Unipolar vs complementary input stage

Status
Not open for further replies.
Jump to Latest
#22
PMA said:
Class A, push-pull CFP follower, in symmetrical "diamond buffer" topology, which has about 1.5 Mohm input impedance and 0.07 ohm output impedance, is my choice. It has very low distortion (both THD and CCIF IMD), it is fast, and creates negligible load for VAS stage.
Click to expand...

Are you talking about the output stage? Can you show a schematic? I'm interested in the "diamond buffer" topology with CFP follower.
 
#23
Giaime said:


Are you talking about the output stage? Can you show a schematic? I'm interested in the "diamond buffer" topology with CFP follower.
Click to expand...

Basically it looks like this. You can use CCS instead of R5, R6 resistors, but then you get sharper transition into AB, in case of low ohm load.
 

Attachments

  • cfp.gif
    cfp.gif
    5.9 KB · Views: 2,945
#24
#26
forr said:
Hi QSerraTico_Tico

---I am not so sure....---

Me too. Not at all.

At least conceptually, complementary (differential) symmetry is puzzling.
The output voltage (more precisely, the voltage at the node of the feedback network defining the voltage gain) is under the control of two loops, with two input stages of opposite polarity. I am not aware of a detailed analysis of such circuits. Having two servo-mechanisms controlling one function makes me feel uneasy. I also remember the objection from Douglas Self who seems to have never been tempted by explorations in this territory.


The main objective of complementary (differential) symmetry in low power stages seems to rely in getting a push-pull VAS for high slew-rate. However there are some examples of this being achieved with a single differential input stage :
- Hitachi low power stage for its first Mosfet amps : a current mirror loads the VAS (I see Bob Cordell's amp as a sophistication of this basic circuit).
Being a cascade of two differential stages and adding suitable current sources for the coupled emitters, these circuits should have higher PSRR than that of the complementary symmetry. Note that differential stages are push-pull circuits of same polarity devices and suffer less of characteristics dispersion than complementary pairs, and so, should be more linear.
- Sansui "diamond" circuit : the VAS is fed by a symmetrical differential intermediate stage which bears some resemblance with John Curl's Fet input stage and its floating tail. But it is bipolar and is biased by diodes networks included in the loads of the single differential input stage.
- Circuit used in some Harman-Kardon amps like the PM640 : the VAS is fed by two complementary emitter followers connected to the opposite sides of a high value resistor. One of the emitter follower is being directly connected to one of the branches of the differential input stage, and the other one is preceded by a phase reversing common emitter connected to the other branch of the differential input stage.
- Hafler 1500 circuit : same basic idea as the Harman-Kardon. One of the branches of the differential NPN input stage directly feeds the PNP transistor of the VAS stage, the other one being phase reversed by a current mirror to feed the NPN transistor of the VAS stage.
- Hafler 4000 "diamond" circuit (electronicians love diamonds…) : the differential input stage has a basic current mirror load, its output current being connected to the resistors biasing the VAS through two intermediate common base transistors.
- I would like to mention the Bissmire's circuit that I've not seen used in audio. It is not a really push-pull circuit but consists of a cascode configuration with two common base transistors in parallel, but passing different currents, one of them modulating the constant (with no signal) current source load of the other one.

At last, I wonder if a class push-pull follower loading a conventional CCS loaded VAS and feeding the output stage is not able to do exactly the same job as full symmetry without the need to look for closely matched complementary pairs.
Click to expand...


Most of the objections to fully symmetrical complementary differential input stages mentioned here so far (low input LTP gain, VAS noise, device matching) are not necessarily relevant if using BJT's for the complementary LTP's (particularly low noise ones).

JFET inputs are simply not worth using here, IMHO. Personally, I wouldn't use JFET LTP inputs in any topology which does not provide the JFET LTP with a rather high load impedance, for both gain an linearity reasons.

With BJT's, 30-40dB of first stage gain is easily had, which is more than enough to make VAS noise irrelevant, even with emitter degeneration, and linearity will rival any JFET stage. And if each LTP is provided with an adequate tail CCS, PSSR isn't an issue either.

For something different, I made basic variation (on my webpage in a 12W class A implementation) which sims 8ppm THD-20 at full power (<1ppm below 1kHz) Basic preliminary measurements of THD-20 on the built unit confirm that the sim was pretty much spot on.
Edmond and a friend have expanded the circuit a bit (well actually more than a bit) and are getting less than 1ppm THD-20 at 200W in a functional prototype.

Complementary symmetry can work rather well. I spent a great deal of time experimenting with various topologies for my 1kW+1kW amp to get the VAS distortion really low (including Bobs variation of the generic Hitachi topology), and the best results came from a BJT complementary symmetrical differential input driving a BJT complementary symmetrical VAS with a variation of Hawskford style cascoding for minimising "slope distortion".

Cheers,
Glen
 
#27
Forr, I am surprised by your concerns. We have been making complementary differential input stages for at least 39 years, and 100's of design examples, maybe 100,000's of thousands, of individual units are still running today.
We did all the ground work, decades ago. There are few surprises today.
It is not absolutely necessary to use complementary differential, but it does have advantages.
The so called 'diamond differential' is something that we also developed as an alternative to full differential, back in the mid '70's. In those days, like once again today, p channel devices were not as good as n channel devices, especially as duals, and they were much noisier. The 'diamond differential' topology could also be implemented as an all fet design as well. This is how I did it, back then.
As low noise p channel jfets became available, first from Hitachi, then Toshiba, with the 2sj73, it became easy to make preamps and power amps with a true complementary differential jfet input stage, and Hafler, Krell, and just about everybody else, followed the early Levinson lead.
A few years later, these parts became nearly impossible to purchase, so many manufacturers went back to bipolar parts, strictly for practical, commercial reasons, where they remain today. It is almost impossible to second source a Toshiba jfet, especially the p complement. Therefore, it is impractical for many large manufacturers to use Toshiba jfets.
Parasound has used Toshiba jfets for at least 15 years straight, but we are running out of them as well. We will have to wait and see if there is a good solution, in future.
 
#29
John,
My concerns may be academic as there is no doubt that there are lots of complementary differential input stages perfectly working since I saw the first circuit using it, an SAE power amp, in 1973, I think.
Till now, I failed to find an explanation for what I see as two feedback loops to be able to control a single voltage at the amp output node. I can only suppose that there is something like a hidden compromise in the real functioning.
Said otherwise, what happens internally when an input pair has more transconductance than the input pair of the other polarity ?
 
#31
Now I am going to give you fellow designers a real challenge. Did you know that Doug Self is right that bipolar transistor differential pairs actually generate LESS harmonic distortion than jfet differential pairs? There is an IEEE paper on the subject. Study that, and it would appear that you should not use differential fet pairs in any configuration.
 
#32
john curl said:
You are referring to the Widler 101 based current source with an added cascode making it a 4 transistor combination? Yes, I agree that that is a good way to go. I used a 4 transistor cascoded current mirror instead that also kept the temp and voltage differences to a minimum, when I designed with the general topology that you prefer. Your circuit might have less 2'nd harmonic than mine. I'll give you that.
Click to expand...


Hi John,

The differential current source load I use in my amplifier is quite unlike the current source Bob Widlar used in the LM101. Once you get past his funky input stage using quasi-cascoded lateral PNPs, you see that his current source load is merely a conventional current mirror that provides only a single-ended output to his VAS.

My differential current source load takes differential inputs and produces differential outputs. It presents an extremely high impedance in the differential mode while providing a very low impedance in the common mode. This provides further common-mode rejection in the signal path. The circuit is self-biasing and produces a well-defined common-mode output voltage that is ideal for driving a VAS differential pair. Inherent in its function is the buffering provided by its two emitter followers, also very desirable for driving a VAS differential pair. All of this takes only four transistors.

Cheers,
Bob
 
#34
forr said:




The main objective of complementary (differential) symmetry in low power stages seems to rely in getting a push-pull VAS for high slew-rate. However there are some examples of this being achieved with a single differential input stage :
- Hitachi low power stage for its first Mosfet amps : a current mirror loads the VAS (I see Bob Cordell's amp as a sophistication of this basic circuit).
Being a cascade of two differential stages and adding suitable current sources for the coupled emitters, these circuits should have higher PSRR than that of the complementary symmetry. Note that differential stages are push-pull circuits of same polarity devices and suffer less of characteristics dispersion than complementary pairs, and so, should be more linear.
Click to expand...

I agree that the main objective, one way or another, is to arrive at a push-pull VAS. Although that probably is an advantage for slew rate, I view it more importantly as an advantage for linearity.


Do you have a schematic of that Hitachi circuit? I don't think I'm familiar with it, but would certainly be interested in seeing any earlier references to circuits like the one I used.

Cheers,
Bob
 
#36
forr said:
John,
My concerns may be academic as there is no doubt that there are lots of complementary differential input stages perfectly working since I saw the first circuit using it, an SAE power amp, in 1973, I think.
Till now, I failed to find an explanation for what I see as two feedback loops to be able to control a single voltage at the amp output node. I can only suppose that there is something like a hidden compromise in the real functioning.
Said otherwise, what happens internally when an input pair has more transconductance than the input pair of the other polarity ?
Click to expand...


Your concern here harks back to the discussion about "fighting VAS's" awhile back in one of these threads. It can happen, but it can also be avoided with proper design, especially with regard to how the feedback compensation is applied.

Cheers,
Bob
 
Status
Not open for further replies.
Top Bottom