Benchmark...first watt....ABX...facepalm!

[restorer-john]

I do! Always good to validate that simulation can approach reality

I'll do it after I've eaten lunch (starving).

Maybe I'll record the signals via the A/D to listen to as well.

I can match the scale, Benchmark have the residual 60dB down (X 1000). Obviously there is noise too that far down, but scope averaging helps (or the analog scope) The Benchmark "scope" shot I believe is a complete fabrication, simply due to the fact there is no noise, just a clean signal. That's not what analogue scope shots at that level look like. But we'll do both a DSO and CRO comparison as I have both on my bench.

 

[restorer-john]

I'll do it after I've eaten lunch (starving).

Maybe I'll record the signals via the A/D to listen to as well.

I can match the scale, Benchmark have the residual 60dB down (X 1000). Obviously there is noise too that far down, but scope averaging helps (or the analog scope) The Benchmark "scope" shot I believe is a complete fabrication, simply due to the fact there is no noise, just a clean signal. That's not what analogue scope shots at that level look like. But we'll do both a DSO and CRO comparison as I have both on my bench.

OK, well that was a bit of a waste of time. I picked a random amplifier here, a nice Rotel RA-970BX. Rated 80wpc@8R THD<0.03% at full power.

Its THD was less than 0.002% at 80wpc and it will put out 100wpc. (200kHz bandwidth) The signal source has an inherent THD of ~0.0015% itself.

To match the Benchmark 'test' I tried a 282mV (0.01W) output into 8R, but the noise with respect to that output is too great and the analyzer residual shows nothing useful, but certainly no X/over distortion.

At 1V@8R output, the distortion is completely buried in the noise on the output of the THD analyzer. There are no crossover spikes and even amplifying the residual and running a 1024 averaging on the scope shows nothing of note- no spikes. I backed the bias right down, but the range is such that there is still enough bias (and NFB) to essentially leave everything still buried in the noise floor.

This is the distortion residual amplified by 1000 times (60dB) with a 1024 averages to kill the noise.

I could say the THD+N at 2.83V@8R output is (left) 0.047% and (right) 0.044%, but it is all N. I cannot see on either scope any hamonics at all and certainly no crossover distortion. I have no reason to doubt Rotel's spec as being 250mW to rated output.

So I need an amplifier with an extremely low residual noise floor, then kill the bias altogether, view the spikes and send the analyzer output after the notch to the FFT.

 

[pjug]

OK, well that was a bit of a waste of time. I picked a random amplifier here, a nice Rotel RA-970BX. Rated 80wpc@8R THD<0.03% at full power.

View attachment 181210

Its THD was less than 0.002% at 80wpc and it will put out 100wpc. (200kHz bandwidth) The signal source has an inherent THD of ~0.0015% itself.

To match the Benchmark 'test' I tried a 282mV (0.01W) output into 8R, but the noise with respect to that output is too great and the analyzer residual shows nothing useful, but certainly no X/over distortion.

At 1V@8R output, the distortion is completely buried in the noise on the output of the THD analyzer. There are no crossover spikes and even amplifying the residual and running a 1024 averaging on the scope shows nothing of note- no spikes. I backed the bias right down, but the range is such that there is still enough bias (and NFB) to essentially leave everything still buried in the noise floor.

This is the distortion residual amplified by 1000 times (60dB) with a 1024 averages to kill the noise.

View attachment 181209
View attachment 181206

I could say the THD+N at 2.83V@8R output is (left) 0.047% and (right) 0.044%, but it is all N. I cannot see on either scope any hamonics at all and certainly no crossover distortion. I have no reason to doubt Rotel's spec as being 250mW to rated output.

So I need an amplifier with an extremely low residual noise floor, then kill the bias altogether, view the spikes and send the analyzer output after the notch to the FFT.

When you set bias to minimum, do any high order odd harmonics show up on a FFT?

 

[restorer-john]

When you set bias to minimum, do any high order odd harmonics show up on a FFT?

I'll check in the morning. 10:49pm here down under. Cheers. John

 

[DSJR]

I used to like the Rotel. More 'solid feeling' to use than an equivalent grey NAD (ok, all visual), but a good sensible amp at sensible money and matching power amp for active/bi-amp use too. Nice to see the performance is still pretty much up there after twenty or more years.

 

[pjug]

I don't know why I didn't think of this to begin with... There is a simpler and more effective way to simulate with Audacity. I just mixed a 1kHz sine wave and 1 kHz square wave to make a sine wave with crossover distortion. Then, notching out the 1KHz in the mix while playing with notch filter Q, I was able to get a distortion waveform that looks pretty similar to the one in the Benchmark scope shot.

edit: now my question is whether the Benchmark distortion waveform should be closer to a square wave, and just has some 1kHz mixed in because of limitations with the notch filter that they use?

 

[DonH56]

OK, well that was a bit of a waste of time. I picked a random amplifier here, a nice Rotel RA-970BX. Rated 80wpc@8R THD<0.03% at full power.

View attachment 181210

Its THD was less than 0.002% at 80wpc and it will put out 100wpc. (200kHz bandwidth) The signal source has an inherent THD of ~0.0015% itself.

To match the Benchmark 'test' I tried a 282mV (0.01W) output into 8R, but the noise with respect to that output is too great and the analyzer residual shows nothing useful, but certainly no X/over distortion.

At 1V@8R output, the distortion is completely buried in the noise on the output of the THD analyzer. There are no crossover spikes and even amplifying the residual and running a 1024 averaging on the scope shows nothing of note- no spikes. I backed the bias right down, but the range is such that there is still enough bias (and NFB) to essentially leave everything still buried in the noise floor.

This is the distortion residual amplified by 1000 times (60dB) with a 1024 averages to kill the noise.

View attachment 181209
View attachment 181206

I could say the THD+N at 2.83V@8R output is (left) 0.047% and (right) 0.044%, but it is all N. I cannot see on either scope any hamonics at all and certainly no crossover distortion. I have no reason to doubt Rotel's spec as being 250mW to rated output.

So I need an amplifier with an extremely low residual noise floor, then kill the bias altogether, view the spikes and send the analyzer output after the notch to the FFT.

I suppose you could try reducing the feedback. I do wonder just how severe a problem crossover distortion is in a decent modern amplifier from... anyone. I suspect, but do not know, that feedback factor (loop gain) and loop bandwidth are much greater than in years (decades) gone by, so the additional feedback that gets all these 0.0nm% distortion figures have effectively obviated crossover distortion as an audible concern.

 

[tomelex]

I suppose you could try reducing the feedback. I do wonder just how severe a problem crossover distortion is in a decent modern amplifier from... anyone. I suspect, but do not know, that feedback factor (loop gain) and loop bandwidth are much greater than in years (decades) gone by, so the additional feedback that gets all these 0.0nm% distortion figures have effectively obviated crossover distortion as an audible concern.

Yes agreed DonH56, reducing feedback is one way, or instead of taking feedback from the output, take it from a drive stage output, then what is left will be the output stage and its crossover distortion, but we know it's there, as shown and proven by some of Douglas Selfs work shown just a few posts before here. Seems a waste of time really though at this point.

 

[pjug]

So I did a quick listen of the 1KHz tone with and without crossover distortion at 53dB below the fundamental, and playing at 67dB. I have to say it is surprisingly easy to hear it, so Benchmark is right about that. You just have to adjust your head position to hear it. Move your head a few inches and the distortion fades away. I am not sure ABX is worth doing; I think everyone would pass if they listen to the distortion to know what to listen for, then get an ear in a good location.

Of course, the facepalm for the Benchmark Application Note is still deserved since an amp with this much crossover distortion would not be typical.

 

[rwortman]

I was troubleshooting an NAD receiver a few years back for channel imbalance. I was driving it from a sinewave generator. The output wave looked beautiful on both channels although one was slightly lower In amplitude. The output of the driver on one channel looked like it had about 25% distortion. I was astonished. Turned out the output transistor bias was way off. Crossover distortion of class B and then some. The feedback loop was taking care off all of it, at least what I could see by eye on the ‘scope.