Workhorse said:
Hi Kanwar,
You are asking the right question. In an earlier post I asserted that it is not a serious problem. Moreover, the capacitance changes in BJTs are even bigger when you properly take into account the stored charge in the base.
As long as you drive the output MOSFET with a fairly low impedance, and with a source that can source and sink the necessary worst-case charging currents for the desired slew rate, you should have no problem, even without the use of EC, from this effect. This is a very good reason NOT to drive a MOSFET directly from the VAS, as some have done. The low impedance drive for the MOSFET is desirable both from the point of view of feedback loop stability under worst case conditions and from the point of view of distortion.
The varying input capacitance of the MOSFET and its desirable driving impedance should also temper one when choosing gate stopper resistors.
The nonlinearity of the input capacitance of a MOSFET is real, but it is not a huge effect and is not a serious problem in a proper amplifier design.
Cheers,
Bob
estuart said:
I would definitely call it Class AB, but I have always thought of a BJT stage optimally biased as being class AB as well (somewhat in contradiction to Self's (re) definition of Class B).
These are impressive results you are getting!
Cheers,
Bob
estuart said:
Hi Edmond,
sorry - I did miss your message earlier 🙁 . And the answer is no, I don't know. However you can easily check it in a simple circuit like I've described earlier - looking at the Vgs and driving the gate with a current source keeping Vds constant.
Alex
Bob Cordell said:
Hi Bob, I can easily agree with most of what you've said above. However two points need to be made. The simulation results, especially re. distortion behaviour are unreliable. And second - if you know that there is a possible problem you have a better chance to get things right 🙂 .
One of your points could be useful to elaborate further:
the gate-stopping resistor creates a weak point in this situation. I knew from my experience that if you increase it's value, the sound quality may suffer, and I did find a very simple solution to that at the time - to put a small inductance in parallel to the gate-stopping resistor. Then I knew it worked but now I know why it does help. The inductance creates much lower driving impedance for the gate at audio frequencies, reducing unlinearity, and at the same time does not affect the function of the gate-stopping resistor at HF where it is important.
Cheers
Alex
Bob Cordell said:
Hi Bob,
As far as your design concerns -driving the MOSFET's by a low impedance- you are right, but some people like to explore different topologies -common source for example- in which the MOSFET's has to be current driven. In this case, it's really important to know exactly how the gate (Miller) capacitance changes under various conditions.
Cheers,
Bob Cordell said:
HI Bob,
So, from now on it's called class-AB, definitely. No confusion any longer.
As for the results, these are only simulations, but the problem of fighting VAS's (or is it VASes?) (by means of the common mode loop) seems to be solved.
Cheers, Bod
I would call it class AB(2), where 1A, Iq and above is class AB(1) This comes from the tube amp characterizations in the past.
x-pro said:
Hi Alex,
Sure, I can repeat your experiment , but there is no need to do this, as I assume your measurements are correct. But, actually, that's not my point, rather how can we model a MOSFET more precisely. BSIM 3.3 claims to model a MOSFET in a more physical way. However there's no data available for all these parameters. I did manage to get the DC parameters right (including the weak inversion region!), but failed to model that evil capacitances. So, I revert to the simplistic capacitance of a diode. I really do hope that you provide us with some suggestions how tot tackle the capacitance model more accurately .
Cheers, Edmond
john curl said:
Hi John,
My fuzzy recollection from the old tube days was that Class AB2 was when the grid was allowed to draw current. I certainly agree that heavily-biased Class AB should be called something a little different; I like to call it Class AAB, since it intentionally seems to combine some of the Class A operating region with the more traditional Class AB region. I like your value of 1A as the dividing line.
Bob
Bob Cordell said:
Mr. Curl, Mr. Cordell
Though this may sound picky on details, I belive there is a definite difference among class AB (and B) where one half is allowed to cut off completely during part or all of the opposite cycle, in comparison with a topology where even during the opposite cycle there is a controled amount of current (not following the signal as is the clase for class A).
Perhaps the concept deserves a designation of its own, and the dividing line is not really a certain amount of current but a fundamentally different mode of operation bridging class AB-B with class A.
Rodolfo
PS. AAB points in the general direction
estuart said:
Hi Edmond,
I was actually referring only to a simulation 🙂 .
estuart said:
My only approximation so far was to use a second MOSFET with small self-capacitances and a similar threshold voltage to switch in an additional capacitor near the threshold. Also it looks like some of the models provided by Fairchild may attempt to simulate at least some of that behaviour, however I did not try these yet. I am pretty sure that there are some better (in this respect) models - but I did not have an opportunity to try these. It is quite clear thought, that with a complex and distributed character of MOSFET behaviour in this respect, there would always be a limit to abilities of a simulator.
I've heard a story that even most elaborate models actually do not represent MOSFET behaviour in an analogue circuit sufficiently accurately for a good distortion analysis - possibly that is just a rumor 🙂.
Cheers,
Alex
I agree that Class AAB makes sense. The point that I wish to make is that if an amp is mostly working in the Class A region for normal program material, and most of the time, then the Class A region should be considered dominant. However, if a circuit is biased so that it is going through a transistion to Class B more than 50% or so of the time, then it is Class B, mostly. Still, pure Class B can be defined as 90-100% of the time, so that is yet another case. So, I guess Class B to Class AB to Class AAB to Class A is OK with me.
john curl said:
Mr. Curl,
Yet I find there is a subtle - or not so subtle - difference on the possible interpretations of class AAB. To put it more clearly, the standard definitions, and what should be debated.
- Class A - Both devices reproduce the same signal in opposite phase.
- Class B - Only one device conducts on each half cycle, the anti phase one being cutoff.
- Class AB - Both devices operate as in class A within a (small) overlapping region, and are respectively cutoff during part of the cycle.
Now for class AAB.
Alternative 1. - Same as class AB, only the region of overlap is substantial whereby at low signal levels operation si truly class A. Full cutoff still holds for a small to moderate part of the cycle for higher power levels.
Alternative 2. - Like class AB, only in this case special means are provided so as to prevent full cutoff at any power level. The minimum anti phase current may be anywhere from usual class AB quiescent bias, to something like the quiescent bias for Alternative 1.
Rodolfo
Great
Hello Bob,
Thats exactly what i wanted to hear...no matter how much worst the capacitances are, But driving the gate with low source impedance, the effect of these capacitances or Nikitin's Jump effect has no significance at all.....
Some designers use Ferrite Rings on the gates of the mosfets, to eliminate parasitic oscillations
regards,
Kanwar
Bob Cordell said:
Hello Bob,
Thats exactly what i wanted to hear...no matter how much worst the capacitances are, But driving the gate with low source impedance, the effect of these capacitances or Nikitin's Jump effect has no significance at all.....
Some designers use Ferrite Rings on the gates of the mosfets, to eliminate parasitic oscillations
regards,
Kanwar
In my personal opinion:
Class A: both sides always conducting
Class AB: only one side conducting at somepoint in the cycle, two at another point.
Class B: only one side at a time.
Class C: significant periods with no device conducting.
I know that's fairly standard, the unique part of my view is the strictness of my definition. IMO the transfer function doesn't matter, all that matters is the current waveforms. An unbiased output stage with significant feedback is class B, bias to (let's say around 1-5ma) and it's now a very lightly biased class AB stage.
As I see it, the class of an output stage is fundamentally defined by it's associated current waveforms and not it's biasing.
P.S. Even an unbiased output stage will draw some current, so obviously we still have to arbitrarily define some transition point between iq=~0 class B, and iq=>0 class AB.
Class A: both sides always conducting
Class AB: only one side conducting at somepoint in the cycle, two at another point.
Class B: only one side at a time.
Class C: significant periods with no device conducting.
I know that's fairly standard, the unique part of my view is the strictness of my definition. IMO the transfer function doesn't matter, all that matters is the current waveforms. An unbiased output stage with significant feedback is class B, bias to (let's say around 1-5ma) and it's now a very lightly biased class AB stage.
As I see it, the class of an output stage is fundamentally defined by it's associated current waveforms and not it's biasing.
P.S. Even an unbiased output stage will draw some current, so obviously we still have to arbitrarily define some transition point between iq=~0 class B, and iq=>0 class AB.
ingrast said:
Hi,ingrast said:
I dislike alternative 2.
IMO, the nearly constant residual current in the "antiphase" device is fixed (or nearly so) and as such does not control the current passing to the load by the "phase" device. Whether the non active device is passing zero current or a low value of constant current makes no difference to the definition of ClassAB. It is the constant current nature (zero or insignificant) of the non active device that makes it completely different from ClassA where both devices are controlling the output current at all times.
Leach even uses the non zero current in the driver as a definition of ClassA drive from the driver stage. I assert that this is completely wrong. The active device is controlling the output (to the bases of the outputs) and the non active device is idling along
AndrewT said:
Hi Andrew,
I would like to support this point of view. It is exactly as I would approach the question of classes. Important bit: is the device actually amplifing the signal or not? If not (i.e. current is constant or 0) it's condition is irrelevant in a definition of the amplification class.
Cheers
Alex
x-pro said:
Makes sense
Rodolfo
PS We may agree nonetheless, from the standpoint of crossover performance, that there is a substantial difference in having the anti phase device completely cutoff or idling at some nonzero current.
This is particularly relevant in global feedback arrangements, where a susbstantial surge in drive can be observed, trying to quickly bring the cutoff device into operation. It was this fact that brought my attention on the issue.
ingrast said:
Makes sense
Rodolfo
PS We may agree nonetheless, from the standpoint of crossover performance, that there is a substantial difference in having the anti phase device completely cutoff or idling at some nonzero current.
This is particularly relevant in global feedback arrangements, where a susbstantial surge in drive can be observed, trying to quickly bring the cutoff device into operation. It was this fact that brought my attention on the issue.
