Pretty much the same efficiency as everything else. The
value of the circlotron in my mind is the ability to get good
direct coupled push-pull operation from matched parts of the
same polarity. After spending more than usual on the extra
supply components, that is.
An ignorant who assumes to be ignorant is actually one step away from being one.
I just realize that this could be easily misread. I meant: "...is actually quite far away from being one."Sorry, the language barrier!
This is probably an old hat in solid state push pull circuitry: In my F5 project I selected the IRFP240s and IRFP9240s for low distortion (a friend helped me, he owns a distortion analyzer - lucky fellow!), because I wanted to try out a few dB less feedback in the circuit. The prints of the distortion curves between the two different Fets were not that similar than I had expected.
I wondered, if this inequality might slightly damage the sound. Maybe in a way of a marginally less relaxed presentation.
Don't get me wrong, I don't hear the fleas coughing (as we say in Germany), it's just speculation.
Could this be another advantage of a circlotron?
P.S. I consider a circlotron as a three-dimensional Möbius strip, but maybe I'm just too romantic...
Nelson,
Would it be possible to use the Semisouth silicon carbide high voltage jfet in the frontend of the amplifier to allow higher overall rail voltages. Have they got the required specs for the differentials and drive units. If the Semisouth part replaced the Loveltech the amp would be considerable less fragile. You could then dispense with the cascode element and simplifiy the circuit. I need very simple these days. The datasheet states very low initial gate voltage and very high effeciency.
I doubt I will ever figure out how all of this stuff works. However, most of my amps turn out smokeless. That is most not all. I have to check everything a dozen times before I power up. The J2 seems to be a good choice for direct drive on a ribbon.
Tad
Would it be possible to use the Semisouth silicon carbide high voltage jfet in the frontend of the amplifier to allow higher overall rail voltages. Have they got the required specs for the differentials and drive units. If the Semisouth part replaced the Loveltech the amp would be considerable less fragile. You could then dispense with the cascode element and simplifiy the circuit. I need very simple these days. The datasheet states very low initial gate voltage and very high effeciency.
I doubt I will ever figure out how all of this stuff works. However, most of my amps turn out smokeless. That is most not all. I have to check everything a dozen times before I power up. The J2 seems to be a good choice for direct drive on a ribbon.
Tad
The IR P channel parts have higher distortion. You could use
the Fairchilds, or take a look at Toshiba or Harris (if you can
find them).
The N vs P channel mismatch is a small source of distortion
down around a watt or so, which is why you want to be
able to trim the Mosfet Source resistors values a bit (if you
have the analyzer handy). Also, it is effective to use a
couple hundred ohm pot, wiper grounded, connected across
the Sources of the input JFETs and trim for least distortion at
1 watt.
The J2 schematic in the 6moons article used parallel fets ins the current source and the diff. amp. Did you use it to lower the noise floor, or was it due to drive requirement???
What is the advantage compared to this topology:
Single fet diff amp, cascoded with small bipolar devices, diff amp build with led/bippolar current source instead of jfet.
Mosfet source follower, to deliver drive current.
Rest: As is.
What is the advantage compared to this topology:
Single fet diff amp, cascoded with small bipolar devices, diff amp build with led/bippolar current source instead of jfet.
Mosfet source follower, to deliver drive current.
Rest: As is.
Another question, showing up my bad knowledge:
I would understand a SRPP with only a single resistor made. But here the output is taken between 2 resrstors. Why???? I also have another question concerning the 6-critter: Why don't we use a few resistors and a capacitor to form the voltage source???
Thanks, Dirk
I would understand a SRPP with only a single resistor made. But here the output is taken between 2 resrstors. Why???? I also have another question concerning the 6-critter: Why don't we use a few resistors and a capacitor to form the voltage source???
Thanks, Dirk
I don't know the circuit that Nelson has designed, but optimizing the SRPP to it's load may be the reason for using two resistors here. See the recent series of articles by John Broskie on http://www.tubecad.com/ .
Jim
Jim
well ..... first thing is that I don't have SS 3-legged fuses near (enough to) me , and second ( pretty more important) thing is that I'm certainly not Papawakoo enough to know many things .......
anyway - imagine that lower part of amp - made of input Jfet+lower output Jfet have some ..... ok - xconductance (maybe better word is transfer function );
probably upper part of amp - comprised of upper output Jfet have another xconductance (transfer function ) ;
these two resistors are (probably) there to make these two transfer functions "same" , to make funny SRPP output halves to working in unison ;
symmetrical clipping ..... blah blah .....
anyway - my crystal ball is lately pretty muddy , and usually all I can see is my lack of official education in field of electronic ; not to mention evident lack of talent ......
so - this is how little ZM imagine how white puff works inside these critters ...
Though my knowledge in electronics is rather limited – I do not understand the mu follower or SRPP subtleties - I think I got the mainlines of how the J2 works.
Let-s call R1 and R2 the upper and lower resistors.
Imagine the bottom leg of the voltage source has moved to R1/ R2 middle point.
The upper jfet is oubviously a current source, the lower one a single ended amplifier.
That is like a Zen output.
Here we have to remember how current varies into the two output devices when a signal is applied to such an amp.
The upper device ( and so does R1) see a constant current.
The current into the lower device varies following the current into the load. From nearly zero when the device is quite “off” to 2 x Bias current when the same device is quite “on”
So, if the voltage across R1 remains constant, the voltage across R2 is an image of the output current.
Moving the lower leg of the voltage source to the other side of R2 makes the output current modulate the current source. Such as an Aleph one.
...But may be I am wrong.
Let-s call R1 and R2 the upper and lower resistors.
Imagine the bottom leg of the voltage source has moved to R1/ R2 middle point.
The upper jfet is oubviously a current source, the lower one a single ended amplifier.
That is like a Zen output.
Here we have to remember how current varies into the two output devices when a signal is applied to such an amp.
The upper device ( and so does R1) see a constant current.
The current into the lower device varies following the current into the load. From nearly zero when the device is quite “off” to 2 x Bias current when the same device is quite “on”
So, if the voltage across R1 remains constant, the voltage across R2 is an image of the output current.
Moving the lower leg of the voltage source to the other side of R2 makes the output current modulate the current source. Such as an Aleph one.
...But may be I am wrong.