is totally determined by the Rsource itself, as it should.
For AC signal, the cap is always effective as long as it is large enough in value to guarantee low corner frequency.
Not the case for the diode.
Patrick
Or John Linsley Hood 1997 Valve & Transistor Amps Fig. 3.3c & associated text.
A cap has an advantage over the diode in that it only acts for AC.
So the DC bias is totally determined by the Rsource itself, as it should.
For AC signal, the cap is always effective as long as it is large enough in value to guarantee low corner frequency.
Not the case for the diode.
Patrick
Last edited:
Patrick
What about the biasing thermistor? There is plenty of non-linearity there too.
Are you also suggesting the patented Aleph Current Source is also a stupid idea?
What about a current source paralleled on one of the mosfets of a push pull output stage used on an Amp worth $65,000?
There is also some weird crap going on there too.
I notice PASS LABS are still in business.
According to your thinking they are incompetent, amplifier builders.
That may not be your intent, but your ideas and thinking are definitely implying that view.
There are plenty of text books that have opposing views on things.
There are also plenty of authors of text books who don't make amplifiers for a living.
Come on Patrick.
I understand your concerns about the diode, but I'm not giving in that easy.
What about the biasing thermistor? There is plenty of non-linearity there too.
Are you also suggesting the patented Aleph Current Source is also a stupid idea?
What about a current source paralleled on one of the mosfets of a push pull output stage used on an Amp worth $65,000?
There is also some weird crap going on there too.
I notice PASS LABS are still in business.
According to your thinking they are incompetent, amplifier builders.
That may not be your intent, but your ideas and thinking are definitely implying that view.
There are plenty of text books that have opposing views on things.
There are also plenty of authors of text books who don't make amplifiers for a living.
Come on Patrick.
I understand your concerns about the diode, but I'm not giving in that easy.
Hey, Thats what I said, well not really, but thats what I was eluding to. Melon Head, the ACS is nothing special. Anybody could have come up with it.....after looking at!
- Status
- Not open for further replies.