john curl said:
Classic ad hom.
Guys, let's keep the arguments technical instead of directed toward each other's credentials, abilities, or whatever.
PIM, yet another fancy TLA?
And, as a bonus, opposed to the classical Miller compensation, no TIM to worry about.
PS: Sorry for the delayed response.
G.Kleinschmidt said:
And, as a bonus, opposed to the classical Miller compensation, no TIM to worry about.
PS: Sorry for the delayed response.
Hello guys
It looks like for 3 average posters here you'll get 4 different opinions on Otala's legacy.
C'mon, that's nearly 40 years old knowledge, well digested and corrected by Cordell (cheers!), Cherry and others. After reading them you get enough opinions to make your own out of those.
Anyway, the man from the thread title deserves some respect, at least for historical point of view and for making claims engineers would quarrel about 40 years later.
It looks like for 3 average posters here you'll get 4 different opinions on Otala's legacy.
C'mon, that's nearly 40 years old knowledge, well digested and corrected by Cordell (cheers!), Cherry and others. After reading them you get enough opinions to make your own out of those.
Anyway, the man from the thread title deserves some respect, at least for historical point of view and for making claims engineers would quarrel about 40 years later.
john curl said:
I have read the other side of the argument and the explanation given by Bob Cordell still stands. All the Otala paper suggests is that applying negative feedback around an otherwise PIM free amplifier does create PIM. It says that in its conclusion. The paper does not prove that increasing feedback, increases the amount of PIM although it does allude to that effect without actually quantifying it. This is where you are taking the results of this analysis way out of context and drawing the wrong conclusion.
regards
Trev
After Otala, you then read what Barrie Gilbert had to say about the subject. Done that yet, oh engineers? 
Trevor, do you know my record on making power amplifiers? Have you compared the MEASURED results of my JC-1 power amp to ANY Halcro power amp? Yes, due to lots of feedback, the Halcro measures somewhat lower in distortion. Any other attributes? How about rated power, slew rate, high frequency output impedance, peak current, and cost?
Is Dr. Candy's amp a combination of Doug Self with a Hawksford (Cordell) output stage? Even HK knew that there were better input topologies to choose from, back in 1980.
Isn't the real effort in the amp in the interesting switching power supply, that doesn't seem to make the unit much smaller in size, and limits the peak current to 15A or so? Hasn't this power supply been implicated in effecting surrounding audio components, even when it is not directly connected in the audio system being used, but just operating in a warm-up mode in the same room? Trevor, you might note that I may a little something about Halcro. Yet, I treat and respect Dr. Bruce Candy as a colleague, even though I may have never met him. Good work Dr. Candy, just don't bring out your ruler, until you have looked at the competition.
Is Dr. Candy's amp a combination of Doug Self with a Hawksford (Cordell) output stage? Even HK knew that there were better input topologies to choose from, back in 1980.
Isn't the real effort in the amp in the interesting switching power supply, that doesn't seem to make the unit much smaller in size, and limits the peak current to 15A or so? Hasn't this power supply been implicated in effecting surrounding audio components, even when it is not directly connected in the audio system being used, but just operating in a warm-up mode in the same room? Trevor, you might note that I may a little something about Halcro. Yet, I treat and respect Dr. Bruce Candy as a colleague, even though I may have never met him. Good work Dr. Candy, just don't bring out your ruler, until you have looked at the competition.
Trevor, have you analyzed the good, the bad, and the 'ugly' aspects of SMPTE IM?
First do it, and we can talk about Cordell's PIM measurement method.
First do it, and we can talk about Cordell's PIM measurement method.
jcx, I appreciate your input, but I don't think that this is the forum to discuss this, as I am under a cloud, here, so to speak. Things will just progress too slowly in this environment. Please feel free to E-mail me.
john curl said:Trevor, do you know my record on making power amplifiers? Have you compared the MEASURED results of my JC-1 power amp to ANY Halcro power amp? Yes, due to lots of feedback, the Halcro measures somewhat lower in distortion. Any other attributes? How about rated power, slew rate, high frequency output impedance, peak current, and cost?
Is Dr. Candy's amp a combination of Doug Self with a Hawksford (Cordell) output stage? Even HK knew that there were better input topologies to choose from, back in 1980.
Isn't the real effort in the amp in the interesting switching power supply, that doesn't seem to make the unit much smaller in size, and limits the peak current to 15A or so? Hasn't this power supply been implicated in effecting surrounding audio components, even when it is not directly connected in the audio system being used, but just operating in a warm-up mode in the same room? Trevor, you might note that I may a little something about Halcro. Yet, I treat and respect Dr. Bruce Candy as a colleague, even though I may have never met him. Good work Dr. Candy, just don't bring out your ruler, until you have looked at the competition.
At the end of the day Stereophile have given class A ratings to all of the Halcro amps that it has reviewed over the years irrespective of your criticisms. Now wasn't it you talking about getting poor reviews in Stereophile from clone amps of yours ??
Also nearly everyone on this forum knows a little bit about the Halcro. The reason being is that it has buck the trends and flies in the face of the guru philospohy of little or no global feedback. Sure it may have the supposed limitations that you outline such as current limits but that doesn't stop it being used with hard to drive Wilson and B&W speakers. At the end of the day the proof is in the eating of the pudding. The Halcros's always get good reviews both subjective and objective which leads me to believe that all of this talk about how much feedback that should be applied is just another distraction to getting the job done !!
regards
trev
A lot of stuff gets rave reviews in Stereophile. There are two different LP demagnetizers that appear on their recommended components list for example. A positive review in Stereophile doesn't mean a dang thing, except that maybe the dealer, distributor or manufacturer bought the reviewer an expensive dinner.
Trevor, I am not being critical of Halcro. It is you who is not giving me equal billing on the same stage, that is all. The real critic was Martin Colloms in 'Hi Fi News'. PLEASE read his review and his credentials before passing judgement, yet again.
Good morning to you Jan,
And: a massive amount of GNB is needed to mitigate it (maybe even more than 10dB extra, who knows?)
Solid state devices are unusable without some form of negative feedback. "Bad" was not my wording. GNF is, however, harmful, does not "remove" distortions surgically as it is widely assumed and should not be used unmindfully.
High open-loop gain and the related low open-loop and closed-loop bandwidth cannot result in transparent sound.
Distortions rise with decreasing bandwidth (the ideal level of bandwidth is infinite).
Capacitances raise distortions (the ideal value of capacitance in any location is zero).
People who complain about GNF, primarily mean the harmonic spectrum created, not the dynamic distortions.
Kind regards
Certainly, another reason is availability of better and for their task more appropriately chosen devices
Yes: for instance, the exceedingly inconvenient crossover distortion produced by low biased push-pull output stages.
And: a massive amount of GNB is needed to mitigate it (maybe even more than 10dB extra, who knows?)
OK, my statements can be be considered black-and-white, fragmentary, easy to misunderstand. Nevertheless, I try to avoid being vague or muddy.
Solid state devices are unusable without some form of negative feedback. "Bad" was not my wording. GNF is, however, harmful, does not "remove" distortions surgically as it is widely assumed and should not be used unmindfully.
High open-loop gain and the related low open-loop and closed-loop bandwidth cannot result in transparent sound.
Distortions rise with decreasing bandwidth (the ideal level of bandwidth is infinite).
Capacitances raise distortions (the ideal value of capacitance in any location is zero).
More explicitly, the transformed (low and high order) harmonics are "pushed" outside the range of audibility and measurement (but do have impact on the audible range, if someone believed differently).
People who complain about GNF, primarily mean the harmonic spectrum created, not the dynamic distortions.
Kind regards
homework
Zero capacitances, infinite bandwidth?
They would guarantee instability. 😀
Lumba Ogir said:
Zero capacitances, infinite bandwidth?
They would guarantee instability. 😀john curl said:
John it was you who bought up the class A rating of Stereophile not me. I am just stating the obvious so don't shoot the messenger.
The Halcro amplifier and the reputation that proceeds it is just but one example of a high negative feedback design amplifier that flies in the face of the negative feedback critics.
regards
Trev
Not in the least, any competent engineer could make a comparable design. We just don't bother to complicate our designs to that extent. You might notice that Halcro's best designs save no money, offer no more output, and really don't sound any better than comparable designs. Since, you are from Australia, perhaps the local press has given you a more one sided input.
For example, when 'Stereophile' had to evaluate a very expensive speaker, that had a VERY LOW overall drive impedance, they turned to the Parasound JC-1 to do the job, NOT Halcro.
This is objective reality, not marketing-sales, Aussie style or anywhere else.
To not understand Halcro's true position, you also impugn Nelson Pass, who is now 'top doggie' on the hydrant, according to 'Stereophile'.
For example, when 'Stereophile' had to evaluate a very expensive speaker, that had a VERY LOW overall drive impedance, they turned to the Parasound JC-1 to do the job, NOT Halcro.
This is objective reality, not marketing-sales, Aussie style or anywhere else.
To not understand Halcro's true position, you also impugn Nelson Pass, who is now 'top doggie' on the hydrant, according to 'Stereophile'.
John, I'm wondering what you think of null tests such as the Hafler to demonstrate that all forms of distortion are below a particular level.
Carver also nulled his amp in the Stereophile challenge, suggesting that high cost parts are not required to duplicate "high end" or the best sound.
Thoughts?
Pete B.
Carver also nulled his amp in the Stereophile challenge, suggesting that high cost parts are not required to duplicate "high end" or the best sound.
Thoughts?
Pete B.
Null tests are difficult, because of phase/ time problems, and don't tell as much about the sound of the actual component as you might think.
From actual experience, it is almost impossible to make a satisfactory differential test of just 2 caps. Just think about comparing a complete amp (with coupling caps, if even in the feedback loop) and a straight wire? Lots of compromises, to make it even somewhat measurable. This might be the problem with the test.
From actual experience, it is almost impossible to make a satisfactory differential test of just 2 caps. Just think about comparing a complete amp (with coupling caps, if even in the feedback loop) and a straight wire? Lots of compromises, to make it even somewhat measurable. This might be the problem with the test.