I think it can be accomplished with one gain stage.
Figure 1 from the patent - delete the output stage - Hmmm - Fascinating.
Figure 1 from the patent - delete the output stage - Hmmm - Fascinating.
From SOn of Zen to X SoZ
Here is how I worked te get te X version of Son of Zen.
( this time seriously )
Patent by N. Pass number 5.376.899 and SoZ circuit at my hand:
first scaled R1 and R2 values from 8 to 4.7 ohm each.
Than take 2 P channel gain devices IRF 9240 and following patent, joint each Source to the Drain(s) of new Soz .Put 12V across Gates to cascodin' this devices each with zener diods .
Following step : current sourcing to output this folded cascode .
I put 10 ohm power resistor to each halve .
Now ; susy connection (or feedback) .I choose a 100K ohm resistor.
The circuit went up with 12V rails . For my preference i set gain resistor (R7 soz circuit) at 8.2 ohm.
It works fine .
Thanks all for suggestions .
Ps. I let you imagine how much current it draws.
Here is how I worked te get te X version of Son of Zen.
( this time seriously )
Patent by N. Pass number 5.376.899 and SoZ circuit at my hand:
first scaled R1 and R2 values from 8 to 4.7 ohm each.
Than take 2 P channel gain devices IRF 9240 and following patent, joint each Source to the Drain(s) of new Soz .Put 12V across Gates to cascodin' this devices each with zener diods .
Following step : current sourcing to output this folded cascode .
I put 10 ohm power resistor to each halve .
Now ; susy connection (or feedback) .I choose a 100K ohm resistor.
The circuit went up with 12V rails . For my preference i set gain resistor (R7 soz circuit) at 8.2 ohm.
It works fine .
Thanks all for suggestions .
Ps. I let you imagine how much current it draws.
From SOZ to X-SOZ
Stefanobilliani, it is hard to visualise what you mean without a schematic. I suspect I will have many questions, but first it seems to me that we need a common reference from which to start. To this end I have attempted to interpret your description and translate it to a schematic. Let me know in what respects it is wrong (as I am sure it will be) and I will try and update it. I look forward to some interesting discussion!
Ian.
Stefanobilliani, it is hard to visualise what you mean without a schematic. I suspect I will have many questions, but first it seems to me that we need a common reference from which to start. To this end I have attempted to interpret your description and translate it to a schematic. Let me know in what respects it is wrong (as I am sure it will be) and I will try and update it. I look forward to some interesting discussion!
Ian.
Ian , thank you very much for help. With this schematic everything
will be clearer!
It is exatly what i intended exept for gain devices that are IRFP240 and IRFP9240.
will be clearer!
It is exatly what i intended exept for gain devices that are IRFP240 and IRFP9240.
Hi!
Concerning X-SOZ, I have questions.
1. SOZ already has balanced topology.
This means that common noise or distortion can be canceled.
Is there any special reason to design X-SOZ topology?
2. How about the efficiency of X-SOZ compared with SOZ?
3. I already have 8ohm power resistors for SOZ.
If I replace 4ohm resistors with 8ohm resistors, what will
happen?
Concerning X-SOZ, I have questions.
1. SOZ already has balanced topology.
This means that common noise or distortion can be canceled.
Is there any special reason to design X-SOZ topology?
2. How about the efficiency of X-SOZ compared with SOZ?
3. I already have 8ohm power resistors for SOZ.
If I replace 4ohm resistors with 8ohm resistors, what will
happen?
Rothko,
I drive soz like source follower for a period , this means remove R1 and R2 and circuit still have not problems. This with 12volt rails and power supply rated at about 150va par channel. Any way it will need a large heat dissipation.
Efficiency , i think should be the same. Especially the current throug complete circuit will be about double.
At Passlabs web site there are interesting articles about Supersimmetry, that will give you some careful informations .
One of these aims even low distorsion and noise in audio circuitry.
Please take this as an exploration. I remember you that Nelson Pass , on another thread , posts that later, x series will compare as variatons of the original zen amps at Passdiy web site.
Stefano
I drive soz like source follower for a period , this means remove R1 and R2 and circuit still have not problems. This with 12volt rails and power supply rated at about 150va par channel. Any way it will need a large heat dissipation.
Efficiency , i think should be the same. Especially the current throug complete circuit will be about double.
At Passlabs web site there are interesting articles about Supersimmetry, that will give you some careful informations .
One of these aims even low distorsion and noise in audio circuitry.
Please take this as an exploration. I remember you that Nelson Pass , on another thread , posts that later, x series will compare as variatons of the original zen amps at Passdiy web site.
Stefano
Jam, regarding your question about current sources. Please note that this is not my design – I was just acting as scribe for Stefano. I do have some thoughts though regarding current sources and these appear below.
Stefano, glad to be of help and I’m pleased that my interpretation was not too far off the mark. Sorry about the MOSFETs – I don’t current have symbols/models for those you used (but will rectify this shortly). Now that we have a point of reference I can raise all those questions that I have had since first hearing about this concept! Some of these are not directly about your implementation and are intended as open questions to the forum.
First off, I find myself questioning why you would want to use folded cascades (or cascades at all for that matter). I appreciate that Nelson’s patent 5,376,899 shows this topology, but it is perhaps worth pointing out to the uninitiated that SuSy is not dependent on it. You can in fact convert the SOZ design to use SuSy with just a couple of resistors. I attach a schematic of one such variant but I should also say that I have not actually built it.
Current sources or resistors, how to decide? The previously mentioned patent shows current sources for diff pair tail, loads and cascode output loads (R12 and 13 in my previous post). But this was intended as a line level or power amp front end. When this topology is called upon to deliver substantial current (e.g. driving a speaker) without an output stage, one is effectively driving the speaker with a pair of current sources. In general this is not too clever as the circuit will show substantial output impedance, even with feedback applied. Current sources come into their own in situations where it is possible to take advantage of lower power dissipation, or when the additional impedance is welcome, e.g. to increase gain. I don’t see that either of these apply in this situation.
I admit to being intrigued by the idea of an X Zen that uses both SuSy and Nelson’s ‘Aleph’ current source. But maybe I am simply going off at a tangent in what will turn out to be a fruitless pursuit. I cannot see how to apply this to the folded cascode topology and suspect this to be the wrong direction. The attached schematic could quite easily be converted to use ‘Aleph’ current sources but I think the diff pair tail would have to remain resistive. A pity as the circuit will have poor CMRR and poor performance if driven by a single ended input like this. Power dissipation in the tail resistor will also be quite high L
Anyway, I am now just thinking out loud. How about some of you clever folks making some suggestions?
Ian.
Stefano, glad to be of help and I’m pleased that my interpretation was not too far off the mark. Sorry about the MOSFETs – I don’t current have symbols/models for those you used (but will rectify this shortly). Now that we have a point of reference I can raise all those questions that I have had since first hearing about this concept! Some of these are not directly about your implementation and are intended as open questions to the forum.
First off, I find myself questioning why you would want to use folded cascades (or cascades at all for that matter). I appreciate that Nelson’s patent 5,376,899 shows this topology, but it is perhaps worth pointing out to the uninitiated that SuSy is not dependent on it. You can in fact convert the SOZ design to use SuSy with just a couple of resistors. I attach a schematic of one such variant but I should also say that I have not actually built it.
Current sources or resistors, how to decide? The previously mentioned patent shows current sources for diff pair tail, loads and cascode output loads (R12 and 13 in my previous post). But this was intended as a line level or power amp front end. When this topology is called upon to deliver substantial current (e.g. driving a speaker) without an output stage, one is effectively driving the speaker with a pair of current sources. In general this is not too clever as the circuit will show substantial output impedance, even with feedback applied. Current sources come into their own in situations where it is possible to take advantage of lower power dissipation, or when the additional impedance is welcome, e.g. to increase gain. I don’t see that either of these apply in this situation.
I admit to being intrigued by the idea of an X Zen that uses both SuSy and Nelson’s ‘Aleph’ current source. But maybe I am simply going off at a tangent in what will turn out to be a fruitless pursuit. I cannot see how to apply this to the folded cascode topology and suspect this to be the wrong direction. The attached schematic could quite easily be converted to use ‘Aleph’ current sources but I think the diff pair tail would have to remain resistive. A pity as the circuit will have poor CMRR and poor performance if driven by a single ended input like this. Power dissipation in the tail resistor will also be quite high L
Anyway, I am now just thinking out loud. How about some of you clever folks making some suggestions?
Ian.
Ian,
I looked at your schematic (thank you very much for posting it)....
Maybe I am being tricked by the way it is drawn but I don't see the "X" topology in your schematic. What I see is an soz with an additional inverted gain stage.
I looked at your schematic (thank you very much for posting it)....
Maybe I am being tricked by the way it is drawn but I don't see the "X" topology in your schematic. What I see is an soz with an additional inverted gain stage.
X or not an X
Grataku,
Presumably you are referring to my earlier post since you mention an additional gain stage...
The "inverted gain stage" is not really an additional gain stage at all, it is a folded cascode (non-inverting, voltage gain only) and hence the circuit is more of a glorified diff pair. It is however, SuSy in that distortion from one side of the diff pair is fed, in phase, to the other (whilst signal is in inverse phase). The 'trick' if one can call it that, is the addition of feedback from output to input gate of each side. This causes distortion from one half to be fed to the other (via their sources) by a process that Nelson referrs to as input gain conduction. Does this help any?
Ian.
Grataku,
Presumably you are referring to my earlier post since you mention an additional gain stage...
The "inverted gain stage" is not really an additional gain stage at all, it is a folded cascode (non-inverting, voltage gain only) and hence the circuit is more of a glorified diff pair. It is however, SuSy in that distortion from one side of the diff pair is fed, in phase, to the other (whilst signal is in inverse phase). The 'trick' if one can call it that, is the addition of feedback from output to input gate of each side. This causes distortion from one half to be fed to the other (via their sources) by a process that Nelson referrs to as input gain conduction. Does this help any?
Ian.
Excellent job, Ian, that's twice you have won the
(single) prize offered.
I don't recognize the phrase "input gain conduction", tho.
(single) prize offered.
I don't recognize the phrase "input gain conduction", tho.
Input gain conduction
I was quoting from memory. The actual words from the Patent:
Thanks for your kind words.
Ian.
I was quoting from memory. The actual words from the Patent:
Thanks for your kind words.
Ian.
Lawyers wrote that.
I often refer to cross-coupled error correction, or
cross-coupled feedback in reference to the circuit.
I notice that a lot of people still don't "get" the circuit,
and that is understandable.
The essence is still using a symmetric feedback arrangement
around a differential pair which isolates error from signal
and uses it to make distortion on two halves identically
in phase and thus cancelled.
I often refer to cross-coupled error correction, or
cross-coupled feedback in reference to the circuit.
I notice that a lot of people still don't "get" the circuit,
and that is understandable.
The essence is still using a symmetric feedback arrangement
around a differential pair which isolates error from signal
and uses it to make distortion on two halves identically
in phase and thus cancelled.
SuSy
Took me a logn time to 'get' the concept too... Sometimes when I get caught up in the details it still eludes me 🙁
Ian.
Took me a logn time to 'get' the concept too... Sometimes when I get caught up in the details it still eludes me 🙁
Ian.
"I notice that a lot of people still don't "get" the circuit,
and that is understandable."
Don't mind me Nelson, I am just a chemist. The principle is trivial to understand (now that you worked it out) but I can't follow the schematic. I 'll come around, eventually.
What's most important to me it's how it sounds. If the whole idea sounds anything like the BOSOZ I must build something with susy. If I was a little richer I wouldn't have to understand anything and I could just buy an XA.
and that is understandable."
Don't mind me Nelson, I am just a chemist. The principle is trivial to understand (now that you worked it out) but I can't follow the schematic. I 'll come around, eventually.
What's most important to me it's how it sounds. If the whole idea sounds anything like the BOSOZ I must build something with susy. If I was a little richer I wouldn't have to understand anything and I could just buy an XA.
A little richer
For many of us I think it is more a case of a lot richer 🙁 Actually I would love to hear a real XA but unfortunately there are not that many over here in the UK.
Ian.
For many of us I think it is more a case of a lot richer 🙁 Actually I would love to hear a real XA but unfortunately there are not that many over here in the UK.
Ian.