Class A push-pull

[Nikolas Ojala]

This is especially for Mr. Nelson Pass, but anyone who knows may answer.

I read this article written by Nelson Pass: http://www.firstwatt.com/downloads/seclassa.pdf

He wrote:

Problems with push-pull amplifier designs associated with crossover distortion have been discussed elsewhere at length, and one of the primary results is non-monotonicity. Class B and many AB designs have distortion products which dramatically increase with decreasing signal. This is reduced greatly by Class A mode, but crossover distortion remains as a lower order discontinuity in the transfer curve.

I emphasized the important sentences.

Some questions arise:

• Is that a known feature of Class A push-pull amplifiers?
• How does it appear in distortion measurements?
• Any references for further study?

[Nelson Pass]

Originally posted by Nikolas Ojala

• Is that a known feature of Class A push-pull amplifiers?
• How does it appear in distortion measurements?
• Any references for further study?

1) Yes, although it is greatly reduced and it depends on the specific
design, bias, power and load.

2) It shows up as harmonics greater than 3rd at higher power levels

3) "Leaving Class A" at www.passlabs.com discusses some of the issues.

[tiefbassuebertr]

The significant difference in sonic quality between "single ended class A" and "push pull Class A" by commercial devices is in my opinion due to the fact that the so-called "push pull Class A" in real life is often push pull class AB. Just because the area where will be leaving the Class A mode, is a little above typical values arround 30 to 50 mA (e. g. 100-300mA), some manufacturers already believe that they offer "pure Class A" amplifier.

[AndrewT]

Hi,
every amplifier produces distortion along with the signal in the output.
The advantage of a ClassA amplifier is that is has Zero Crossover distortion. It still has all the other distortions and noises etc. It very specifically addresses one distortion issue.

Single Ended ClassA has a hard current limit. The ClassA limit = peak output current limit = output stage bias current.

Push Pull ClassA has a "ClassA" current limit. The ClassA limit = twice the output stage bias current.
But a major difference, (I'll let you decide if this could be interpreted as an advantage) between Single Ended and Push Pull is that the PP can deliver peak output current that can very significantly exceed the ClassA current limit. Appropriate design decisions can use this ability to give performance improvements that ClassAB and SE cannot achieve.

[tiefbassuebertr]

Originally Posted by AndrewT

Hi,
every amplifier produces distortion along with the signal in the output.
The advantage of a ClassA amplifier is that is has Zero Crossover distortion. It still has all the other distortions and noises etc. It very specifically addresses one distortion issue.

Single Ended ClassA has a hard current limit. The ClassA limit = peak output current limit = output stage bias current.

Push Pull ClassA has a "ClassA" current limit. The ClassA limit = twice the output stage bias current.
But a major difference, (I'll let you decide if this could be interpreted as an advantage) between Single Ended and Push Pull is that the PP can deliver peak output current that can very significantly exceed the ClassA current limit. Appropriate design decisions can use this ability to give performance improvements that ClassAB and SE cannot achieve.

I agree with all mentioned things except of the last sentence:
================================================== ====
Appropriate design decisions can use this ability to give performance improvements that ClassAB and SE cannot achieve
================================================== ====


Are there topologies about such designs? Please let me know

[AndrewT]

Hi,
all right, you can have that opinion.
Tell me, how does a SE give peak current that exceeds the bias current in signals of both polarities?
I consider that to be an advantage of PP, particularly when driving reactive loads (all speakers). Highly reactive loads demand even more peak current on fast starting/stopping transients. A recent post showed experimental data for real speakers with real music signals where the peak transient currents were approaching five times the current predicted by the nominal resistive impedance of the (commercial) speakers included in the testing.

Tell me how a ClassAB amplifier can completely eliminate the crossover distortion that is inherent in the topology?

Now, I'll give a simple example of a PP 75W ClassA amplifier where both these limitations are to some extent overcome.
Take a 10pair output stage running on a +-40Vdc PSU. Re=0r1, optimum ClassAB bias per device ~220mA.
Total ClassA bias 2.2A. Peak ClassA output current ~4.4A
Dissipation ~180W/channel, ~18W /device.
Peak transient output current >20A into 1r5
This is a pure ClassA 75W amplifier (for 8ohm loading) that can drive upto 5times nominal maximum current into a reactive load and has Zero crossover distortion while within it's ClassA 75W rating.

Show me a 75W into 8r0 SE or ClassAB amplifier that can achieve both these criteria.