I have been running ideas in my head for a next amp project and rereading the various materials about amp design. Reading the article my Mr. Pass about cascoding put me thinking about possible improvements that could be achieved by cascoding the power transistor in ZEN circuit. May be also cascoding the current source.....
Most of the distortion and problems remaining in this design seem to be caused by parasitic capacitances in power transistors - so cascoding should help? I know the downside is that we lose some output voltage swing capability and generato more heat.... + some more components overall.
Ergo
Most of the distortion and problems remaining in this design seem to be caused by parasitic capacitances in power transistors - so cascoding should help? I know the downside is that we lose some output voltage swing capability and generato more heat.... + some more components overall.
Ergo
I have built some cascoded Zen output stages, and have
planned some of these for later Zen Variations projects.
Ir's easy enough to do; you will find that you simply hook up
a power transistor for the cascode and feed its Gate enough
voltage to leave a minimum of 2 volts across the Gain transistor.
(You might refer to my cascoding article in Audio magazine, which
you can navigate to from www.passlabs.com )
The results are quite good, the primary effect being that it lowers
the distortion at high frequencies with the higher input and
loop impedances that are so desirable.
It does change the sound, and I would be interested in your
reaction after you have tried it.
😎
planned some of these for later Zen Variations projects.
Ir's easy enough to do; you will find that you simply hook up
a power transistor for the cascode and feed its Gate enough
voltage to leave a minimum of 2 volts across the Gain transistor.
(You might refer to my cascoding article in Audio magazine, which
you can navigate to from www.passlabs.com )
The results are quite good, the primary effect being that it lowers
the distortion at high frequencies with the higher input and
loop impedances that are so desirable.
It does change the sound, and I would be interested in your
reaction after you have tried it.
😎
jh6you said:
The philosophical simplicity is fading out...?
JH
Why? The cascode operation let the buffer out and allows less feedback ...
Also remainds me the year that I lived with the SOZ cascoded , and then the discover of the essential X ...
Indeed the cascoding would not be the step to make it more simple. I think Nelson himself has already been there and done it as simple as at all possible and also he has shown it to us.
What I think is that this step, even so more complicated, could provide us with better subjective and objective performance. At this point I think the saying "keep it as simple as possible but no simpler" applies😉
**
Anyway, would you guys keep the power transistors the same type and use the same for cascoding or would it make sense to use a smaller and at the same time faster mosfet as gain device and the high power one for cascoding. With much less voltage on the gain device it would broaden the possible choises I think.
Ergo
What I think is that this step, even so more complicated, could provide us with better subjective and objective performance. At this point I think the saying "keep it as simple as possible but no simpler" applies😉
**
Anyway, would you guys keep the power transistors the same type and use the same for cascoding or would it make sense to use a smaller and at the same time faster mosfet as gain device and the high power one for cascoding. With much less voltage on the gain device it would broaden the possible choises I think.
Ergo
ergo said:
It works beautifully with the IRF610 as gain device .
I find that the small transistor produces a *meaningful* better sound
than the IRFP 044
This would be very subjective thing. Having tried a few variations, I can say that some will find a smaller gain chip like the 610 more accurate on the top and certainly faster, but may lack other characteristics at the lower frequencies. If you are looking at multi-amplification in particular, there is a lot of room for exploration.stefanobilliani said:
Bricolo said:
Even Nelson's complicated designs are extremely simple, compared to other amps. What is the problem with inserting a cascode or active current source, if it makes the performance of a simple circuit better?
Nelson Pass said:
Yes , I find this way , turn into bi-amplification and
choose pratically for 2 Zen V2 , one of which is cascoded for the high frequencies speaker .
I'm now in process of thinking all this through in my head....
It seems that IRFP044 is a transistor choise hard to beat here. Most of the other choises seem to have much lower gm and thus lower open loop gain.... but unfortunately this is not available near Estonia (don't you guys just hate such situations)
The 2V on gain transitor seemed a bit low at first but after studing the datasheet I can see why Nelson suggests it. The characteristic is indeed very linear around 2V.
***
Now some questions.
Nelson and stefanobilliani, have you tried also cascoding the current source and if so, was this helpful?
What would be the things to consider with circuitry providing bias for cascode transistors. I have never before cascoded high current stages and there might be some additional problems to consider.
Ergo
It seems that IRFP044 is a transistor choise hard to beat here. Most of the other choises seem to have much lower gm and thus lower open loop gain.... but unfortunately this is not available near Estonia (don't you guys just hate such situations)
The 2V on gain transitor seemed a bit low at first but after studing the datasheet I can see why Nelson suggests it. The characteristic is indeed very linear around 2V.
***
Now some questions.
Nelson and stefanobilliani, have you tried also cascoding the current source and if so, was this helpful?
What would be the things to consider with circuitry providing bias for cascode transistors. I have never before cascoded high current stages and there might be some additional problems to consider.
Ergo
ergo said:
Inserting a transistor between the +V and the current source ,----is this a right understanding of your question?----is the same at the use of a source follower-voltage regulator , but ZV4 is already featured for this .
ergo said:
There are not too much things to consider I think...;
Just decrease the resistor value in the active portion of the current source if a small chip like IRF610 is used as a gain device( R16 from ZV4)
I have not seen significant improvement cascoding the current
source, but this depends on the nature of the current drive
and/or feedback on the current source circuit.
Regarding the simplicity issue, I regard this as a nice quiding
philosophy, but in reality these projects are exploratory in
nature, and we will wander around the landscape as we please.
If I were to publish the most absolutely simple amplifier I can
come up with and call it a day, what would we do next for
entertainment?
😎
source, but this depends on the nature of the current drive
and/or feedback on the current source circuit.
Regarding the simplicity issue, I regard this as a nice quiding
philosophy, but in reality these projects are exploratory in
nature, and we will wander around the landscape as we please.
If I were to publish the most absolutely simple amplifier I can
come up with and call it a day, what would we do next for
entertainment?
😎
how could you get any more simplier than the SOZ?
m.
/ is very interested to see what's in the landsape next! : )
m.
/ is very interested to see what's in the landsape next! : )
For the input , read this post from Zen lightement thread:http://www.diyaudio.com/forums/showthread.php?postid=182457#post182457 .
For the output , it lacks a resistor between V+ and the Zener.
I use 7.5v zener
🙂
For the output , it lacks a resistor between V+ and the Zener.
I use 7.5v zener
🙂
- Status
- Not open for further replies.