Can one turn a Stereo amp into a Parallel Monoblock? (Not Bridged!)

[dlaloum]

Note: this is definitely an academic question... my amp with a spare channel is a Crown XLS2500 - which in no way requires more power or current with the speaker it is driving....

 

[MakeMineVinyl]

I read that.... but I also saw the Peter Walker approved mod for his 405 design.... a highly respected amplifier designer - and if you read the linked PDF in my OP - you will see the circuit and setup/tuning details.

Yes the channels need to be matched.

There may be other "gotcha's" with different designs - eg: if the two channels being linked differ in their response to the environment (thermal or load response) - the more different the channels are the more issues are possible.

But the design assumes individual testing of each amp to match the channels.

Are there additional risks, with differing amp designs?

Is this something that Walker could do with his design, partly due to the relatively indestructible nature of his design? (some of these legendary amps survived extended periods of short circuit, in use at the BBC)

And if you do achieve it successfully - what would be the expected result?
What would be the expected change in specification of the resulting Frankenamp?

Then have at it.

 

[sarumbear]

So reading random interesting bits of the internet, I came across the concept of the Parallel Amplifier....

At the component level, two amplification circuits, run in parallel, keeping the same V output, but doubling the current - sort of the Yin to a Bridged setup's yang...

The amp would have a lot more current and ability to handle difficult loads (low impedance, capacitive loads) - but it would not get higher power output.

At least I think that is how that would work....

Complexities include needing to match the two channels, ensuring no DC offset... others?

Apparently in the early days of the Quad 405, Quad used to have a Parallel amplification conversion kit.

This piqued my interest, as I may have a spare channel lying around, once I use one of the channels of a stereo amp for my center.

Thought I would throw it out there to the collective brain trust.

(instructions and details of the Quad 405 conversion are here: https://www.dadaelectronics.eu/uplo...Documents/Quad-405-Monoblock-Instructions.pdf )

The short answer is no. The example you posted is from the days (hall a century ago) of low power and low damping factor values on power amplifiers. It worked by increasing the damping factor. It would be silly to do that now. Damping factor is a very important factor. It effects FR for starters.

 

[sergeauckland]

The short answer is no. The example you posted is from the days (hall a century ago) of low power and low damping factor values on power amplifiers. It worked by increasing the damping factor. It would be silly to do that now. Damping factor is a very important factor. It effects FR for starters.

Actually, the Quad modification, and any other paralleling would decrease the damping factor by the value of the balancing resistance, which far exceeds the output impedance of the amplifiers being paralleled. With valve amps that's different, as there is already a higher output impedance that won't necessarily need any external balancing resistors. Although, with valve amps, getting the two halves identical enough to maintain balance over the full frequency range is doubtful.

As to the importance of damping factor, this is grossly overrated, as once the output impedance gets to 10% of the minimum load, any further reduction has a minimal effect on frequency response. A damping factor of 20 is quite adequate, and would account for any sensible impedance dips from nominal.

S

 

[sarumbear]

Actually, the Quad modification, and any other paralleling would decrease the damping factor by the value of the balancing resistance, which far exceeds the output impedance of the amplifiers being paralleled.

In other words the damping factor will be dramatically increased decreased.

As to the importance of damping factor, this is grossly overrated,

I disagree.

 

[egellings]

D.F. is very important, but you don't need huge amounts of it. After about 20 or so, you will likely not hear a difference between that and a lot more on most speakers. Speakers that impose a severe load on the amplifier may benefit by a bit more however.

 

[DonH56]

The output impedance of a single amp is Zamp, to which you add a series resistor Rs to help perform load-balancing (similar to the power combiner in the RF example). This keeps the amplifier's outputs from being directly shorted to help protect the amp channels. The output impedance goes from Zout = Zamp for a single amp to Zout ~ (Zamp+Rs)/2 for the parallel combination, assuming identical amplifier channels. If Rs is greater than Zamp, then the effective output impedance of the paralleled outputs (Zout) will be higher than that of a single amplifier, and thus damping factor (Zload/Zout) will decrease (larger Zout --> smaller damping factor). If Rs is less than Zamp, then the parallel impedance will be lower, and damping factor will increase. (And is Rs = Zamp then the parallel combination will have the same output impedance and damping factor, though the amp's Zamp will change with frequency whilst Rx probably does not, at least in the audio band). I do not know the output impedance of a Quad 405 but that will determine how much you actually gain from such an arrangement.

Note the series Rs are outside the feedback loop so their extra resistance will not be compensated.

 

[Head_Unit]

An idea not taken up by anyone else as far as I know,

Bryston monoblock. But no, not common. Probably tricky to engineer. In practice, usually gotten around by using multiple subs and/or dual voice coils I suppose?

 

[polmuaddib]

Would making it balanced help? And would that be simple and safe?
Just connect + to left (or right) positive and - to right positive (or left) and use left speaker output as + and right output as - ?
Would that double the voltage? Increase power? or that doesn't have anything to do with OP question?

 

[dlaloum]

Bryston monoblock. But no, not common. Probably tricky to engineer. In practice, usually gotten around by using multiple subs and/or dual voice coils I suppose?

Yes, I looked up the Bryston monoblocks up, and there is a factory option to have them set up in parallel rather than their default bridged mode... conversion from one mode to the other requires shipping it back to the factory - so presumably they do a similar form of tuning to what Quad recommended for the 405...

 

[sergeauckland]

In other words the damping factor will be dramatically increased.

I disagree.

No, the damping factor will be reduced, i.e. it will get smaller as the output impedance of the parallel pair will be increased by the value of the balancing resistors, albeit halved by the paralleling. As the balancing resistors have a value well in excess of the output impedance of a SS amplifier like the Quad, the overall damping factor is worse than with unparalleled amps, but not by enough to make any difference.

Once the damping factor is above 10 or so, that's quite enough.

S

 

[sarumbear]

No, the damping factor will be reduced, i.e. it will get smaller as the output impedance of the parallel pair will be increased by the value of the balancing resistors, albeit halved by the paralleling. As the balancing resistors have a value well in excess of the output impedance of a SS amplifier like the Quad, the overall damping factor is worse than with unparalleled amps, but not by enough to make any difference.

Once the damping factor is above 10 or so, that's quite enough.

S

I made a typo, I meant decreased but wrote increased instead, thinking impedance. I corrected my post.