[SoundGame]

I've seen a number of manufacturers, in addition to their THD ratings at rated power output, provide specifications at somewhere between (50% and 80%).  Where this is of interest is when you're contemplating purchasing the next power amp up in say the Bryston line.
 
For example, if you own a speaker with a nominal 8 ohm load that has say a minimum load of 5 ohms and a power handling of 20 - 100 wpc, with a benign impendence and phase curve, you may be debating buying a 2B, 3B or 4B.  Given that the speaker is not difficult to drive - a 2B may be more than adequate but the question begs if a 3B or 4B might provide some advantage.
 
Dynamics of the higher power amps (if the speaker has high dynamic capabilities) may be a reason to move up but I was thinking of whether also THD might be a factor.
 
The question is whether the distortion on Bryston amps is significant lower from 20 to 20,000Hz at say three quarters or half power?  Since at full rated power output the THD ratings of the 2B, 3B and 4B are the same - would the 3B be significantly lower at 75% and what would be the 4B's THD rating at 50% of its rated power output i.e. 150wpc??

[sharpsuxx]

By no means do I know if there is anything beyond placebo to this...in other words YMMV.  But I usually turn my pre-amp up to about 50% then start turning up the amp gain until I hear distortion through the speakers with no signal, then I go a few notches below that to make sure I keep as silent of a noise floor as possible, while giving the speakers as much dynamic range as I can.  If I can't get silence from the speakers, something is wrong with one of the components.  I know this throwing two attentuators in sequence and all that...but hey silence is golden.

I would assume (and we all know what assuming does) that these lower distortion figures would allow me to set the amp gain to a higher volume setting thus giving me higher dynamic range with lower distortion at the speaker level.  I have yet to have any amps go to full bore without at least a little background noise, my budget is not in the bryston region though, obviously there are plenty of dead silent pre-amps, all the way to 100%.  To me the attentuators in sequence can't be worse than a grey noise floor for imaging...Black as possible is best, but I avoid as much of the affect as I can by setting the gain as wide open as possible.  Hopefully at reasonable volumes the amp attentuator would never even be touched.

I am sure you could do much more accurate settings with better measurement, but I usually listen to music with my ears, so they are good enough for this job.  Hopefully someone more technical can chime in.

I hope this answers your question, or at least helps you out.

[SoundGame]

Here's an example, using the Soulution 710 power amplifier, as reviewed in Stereophile:
"The THD+noise vs power (8 ohms) is a mind-bogglingly low 0.00075% up to around 115W, where it rises sharply to a still-negligible 0.1% at the rated 130W or 260W into 8 ohms."
 

[James Tanner]

Hi;

Thanks for your very interesting question.  Generally, Bryston amplifiers all use the same circuit topology from input to output. The only differences lie in their output stages: The smallest amplifiers, like the 2B SST and the 2B-LP use somewhat simplified output stages with 2 output transistors per channel. The largest, 28B-SST2, uses 32 output transistors for one channel, with differing numbers in the middle ranges of amplifiers depending upon the amount of peak and average current that might be required.

The different sizes of amplifier begin to depart from each other in measured performance only as they reach their respective maximum power levels.  In other words, at 60 Watts output all Bryston amplifiers measure very similar, and very low, THD.  As the output power reaches towards 100 Watts, then 150 Watts, then 300, 600 and 1000 Watts, the smaller amplifiers reach their clipping points and are passed in performance by successively larger amps in the lineup. Thus in terms of noise and distortion performance at low power levels, all Bryston amps are very similar to each other.

However, beyond being capable of higher output levels, louder crescendi and wider dynamic range, the larger amplifiers also benefit from the improved load regulation, (‘solidity’), of their power supplies.  It’s simply a fact of the laws of physics that larger transformers have lower internal resistance and thus they better maintain their steady-state Voltage levels under conditions of fluctuating current-draw than smaller transformers.  The very large transformers used in the largest Bryston amps also have significant energy storage in their magnetic cores, reducing the fluctuation of current drawn from the wall.  Larger banks of filter capacitors necessarily used in larger amplifiers hold greater amounts of stored energy, with energy storage going up as the square of the Voltage rating.  All this adds up to the larger Bryston amplifiers showing better focus of the virtual stereo ‘image’ and somewhat deeper sound-field as their power ratings go up.  The differences are subtle, and all Bryston amplifiers have been designed to provide very smooth, lucid and transparent musical reproduction, but I won’t try to pretend that there is no reason other than louder output levels to use a large, powerful amplifier in your optimum music system.  The hugely powerful Bryston 28B-SST2 has been rated by numerous reviewers around the globe as being the most ‘believable’, most focused and transparent, and least ‘identifiable’ amplifier in the world, bar none.

I hope the above is helpful, but please don’t hesitate to contact me if you have any other questions. Thanks for thinking of Bryston!

Chris Russell

[SoundGame]

Hi;

Thanks for your very interesting question.  Generally, Bryston amplifiers all use the same circuit topology from input to output. The only differences lie in their output stages: The smallest amplifiers, like the 2B SST and the 2B-LP use somewhat simplified output stages with 2 output transistors per channel. The largest, 28B-SST2, uses 32 output transistors for one channel, with differing numbers in the middle ranges of amplifiers depending upon the amount of peak and average current that might be required.

The different sizes of amplifier begin to depart from each other in measured performance only as they reach their respective maximum power levels.  In other words, at 60 Watts output all Bryston amplifiers measure very similar, and very low, THD.  As the output power reaches towards 100 Watts, then 150 Watts, then 300, 600 and 1000 Watts, the smaller amplifiers reach their clipping points and are passed in performance by successively larger amps in the lineup. Thus in terms of noise and distortion performance at low power levels, all Bryston amps are very similar to each other.

However, beyond being capable of higher output levels, louder crescendi and wider dynamic range, the larger amplifiers also benefit from the improved load regulation, (‘solidity’), of their power supplies.  It’s simply a fact of the laws of physics that larger transformers have lower internal resistance and thus they better maintain their steady-state Voltage levels under conditions of fluctuating current-draw than smaller transformers.  The very large transformers used in the largest Bryston amps also have significant energy storage in their magnetic cores, reducing the fluctuation of current drawn from the wall.  Larger banks of filter capacitors necessarily used in larger amplifiers hold greater amounts of stored energy, with energy storage going up as the square of the Voltage rating.  All this adds up to the larger Bryston amplifiers showing better focus of the virtual stereo ‘image’ and somewhat deeper sound-field as their power ratings go up.  The differences are subtle, and all Bryston amplifiers have been designed to provide very smooth, lucid and transparent musical reproduction, but I won’t try to pretend that there is no reason other than louder output levels to use a large, powerful amplifier in your optimum music system.  The hugely powerful Bryston 28B-SST2 has been rated by numerous reviewers around the globe as being the most ‘believable’, most focused and transparent, and least ‘identifiable’ amplifier in the world, bar none.

I hope the above is helpful, but please don’t hesitate to contact me if you have any other questions. Thanks for thinking of Bryston!

Chris Russell

A very enlightening response and much appreciated!  This sheds a considerable amount of light on some of the impressions that Bryston owners have shared on this forum and in other venues about perceived improvements in moving up the Bryston amplifier chain - irrespective of the specific power demands of a given speaker.  It also shed light on why the amplifiers also sound so similar, within their respective power ranges.  Thanks for the response - I've learnt something today.    
 
It is

[1oldguy]

Interesting........I like the Sound of that.

[spinner]

 Yes, much appreciate the full story on that. So this is what makes a difference between the 4b and 7b.Thanks,this gives me another area of consideration when I change my amps.

[SoundGame]

Hi;

Thanks for your very interesting question.  Generally, Bryston amplifiers all use the same circuit topology from input to output. The only differences lie in their output stages: The smallest amplifiers, like the 2B SST and the 2B-LP use somewhat simplified output stages with 2 output transistors per channel. The largest, 28B-SST2, uses 32 output transistors for one channel, with differing numbers in the middle ranges of amplifiers depending upon the amount of peak and average current that might be required.

The different sizes of amplifier begin to depart from each other in measured performance only as they reach their respective maximum power levels.  In other words, at 60 Watts output all Bryston amplifiers measure very similar, and very low, THD.  As the output power reaches towards 100 Watts, then 150 Watts, then 300, 600 and 1000 Watts, the smaller amplifiers reach their clipping points and are passed in performance by successively larger amps in the lineup. Thus in terms of noise and distortion performance at low power levels, all Bryston amps are very similar to each other.

However, beyond being capable of higher output levels, louder crescendi and wider dynamic range, the larger amplifiers also benefit from the improved load regulation, (‘solidity’), of their power supplies.  It’s simply a fact of the laws of physics that larger transformers have lower internal resistance and thus they better maintain their steady-state Voltage levels under conditions of fluctuating current-draw than smaller transformers.  The very large transformers used in the largest Bryston amps also have significant energy storage in their magnetic cores, reducing the fluctuation of current drawn from the wall.  Larger banks of filter capacitors necessarily used in larger amplifiers hold greater amounts of stored energy, with energy storage going up as the square of the Voltage rating.  All this adds up to the larger Bryston amplifiers showing better focus of the virtual stereo ‘image’ and somewhat deeper sound-field as their power ratings go up.  The differences are subtle, and all Bryston amplifiers have been designed to provide very smooth, lucid and transparent musical reproduction, but I won’t try to pretend that there is no reason other than louder output levels to use a large, powerful amplifier in your optimum music system.  The hugely powerful Bryston 28B-SST2 has been rated by numerous reviewers around the globe as being the most ‘believable’, most focused and transparent, and least ‘identifiable’ amplifier in the world, bar none.

I hope the above is helpful, but please don’t hesitate to contact me if you have any other questions. Thanks for thinking of Bryston!

Chris Russell

I have some follow-up questions related to this response.  What are the dynamic headroom / peak transient output power specifications on Bryston amplifiers?  Does the headroom and/or the ratio of RMS vs. peak power differ on the various models?  How do the distortion figures stand-up on the various amplifier models when considering transient peak power levels? 
 
Since dynamic headroom is a key advantage of a quality amplifiers - has Bryston considered publishing headroom / peak power specifications on it's amplifiers?   

[Laundrew]

 "...and least ‘identifiable’ amplifier in the world, bar none."

Personally, I think this to be one of the best analogies that I have heard about any amplifier.

Be well...

[HooStat]

From an earlier post by James, they all operate in class A up to about 5% of their rated power.  (I don't know whether this is single ended class A or push-pull class A, or what the differences are, or any of the nuances of class A.)

[SoundGame]

From an earlier post by James, they all operate in class A up to about 5% of their rated power.  (I don't know whether this is single ended class A or push-pull class A, or what the differences are, or any of the nuances of class A.)

Thanks HooStat - I do remember that post on their class A power; however, with all respect that has no relation to dynamic headroom.  My Bryston 4B-SST2 is rated at 300 wpc RMS into 8 ohms from 20-20,000 Hz but that doesn't tell me what power it can peak at on transients - which is typically quoted as Dynamic Headroom in dB or in terms of Peak Transient Power.  Cheers.

[tim92gts]

Thanks HooStat - I do remember that post on their class A power; however, with all respect that has no relation to dynamic headroom.  My Bryston 4B-SST2 is rated at 300 wpc RMS into 8 ohms from 20-20,000 Hz but that doesn't tell me what power it can peak at on transients - which is typically quoted as Dynamic Headroom in dB or in terms of Peak Transient Power.  Cheers.

Isn't peak power on transients dependent on peak current delivery?
I thought the 7 / 14 BSST2 was around 150 Amps peak per channel.
i guess this would only happen when the voltage is peak and impedance is low,
also maybe only for a very low number of wavelengths.
It can't be possible that a pair of 14Bs will output 0.36 MW into 4Ohms though?

[SoundGame]

I have some follow-up questions related to this response.  What are the dynamic headroom / peak transient output power specifications on Bryston amplifiers?  Does the headroom and/or the ratio of RMS vs. peak power differ on the various models?  How do the distortion figures stand-up on the various amplifier models when considering transient peak power levels? 
 
Since dynamic headroom is a key advantage of a quality amplifiers - has Bryston considered publishing headroom / peak power specifications on it's amplifiers?   

To provide an example, in relation to my outstanding questions - NAD quotes dynamic IHF power ratings, for their amplifiers.  Their M2 is quoted as having a dynamic peak power of twice the continuous RMS power, which is a 3 dB Headroom:
 
NAD M2 Direct Digital Amplifier:   

• 2 X 250W Continuous Power at 8 and 4 Ohms
• >500W IHF Dynamic Power

[Mike Pickett]

Peak Transient Power is dependent on a number of factors, but the theoretical limiting factor would be the rail voltage of the amp.  It doesn't matter how short the transient is, or how much current you can back it up with; the output stage cannot deliver more voltage than the internal power supply can provide.  Given a 4BSST2, for example, the rail to rail voltage would be around 170V. 

If you make a handful of fictional assumptions regarding output devices that have no voltage drop and no current limit, a power supply that can deliver unlimited current, and a driver stage that can supply the required voltage drive, a 4B could theoretically put out no more than 170V peak to peak.  This works out to around 60V RMS, which would put 450 Watts across an 8 Ohm load, with 7.5 Amps of current being delivered.  This same 'perfect' amp would be able to put 900 Watts across a 4 Ohm load, delivering 15 Amps of current.  You can continue halving the impedance and doubling the current until you reach infinite current across exactly 0 Ohms, at which point the universe will wink out of existence...

In a real amp, these assumptions can't be made, and as you draw current from the amp, the rail voltages drop and peak output is reduced.  We also incorporate a current limiter on the output stage which is well beyond what any speaker can demand, but which will protect the amp if something bad happens, and this will kick in when a 4B attempts to drive 0.5 Ohms.

I'll try to take some time to run some tests later today, and provide some actual numbers for short term power delivery.  I'm guessing that the 8 Ohm rating will be close to the theoretical max worked out above, minus some allowance for a couple of diode drops, etc...

[SoundGame]

Peak Transient Power is dependent on a number of factors, but the theoretical limiting factor would be the rail voltage of the amp.  It doesn't matter how short the transient is, or how much current you can back it up with; the output stage cannot deliver more voltage than the internal power supply can provide.  Given a 4BSST2, for example, the rail to rail voltage would be around 170V. 

If you make a handful of fictional assumptions regarding output devices that have no voltage drop and no current limit, a power supply that can deliver unlimited current, and a driver stage that can supply the required voltage drive, a 4B could theoretically put out no more than 170V peak to peak.  This works out to around 60V RMS, which would put 450 Watts across an 8 Ohm load, with 7.5 Amps of current being delivered.  This same 'perfect' amp would be able to put 900 Watts across a 4 Ohm load, delivering 15 Amps of current.  You can continue halving the impedance and doubling the current until you reach infinite current across exactly 0 Ohms, at which point the universe will wink out of existence...

In a real amp, these assumptions can't be made, and as you draw current from the amp, the rail voltages drop and peak output is reduced.  We also incorporate a current limiter on the output stage which is well beyond what any speaker can demand, but which will protect the amp if something bad happens, and this will kick in when a 4B attempts to drive 0.5 Ohms.

I'll try to take some time to run some tests later today, and provide some actual numbers for short term power delivery.  I'm guessing that the 8 Ohm rating will be close to the theoretical max worked out above, minus some allowance for a couple of diode drops, etc...

Thanks Mike for taking the time out to lab test this - I look forward to your results for the 4BSST2.  Your explanation is very enlightening.  It would be interesting to apply these theoretical limits to the specs that other manufacturers claims in their specifications - especially in cases where they disclose their rail voltage. 

[*Scotty*]

During the 70s many amplifier manufacturers used the IHF spec to inflate the power output numbers their amplifiers were claimed to have.
 In 1974 the FTC passed its Amplifier Rule to combat this practice. The FTC mandated that continuous power measurements be made with sine wave signals on advertising and specification citations for amplifiers sold in the US.
 Anyone resorting to citing IHF power measurements at this late date is pretty lame.
See the link to the Audio Power wiki below for more information.
http://en.wikipedia.org/wiki/RMS_power
Scotty

[Mike Pickett]

Alrighty... got some numbers here.

In the interest of getting as close to the theoretical max of the amp, I used a very conservative 1:200 ratio of signal to silence to test the 4BSST2.  Signal was 2KHz Sine Burst, single cycles.

At 8 Ohms the amp delivered 435 Watts, at 4 Ohms 845 Watts, and at 2 Ohms I measured 1580 Watts.  I would expect these numbers to drop as the ratio of signal to silence goes up, so I tried it.

At 2 Ohms, I measured the following results for these signal to silence ratios.

1:200 - 1580 Watts
1:100 - 1549 Watts
1:50 - 1509 Watts
1:25 - 1475 Watts
1:10 - 1350 Watts
1:5 - 1219 Watts
1:3 - 1074 Watts

Just for laughs, I tried it at 1 Ohm as well.  With the 1:200 signal to silence ratio I measured over 2650 Watts, which corresponds to just over 51 Amps of current.

Hope this helps...

Mike

[*Scotty*]

Mike not to unduly complicate your life but is there any chance you could measure the rail voltage with a 1kHz sine-wave producing 1 watt into 8ohms and the same test of the rail voltage at 300 watts into 8 ohms. I was wondering how much the rail sagged at full rated power vs 1 watt.
 Or another way of putting this is what the load regulation looked like on the transformer.
I figured it was probably about 4% as Bryston has a reputation for conservative design.
Scotty

[SoundGame]

Alrighty... got some numbers here.

In the interest of getting as close to the theoretical max of the amp, I used a very conservative 1:200 ratio of signal to silence to test the 4BSST2.  Signal was 2KHz Sine Burst, single cycles.

At 8 Ohms the amp delivered 435 Watts, at 4 Ohms 845 Watts, and at 2 Ohms I measured 1580 Watts.  I would expect these numbers to drop as the ratio of signal to silence goes up, so I tried it.

At 2 Ohms, I measured the following results for these signal to silence ratios.

1:200 - 1580 Watts
1:100 - 1549 Watts
1:50 - 1509 Watts
1:25 - 1475 Watts
1:10 - 1350 Watts
1:5 - 1219 Watts
1:3 - 1074 Watts

Just for laughs, I tried it at 1 Ohm as well.  With the 1:200 signal to silence ratio I measured over 2650 Watts, which corresponds to just over 51 Amps of current.

Hope this helps...

Mike

Hey Mike - I converted your figures into Headroom measures for both 8 and 4 ohms.  For the 4B-SST2 they stand as the following:
 
1.6 db @ 8 ohm
2.3 db @ 4 ohm
 
Thanks Mike!  This compares favourably with some another Canadian audio product manufacturer i.e. Anthem - Sonic Frontiers / Paradigm.  On their amplifiers they share a headroom specification.  I thought their measurement was very conservative - given some claims for amplfiers having 3 dB to 6 dB of headroom.
 
Have a great weekend!

[Mike Pickett]

Mike not to unduly complicate your life but is there any chance you could measure the rail voltage with a 1kHz sine-wave producing 1 watt into 8ohms and the same test of the rail voltage at 300 watts into 8 ohms. I was wondering how much the rail sagged at full rated power vs 1 watt.
 Or another way of putting this is what the load regulation looked like on the transformer.
I figured it was probably about 4% as Bryston has a reputation for conservative design.
Scotty

4% would be nice... I measured 171V rail to rail at 1 Watt, and 150V at 300 Watts.