What is a Quad balanced amp?

[ron-c]

Using MC452 as an example of a Quad Balanced McIntosh amplifier, from the Owner's Manual:

'The MC452 is fully balanced from inputs and outputs. It consists of two matched power amplifiers operating in push-pull with their outputs combined in a McIntosh Autoformer. The Quad Balanced configuration cancels virtually all distortion'

Each channel of the MC452 has two independent 225 watt power amplifiers. These amplifiers and all input circuitry mirror each other and they are fed a balanced input signal with one amp inverted from the other.
This design will not allow induced noises and distortions in the output as these would be out of phase with the actual signal which a true balanced design cancels.
In the output Autoformer the two amplifiers are combined and outputted as a negative and positive signal. The terminal posts are 2 Ohm +, 2 Ohm-, 4 Ohm+, 4 Ohm-, 8 Ohm+, 8 Ohm-. The negative connection floats above any chassis ground as it is as 'hot' as the positive output terminal.
using this type of amplifier with and connection that shares ground between channels or grounds to chassis is a short circuit. These would not be the amps to use on SDA Polk speakers as an example.
The disadvantage of the Quad Balanced design is that the parts count is doubled in many parts of the amp, the autoformer and output connections.
Conventional 'balanced' amplifiers typically share chassis ground which means they are actually bridged. The McIntosh Quad Balanced will only work with the unique Autoformer.
A Quad Balanced amp will have a higher signal to noise ratio than other designs by 6 to 16 dB which is a lot. Quad Balanced was introduced on the MC1000 and has been used on many of the best models such as MC352, MC402, MC501, MC601, MC1201, MC1.2K, MC2KW and MC2301. The high signal to noise ratio of the Quad Balanced amps will exceed that of almost preamps so your sources will be reproduced

 

[miner]

Thank you Ron. Is Dan Graves still with McIntosh? He gave my wife and I a tour a few years ago.

 

[ron-c]

Sure, Dan-G is still with McIntosh, he is a youngster!

thanks,
Ron-C

 

[ODS123]

"A Quad Balanced amp will have a higher signal to noise ratio than other designs by 6 to 16 dB which is a lot.."

While this may be true, it's hard to imagine a scenario in which this improvement would be audible. ..This strikes me as a very incremental improvement on what is already stellar performance. With ANY of Mac's current amps, source noise (recording hiss) and background noise, etc. are way louder than amp noise. ..So these Quad Balanced improvements are impressive technically, but in practical application, they are essentially meaningless. ..IMHO.

Would you (or anyone) be able to pick out a quad balanced Mac amp from a non in a blind-folded, volume matched trial?? ..Not a chance, I suspect.

 

[william.keith]

An opposing view on the advantages of balanced topology (see topic #9).

http://www.monoandstereo.com/2013/11/interview-with-roger-sanders.html

 

[jlovda]

"A Quad Balanced amp will have a higher signal to noise ratio than other designs by 6 to 16 dB which is a lot.."

While this may be true, it's hard to imagine a scenario in which this improvement would be audible. ..This strikes me as a very incremental improvement on what is already stellar performance. With ANY of Mac's current amps, source noise (recording hiss) and background noise, etc. are way louder than amp noise. ..So these Quad Balanced improvements are impressive technically, but in practical application, they are essentially meaningless. ..IMHO.

Would you (or anyone) be able to pick out a quad balanced Mac amp from a non in a blind-folded, volume matched trial?? ..Not a chance, I suspect.

Since the dawn of audio manufacturers have been trying to add as many zeros to the right of the decimal point as possible whether it is sonically over the top or not.

 

[Pio1980]

Didn't Nakamichi call this "diamond differential drive"?

Sent from my SCH-I605 using Tapatalk

 

[kevzep]

An opposing view on the advantages of balanced topology (see topic #9).

http://www.monoandstereo.com/2013/11/interview-with-roger-sanders.html

Wrong type of balancing topology.....and I personally think this guy is talking crap, IMO of course......

 

[kevzep]

Didn't Nakamichi call this "diamond differential drive"?

Sent from my SCH-I605 using Tapatalk

Sansui were doing the same thing in some of their designs, they advanced the simple differential designs and used it on the front end of the driver stages in some of their power amps, the late AU intergrated amps, the G series receivers too....it was a major advancement in S/N performance at the time and they came up with some great marketing terms to promote it....

Differential design is nothing new and it has been around since the dawn of SS designs, but McIntosh are the only ones I know of to apply it in the power output stage of an amplifier......
It is all about noise rejection and amplifying the actual signal and not anything else.....

 

[william.keith]

Wrong type of balancing topology.....and I personally think this guy is talking crap, IMO of course......

I think we all could save a lot of time if we just cleared 'everything' through you first.

 

[AudioClassics1]

Sure, Dan-G is still with McIntosh, he is a youngster!

I guess this makes you an old man now Ron.....

 

[Pio1980]

Sansui were doing the same thing in some of their designs, they advanced the simple differential designs and used it on the front end of the driver stages in some of their power amps, the late AU intergrated amps, the G series receivers too....it was a major advancement in S/N performance at the time and they came up with some great marketing terms to promote it....

Differential design is nothing new and it has been around since the dawn of SS designs, but McIntosh are the only ones I know of to apply it in the power output stage of an amplifier......
It is all about noise rejection and amplifying the actual signal and not anything else.....

A-ha! I (mis)remembered Diamond Differential as a Nak term. Nonetheless a major advantage of quad-diffy output is that with a servo amplifier it permits using the entire rail difference range for the output voltage instead of just half against the center. This allows essentially double the power with the same rail voltages as the conventional servo amp, all else being equal, BUT, it will tax the power supply capacity as well as the output capacity twice as hard and will not do well with lower 'Z' output loading without an impedance transformer such as Mac uses. Also without an isolation type output transformer the channel output returns (-) cannot be assumed to be common and should NEVER be joined as they will have a substantial differential between them in normal stereo operation.
FWIW, Some car amps are full-diffy and will powerfully self-destruct if the speaker returns are joined or grounded.

 

[ron-c]

Ryan,

Ah, yes, pretty much!

Thanks,
Ron-C

 

[kevzep]

A-ha! I (mis)remembered Diamond Differential as a Nak term. Nonetheless a major advantage of quad-diffy output is that with a servo amplifier it permits using the entire rail difference range for the output voltage instead of just half against the center. This allows essentially double the power with the same rail voltages as the conventional servo amp, all else being equal, BUT, it will tax the power supply capacity as well as the output capacity twice as hard and will not do well with lower 'Z' output loading without an impedance transformer such as Mac uses. Also without an isolation type output transformer the channel output returns (-) cannot be assumed to be common and should NEVER be joined as they will have a substantial differential between them in normal stereo operation.
FWIW, Some car amps are full-diffy and will powerfully self-destruct if the speaker returns are joined or grounded.

Oh right, yeah I wondered how Nak would have used to same marketing term "Diamond Differential"......anyways, yeah absolutely correct and yeah those Sansui's are pretty well designed to a point, they do use a separate supply to the differential circuit in the front end of the driver stage, which needs to be properly regulated, but they just throw a couple of zeners in there rather than designing a decent regulated supply.....a weak point in those amps....and when those zeners let go all hell breaks loose.....

Correct, the (-) output is not common, and certainly not ground.....it does go to some length to explain this in the manual.....

 

[Hyperion]

What you are talking about sounds remarkably like the system used in for example the Sansui AU-D11 II, and quite a few of their higher end power amps after that, (with a steadily improving design). When you talk about 'hot' speaker terminals, 2 amplifiers per channel, operating in push-pull, linked together, and sensitivity to loading, it all sounds very much the same, or is this something else - if so please explain? I've been following this thread with interest...

Thanks,

 

[kevzep]

What you are talking about sounds remarkably like the system used in for example the Sansui AU-D11 II, and quite a few of their higher end power amps after that, (with a steadily improving design). When you talk about 'hot' speaker terminals, 2 amplifiers per channel linked together, and sensitivity to loading, it all sounds very much the same, or is this something else - if so please explain? I've been following this thread with interest...

Thanks,

Well as much as I can gather John, its not a floating bridge as such, but they do behave as one as the negative side is not referenced to ground and there are two amps running together.
It seems one amp runs out of phase with the other, and thats also like a bridge design....
They say the MC452 (450watts per channel) is two 225watt amps, so that seems to fit for a bridge amp...
But it behaves like a differential pair in that it only amplifies the signal, and cancels out noise and anything else which is not the signal....
So that fits for a differential topology.....

Its not that McIntosh are cagy about it, they only wet your appetite for how it works....
I think I have got a MC402 schematic somewhere, I'll see if I can dig it up and get a better understanding......

 

[Unabubba]

Probably a dumb question, but here it is nonetheless. Is that sizeable increase in the signal to noise ratio exactly why most of us that set up Quad systems 'back in the day' used a decoder that was seperate from our amplifier (s) and it's (their) electrical system (s)?

And since I asked that dumb question, I'll follow up with this one. Just what, exactly, is McIntosh doing differently now that would make this an 'improvement'?

At least that was why I used a seperate decoder.

 

[kevzep]

Probably a dumb question, but here it is nonetheless. Is that sizeable increase in the signal to noise ratio exactly why most of us that set up Quad systems 'back in the day' used a decoder that was seperate from our amplifier (s) and it's (their) electrical system (s)?

And since I asked that dumb question, I'll follow up with this one. Just what, exactly, is McIntosh doing differently now that would make this an 'improvement'?

At least that was why I used a seperate decoder.

No, it has nothing to do with quadraphonic or quadraphonic decoders......

Its a marketing description of the amplifier design topology used, these are all stereo amplifiers.....

What you are describing is 4ch quadraphonic sound or the early surround sound using CD4 or SQ and other decoding to separate the two channel signal into 4 ch surround....
Completely different subject and objective......

 

[Unabubba]

No, it has nothing to do with quadraphonic or quadraphonic decoders......

Its a marketing description of the amplifier design topology used, these are all stereo amplifiers.....

What you are describing is 4ch quadraphonic sound or the early surround sound using CD4 or SQ and other decoding to separate the two channel signal into 4 ch surround....
Completely different subject and objective......

Thanks. I was trying to wrap my head around what set of circumstances a push/pull amplifier would be suitable for discrete quad, there are none. SQ or synthesized quadrophonic did cross my mind, but I had difficulty trying to figure out McIntosh's reasoning in building such an amplifier.

When I did quadrophonic when I was younger, we did it with a decoder and 4 monoblock amplifiers. Push/pull has way too much bleed over for discrete quad. If their objective was quadrophonic sound, this would appear a tragically bad plan.

 

[Pio1980]

There is some confusion here, push-pull is a balanced drive configuration typically used in most tube and solid-state power amplifiers to provide reasonable output levels at reasonably low quiescent current consumption. It has nothing to do with quadraphonic sound other than to be used in the power amps sections to get good speaker drive capability.

Thanks. I was trying to wrap my head around what set of circumstances a push/pull amplifier would be suitable for discrete quad, there are none. SQ or synthesized quadrophonic did cross my mind, but I had difficulty trying to figure out McIntosh's reasoning in building such an amplifier.

When I did quadrophonic when I was younger, we did it with a decoder and 4 monoblock amplifiers. Push/pull has way too much bleed over for discrete quad. If their objective was quadrophonic sound, this would appear a tragically bad plan.