You are on record that you only 'hear what you hear' if you can see what makes the sound.
jan didden
I think you meant this one:
http://www.hypex.nl/docs/Bruno Masterclass/slides.htm
Working in the quote mine?
capacitor DA is of 1st order significance in sample/hold circuits, V coeff, distortion in filter circuits can be measured - materials science even gives some guidance on molecular structural causes – the question of audibility is often way below DBT thresholds but some poor application choices could be expected to be audible – so tossing this up is just chaff to distract from the argument
"flat feedback factor" vs low corner frequency, falling feedback factor "high feedback" effects on PIM have been explored (after Otala's raising the issue in audio design) in theory, by measurement, in sim and the results simply do not support the position you repetitively put forth - that PIM is the cause of your sighted listening preference for flat feedback factor amps that Otala prescribed
the sole evidence for PIM being an audible issue is your assertion - the follow on analysis and measurement clearly showed Otala's prescription was not the sole route to low PIM
high feedback circuits can be designed to be superior in this metric while employing falling feedback factor over the audio range
your and others hearing a difference that you ascribe to flat feedback factor may be "objectively true" - but the link to Otala's theory of PIM isn't proven even if we accept the sighted listening correlation as valid
if you want to live up to your own "live and learn" statement then you should demonstrate how the follow-on analysis in peer reviewed engineering journals is technically flawed - without the usual digression into decades old personality/political conflicts (including the technically vacuous "letter" you have alluded to in the past)
just straight up technical argument - can you do it?
Or present new evidence that PIM/FM IMD is orders of magnitude more audibly objectionable than “AM” IMD products
Or learn to quit asserting the link to PIM as a known fact – “we don’t know” and “theory and measurement suggest PIM is less likely an explanation of audible differences than initially assumed” seem to me to be more fair statements of the current understanding of the subject
"flat feedback factor" vs low corner frequency, falling feedback factor "high feedback" effects on PIM have been explored (after Otala's raising the issue in audio design) in theory, by measurement, in sim and the results simply do not support the position you repetitively put forth - that PIM is the cause of your sighted listening preference for flat feedback factor amps that Otala prescribed
the sole evidence for PIM being an audible issue is your assertion - the follow on analysis and measurement clearly showed Otala's prescription was not the sole route to low PIM
high feedback circuits can be designed to be superior in this metric while employing falling feedback factor over the audio range
your and others hearing a difference that you ascribe to flat feedback factor may be "objectively true" - but the link to Otala's theory of PIM isn't proven even if we accept the sighted listening correlation as valid
if you want to live up to your own "live and learn" statement then you should demonstrate how the follow-on analysis in peer reviewed engineering journals is technically flawed - without the usual digression into decades old personality/political conflicts (including the technically vacuous "letter" you have alluded to in the past)
just straight up technical argument - can you do it?
Or present new evidence that PIM/FM IMD is orders of magnitude more audibly objectionable than “AM” IMD products
Or learn to quit asserting the link to PIM as a known fact – “we don’t know” and “theory and measurement suggest PIM is less likely an explanation of audible differences than initially assumed” seem to me to be more fair statements of the current understanding of the subject
Last edited:
This?
/quote
Predictably, Otala's 'discovery' only added welcome fuel to the fire of those who were quite convinced that these new-fangled transistor amplifiers were genetically incapable of pleasing the audiophile ear, a notion which, like all myths (italic mine-jd), persists unabated, in spite of the fact that the root cause of TIM is now completely understood, and can easily be avoided in transistor amplifiers expressly designed for the most demanding audio applications.
/unquote
jan didden
Barrie can be rude on occasion. How about Barrie calling Otala: 'Mister Otala' ???
It just so happens that Matti Otala has a real PhD, and Barrie has an honorary one. This is an implied insult. It is just too obvious to be an oversight. However, everyone, it gets better, read on.
It just so happens that Matti Otala has a real PhD, and Barrie has an honorary one. This is an implied insult. It is just too obvious to be an oversight. However, everyone, it gets better, read on.
Last edited:
Hi John,
There was really nothing new to Barrie's analysis beyond what was published many years earlier, as in the paper I did on PIM. There is nothing wrong with Barrie's anaysis - he shows how negative feedback can lead to PIM, as Otala and I did. There is no argument there. However, he did not assert or show that low open-loop bandwidth made PIM worse. In fact, it does not. This may be non-intuitive to some.
PIM is really fairly simple. When a signal excursion causes the phase shift through an amplifier to change, PIM has been created. This is basically amplitude to phase conversion. Long before Otala, it was referred to as differential gain and phase in the video community where it really mattered in color TV.
If the 3 dB bandwidth of an amplifier changes with signal swing, PIM is created through the simple mechanism of a bandwidth corner causing phase shift at frequencies well below the corner frequency. Anything that causes the bandwidth of an amplifier to be modulated will cause PIM. Note that we are talking about closed loop bandwidth in the case of a feedback amplifier.
Even amplifiers without feedback can generate PIM. For example, the nonlinear collector-base capacitance in a VAS stage in a no-feedback amplifier can cause PIM. This can also cause PIM in the open-loop of a feedback amplifier. This cause of PIM was not discussed by Barrie.
All Barrie was really doing was showing that incremental gain changes with signal in the input stage (amplitude intermodulation distortion) caused the open-loop gain of the amplifier to change, and thus to cause the negative feedback gain crossover frequency to change. This in turn moves the closed-loop corner around and results in PIM. This mechanism works the same whether the amplifier has low or high open-loop bandwidth as long as the nominal closed loop bandwidth is the same in both cases.
So there are two causes of PIM. One naturally in the open loop of an amplifier, and one brought on by negative feedback. A key thing to keep in mind is that there can be no feedback-generated PIM without significant amounts of AIM in the first place. Secondly, and interestingly, although NFB can cause the generation of some PIM due to AIM (amplitude-to-phase conversion), the NFB actually acts to reduce PIM that was in the open loop in the first place, as via nonlinear collector-base capacitance. There is thus not necessarily an increase in total PIM when NFB is applied.
Finally, it is important to recognize that an amplifier with PIM will show picket-fence distortion sidebands in the 19+20 kHz CCIF IM test, just like HF AIM will. Most of this is covered in my book.
Cheers,
Bob
Well, now that we have gotten the conclusions of the critics, why don't we all read what Barrie has to say, and look at a worked out example of PIM? Please remember the 3 stages of an idea. ;-). When Matti Otala put forth a purely mathematical model of PIM in 1980 or so, and tried to put it in the JAES, it was rejected for being too 'revolutionary'. Then, in 1995 when Barrie got around to studying TIM and PIM, he found something interesting. Now, it is and always has been OBVIOUS to all real engineers.
This is most probably WHY video amps like the AD825 sound so good, even though they have limited acceptance in the audio marketplace.
This is most probably WHY video amps like the AD825 sound so good, even though they have limited acceptance in the audio marketplace.
Well, many of us have used the AD825 for about 7 years 
This part is very good for audio, in case you do not require low noise.
But, it is intended for audio (not only for video), as listed in datasheet:
APPLICATIONS
CCDs
Low distortion filters
Mixed gain stages
Audio amplifiers
Photo detector interfaces
ADC input buffers
DAC output buffers
This part is very good for audio, in case you do not require low noise.
But, it is intended for audio (not only for video), as listed in datasheet:
APPLICATIONS
CCDs
Low distortion filters
Mixed gain stages
Audio amplifiers
Photo detector interfaces
ADC input buffers
DAC output buffers
Last edited:
Well pointed out, PMA. I designed it into a power amp as a reference analog comparator, with output discretes to make it swing +/- 60V. Works extremely well for this task. Thanks again, Scott Wurcer, for sampling me these parts 2 years ago. I had them from years before, since when they first came out, but I did not think to try them for audio, until I had this project to design component modules for the power amp. We also used Bruno's class D power amps (at my recommendation) to make the tracking power supplies and reduce the heatsink area to virtually nothing with a 2 X 250W power amp. Incestuous bunch, aren't we? ;-)
Hi Charles,
In case you are looking in. This is the first chance I've had to respond to you.
This really only means that you've dealt with the circuits you are designing enough to remember what the performance should be like. I would have thought that you might play with quiescent current flows and voltage levels to optimize each stage since you aren't applying GNFB. The question was simply out of my own interest, not designed to trip you up in any way.
As for distortion measurements, a THD measurement is next to useless to the average person. But then, with the state of the current retail network and staff knowledge, so are listening tests. I hate to say this, but after observing some presentations out in the wild, most times what a customer is told is pure fiction. A sale is made mostly on the presented facts a customer is exposed to, along with price, appearance and lastly, performance. I expect the higher priced equipment might frighten salespeople enough to have them fly straight.
-Chris
In case you are looking in. This is the first chance I've had to respond to you.
That's cool then. It did really seem like you had something against the guy. I'm only mentioning this because he is not in attendance. It seemed unfair.
Also cool.
This really only means that you've dealt with the circuits you are designing enough to remember what the performance should be like. I would have thought that you might play with quiescent current flows and voltage levels to optimize each stage since you aren't applying GNFB. The question was simply out of my own interest, not designed to trip you up in any way.
I've seen and used equipment all over the map where THD etc. is concerned. I have no real preference for a number as a high level limit. One of my favorite amplifiers in use here is a tube model (old school - Eico HF-87 stock-o).
If you knew where I stood on these things, you wouldn't have to ask. I've been extremely clear and consistent over the years. Not that you should even notice me here. I suspect you've been talking to "Sir. John" here. Either that or you're paranoid, and there is no evidence of that. I haven't heard any of your products, so there is no way I'm about to comment on them. Just to be clear. I also don't have any agenda beyond what I've asked you for interests sakes.
That's entirely reasonable, and much more like the response I expected. Thank you Charles.
That is actually something I have both skill and training in. No, we don't use amplifiers as a standard.
Charles, on this I completely agree! Thermographs of operating (warmed up) amplifiers would really allow the average person to avoid poorly designed products. That's the technician in me speaking. A manufacturer would also have a more accurate idea of how badly their warranty program might be hit by certain models.
As for distortion measurements, a THD measurement is next to useless to the average person. But then, with the state of the current retail network and staff knowledge, so are listening tests. I hate to say this, but after observing some presentations out in the wild, most times what a customer is told is pure fiction. A sale is made mostly on the presented facts a customer is exposed to, along with price, appearance and lastly, performance. I expect the higher priced equipment might frighten salespeople enough to have them fly straight.
-Chris
Hi John,
The same can be said the other way. How much evidence will you accept that runs counter to what you have assumed for years?
I'm not saying you are wrong, or right for that matter. It's just that if you expect blind acceptance from anyone else, are you prepared to also accept another view? Obviously the answer is no. You want to walk around and "kick the tires". You want to get comfortable with what is being said before you would even consider any other viewpoint. The exact same is true for anyone who may not share your viewpoint.
Subjectivity can be a plus or minus, depending on how you came about that opinion / belief. The more closed minded a person is, the more they will defend their point of view as they would a religion in the dark ages. Someone closed in their beliefs can be recognized by others, and will tend to be avoided. The reasoning would seem to be that if a person has ceased to examine information, what they believe is most probably out of date (since 1968? ). Just kidding John.
How much are you willing to try in your lab?
Now that we've beat our chests, do you feel better? Don't care myself.
But why would you even bring this up? No implications were made that you or anyone else (Charles I guess) fails to measure anything. I was curious, Charles answered and I already know how you go about things - you didn't have any need to reply to this. I asked Charles. I should mention that I've never been critical of your methods, but you do come out with some bizarre statements from time to time.
Take care John, relax. -Chris
The same can be said the other way. How much evidence will you accept that runs counter to what you have assumed for years?
I'm not saying you are wrong, or right for that matter. It's just that if you expect blind acceptance from anyone else, are you prepared to also accept another view? Obviously the answer is no. You want to walk around and "kick the tires". You want to get comfortable with what is being said before you would even consider any other viewpoint. The exact same is true for anyone who may not share your viewpoint.
Subjectivity can be a plus or minus, depending on how you came about that opinion / belief. The more closed minded a person is, the more they will defend their point of view as they would a religion in the dark ages. Someone closed in their beliefs can be recognized by others, and will tend to be avoided. The reasoning would seem to be that if a person has ceased to examine information, what they believe is most probably out of date (since 1968?
). Just kidding John.How much are you willing to try in your lab?
I thought it was interesting that a Japanese manufacturer decided to do this. It might be worth you playing with a couple to see how they perform in the real world compared to a signal circuit of your own design. Hearing is believing, right?
BTW, just as a response, my gear valued in the same manner easily exceeds $50K as well. So?
Now that we've beat our chests, do you feel better? Don't care myself.
But why would you even bring this up? No implications were made that you or anyone else (Charles I guess) fails to measure anything. I was curious, Charles answered and I already know how you go about things - you didn't have any need to reply to this. I asked Charles. I should mention that I've never been critical of your methods, but you do come out with some bizarre statements from time to time.
And we've been hearing about what you've done way back in 1968 for years. Can we get onto some newer engineering feats, or is this it?
Take care John, relax. -Chris
Please give it a break, Anatech. I don't know you, and you don't know me. So far, we are even on that score. I am continually berated for not measuring, yet I have to state that I have purchased measuring equipment for THAT VERY PURPOSE for the last 50 years. Yet, I don't use it as much as I used to. Why? Because I have found it to be only partially useful in predicting audio success with the products that I design. However, WHEN I CAN use my test equipment for something really useful, I fire it up, willingly. Most here don't even have test equipment, but just use simulation, is that enough?
- Status
- Not open for further replies.