Is a stereo amp, when bridged to mono, by definition differential?

Hi Howard,

Engineers as well as non-engineers can easily conflate some of these terms incorrectly, at least in their wording if not in their understanding. In particular, the terms "differential" and "push-pull" are often used in an excessively loose manner IMO.

As I see it "push-pull" is best used to distinguish an amplifier from one having a "single-ended" output stage, such as a SET (single-ended triode) tube amplifier, or the single-ended pentode tube amplifier Whart referred to.

In the first paragraph of Steve M.’s writeup that you quoted he says:

Speaker drive is thus differential or push-pull as well.

Yes, the + and - outputs can indeed be thought of as pushing and pulling current, each of them doing so alternately (with one pushing while the other is pulling) on the positive and negative halves of each cycle of the waveform. And yes, the **outputs** (i.e., the "speaker drive") operate in a differential manner. But his wording misleadingly conflates the two terms, IMO. And I’m not sure if his adjacent reference to the DNA-500 being a "fully differential/balanced amplifier from input to output" is worded as precisely as it should be. That wording could be construed to mean that the internal signal paths of the amp are comprised of differential stages, which does not appear to be the case.

The bottom line to my interpretation of the description you quoted is that I would characterize the amp as being equivalent to two robustly designed bridged mono amplifiers on a single chassis.

Also, the first full paragraph of the post by GS5556 is worded a bit more ambiguously than it should be, IMO:

Push-pull means the output stage is has complementary devices that alternately control about half the signal by sourcing and sinking current -- not voltage -- between the positive and negative voltage rails. How much of the signal each control depends on the biasing so both Class A and Class A/B amps are push pull.

First, a push-pull amp can be class A or class AB, but a class A amp is not necessarily push-pull. A single-ended audio amplifier of necessity operates its output stage with class A bias. (If it didn’t there would be little or no output for part of each cycle of the waveform). A push-pull amplifier may bias its output stage in either class A or class AB, or it may bias the output stage in class A up to a certain power level, and then switch to AB. (Or it may operate in class D or other classes, but those are different animals altogether). The distinction between these bias classes is explained in this Wikipedia writeup.

Also, the quoted statement could be interpreted to mean that a push-pull design cuts off the current that is conducted by a given output device during part or all of either the positive or negative half of the output waveform. However such cutoff may or may not occur in a push-pull design, depending on the bias class.

Not sure if all of that adds to or lessens the confusion, but FWIW. :-)

Best regards,
-- Al

If you are talking about one of Steve McCormick's designs in the context of this question, you have no worries. He knows what he is doing and his inputs work correctly.

I know this post is a bit old but I would like to ask a related question.

Looking at the Clayton M300 for example, promoted to be a pure class A balanced fixed bridged design, only with XLR inputs.  If each channel is composed of two 75 watt pure class A amplifiers in balanced design, is the output also class A as stated by the manufacturer? (rated at 300/600 wpc pure class A). Just curious, as 600 Watts class A into 4 ohms seems to be a lot. They are indeed great amps!  Thanks to any responders in advance.

@audiobrian Bridging two channels makes it possible for twice as much voltage to be applied to the speaker as could be provided by one of those channels, **IF** the amp can supply the correspondingly greater amount of current, and **IF** it can do so without overheating. Since for a given load impedance twice the voltage corresponds to four times as much power (P = E squared/R), two channels capable of supplying 75 watts into 8 ohms and presumably 150 watts into 4 ohms can potentially supply 300 and 600 watts into those impedances when bridged.

The majority of bridged amplifiers fall short of realizing that 4x factor, of course, due to current and/or thermal limitations.

In this case we are dealing with a very well regarded and robustly designed monoblock. However both its manual and a review I found in Positive Feedback state that in its high bias mode its power consumption is 800 watts. When providing 600 watts into 4 ohms that would mean an efficiency of 75%, and I’m not sure how an amp operating in class A can provide efficiency that is anywhere close to being that high.

Regards,
-- Al

Thank you Al,

I’ve always wondered how Clayton Audio gets 300 wpc class A into 8 ohms from a relatively small chassis/amplifier, albeit robustly built. I guess part of the answer is a fixed bridged design with 2 x 75 watt class A amps per channel. If power draw in high bias is indeed 800W, perhaps it truly delivers class A performance into 8 ohms but partly class A/AB into 4 ohms?  If one can build a class A amplifier of considerable power using bridged design (with the needed robust power supply and build parts), allowing for a much smaller amplifier, is this design any less desirable sonically than a non bridged class A amplifier such as Pass, Gryphon etc with considerably larger and heavier amplifiers? Manufacturers that have gone down the fixed bridged monoblocks design, not all in class A like Clayton, include Steve McCormack’s DNA 500, all Constellation amplifiers and Naim Reference Series, amongst others.
Thanks so much!