Soooo, what about enjoying-us with a sample of whatever source you like, and a copy of it across Blowtorch ?
You will take no risk with my low discerning ears, as i still enjoy listening so many awfull sounding CDs from my collection on my mid-fi system.
Yes Richard, that will work to remove any measurable distortion.
However, does it really sound good? That is an important question.
For example: I once designed a special 600 ohm audio driver for Sound Technology, sort of like an 811, that would give only a few thousands of percent or less at 100KHz and below at about 6V out. I did it by using a special design with 100KHz open loop bandwidth, and closing the loop at 12dB/octave. Worked just fine, FOR THE APPLICATION.
However, I would not necessarily recommend it for quality audio.
However, does it really sound good? That is an important question.
For example: I once designed a special 600 ohm audio driver for Sound Technology, sort of like an 811, that would give only a few thousands of percent or less at 100KHz and below at about 6V out. I did it by using a special design with 100KHz open loop bandwidth, and closing the loop at 12dB/octave. Worked just fine, FOR THE APPLICATION.
However, I would not necessarily recommend it for quality audio.
Thanks, however it is a little acrobatic, i would not use that in a industrial amp.
To deviate a little from the technical discussions.
I have a chance to listen to one of John Curl's products.
My present pre-amp is tube-based, Audible Illusions Modulus 3 + MC Phono. The MC Phono part is designed by JC, by the looks of it from the outside, it's constructed with discrete JFETs and MOSFETs.
All I can say is that LPs never before sounded so good on my audio setup.
Hat off to JC.
I have a chance to listen to one of John Curl's products.
My present pre-amp is tube-based, Audible Illusions Modulus 3 + MC Phono. The MC Phono part is designed by JC, by the looks of it from the outside, it's constructed with discrete JFETs and MOSFETs.
All I can say is that LPs never before sounded so good on my audio setup.
Hat off to JC.
Everyone knows this, it is trivial. So why you were asking the full scale for AD797 and distortion magnified 1000x??? And speaking about issues if any trace of 7th is seen then??? The reason why I am so disappointed with your reasoning is that you are misleading the readers.
You have different measures for the BT and AD797 (or another opamps).
Don't you think it's just a wee bit misleading to show spectra from a 797 with distortion magnification, high output voltage, and heavy load, then compare it to something you sell/sold tested with low output voltage, light load, and no distortion magnification?
Man up and admit that your accusation about relative 7th and 9th order distortion was mistaken and not based on any actual data. It's the honest thing to do.
Man up and admit that your accusation about relative 7th and 9th order distortion was mistaken and not based on any actual data. It's the honest thing to do.
Mainly silicon, i believe
I used to look, first, at the simplified schematics of OPAs provided in all the data-sheets in the good old days. It was like a first shake hands where you get a feeling of your interlocutor.
I wonder why they have disappeared from the datas of most of the recent ICs.
Perhaps, but I just like the concept -- which can be applied in many other ways, I'm sure. I do recognise it is another form of feedback overall and what ever implications that might have.
-R
Gentlemen please remember that the measurements shown by JC are made by Samuel Groner / Weiss.
For those interested that haven't seen the measurements, go to Samuel's homepage and download the (op_amp distortion.pdf), (too large to attach)
IMHO Samuel is trying as much as possible to kill all IC OpAmps, why?
Remember that he is working for WEISS and that they have a their own solution. Their own discrete OpAmp.
Think about it.
Stein
For those interested that haven't seen the measurements, go to Samuel's homepage and download the (op_amp distortion.pdf), (too large to attach)
IMHO Samuel is trying as much as possible to kill all IC OpAmps, why?
Remember that he is working for WEISS and that they have a their own solution. Their own discrete OpAmp.
Think about it.
Stein
Last edited:
It allows for another way to get around the issue of BW vs gain for lowering distortion in gnfb designs. Another tool in the designers arsenal. Similar could be (and maybe is) done within stages of amplifiers circuitry.
Looking back, IMO designers around the world tend to converge on what is best for a given app. If there is one best compromise, designers will find it over time and then it can become the popular way -- prices come down and gets copied and integrated and maybe into IC at low overall cost.
We seem to have done this in audio amplifiers to a large degree.... certain topologies et al.
Thx-RNMarsh
Last edited:
I hope everybody here knows this. The problem was that John had not described the method used, so I did that later to show comparison of apples with oranges.
As you know, in this thread we get purely marketing reasoning rather than engineering reasoning.
And that's fine- the problem is the inappropriate comparisons to John's own products tested in a different, much less rigorous way. As Scott suggested, use a datasheet circuit or some other reasonable one, or test his circuits the same way for comparison. John's assertion that ICs are "mid-fi" because of 7th and 9th order distortion is apparently totally unsupported by data- it's a pity that he doesn't seem to be able to say that, and either provide appropriate supporting measurements or admit that he's wrong.
For someone of his competence and stature, it's particularly disappointing.
Chatting with him over the years, I don't think that's his intention. He just wanted to extend and update the work done by Jung and to some extent Self, and accurately characterize the parts we have available today. Because they measure so well, this requires going to more extreme lengths in terms of noise gain and the use of available test equipment. Once one takes the results in that context and given understanding of how noise gain works, one may well conclude that many parts will work just fine in a given application.
Samuel's interests, besides extending the state-of-the-art in analog electronics, include musical performance (he's an accomplished pianist and conductor, among other things) and studio recording. I don't think the business of selling discrete opamps is of that much importance to him. And he would not intentionally influence his measurements to make other parts look bad. He has integrity.
SY,
I was trying to get to one of the important differences before all the noise set in. If THD were the only issue it would certainly be simpler. Pretty much THD below .01 - .001% is all you might need. However if you carry out Bruno P's analysis a bit further and do not assume a linear system you can get to distortion harmonic weighting. But that is still not all.
You can also add in both internal noise and the effects of external noise to get a bit closer. (In power amplifiers output behavior is a bigger deal than in preamps, but the issue never goes away.)
Adding noise properly by the RMS value is not as useful as looking at peak (frequency or period weighted) voltages (or power.)
You might be surprised by how much noise is even in a basic power supply. 120 dB PSSR looks impressive and in practice is actually a bit shy!
5% ripple of a lowly 15 volt IC supply or .75V -120 dB = .75 uV (Duh!) at 1 volt amplifier sensitivity and 20 db of headroom, .75uV/100mV = -102 db below program. Add a silent passage (-60) to the dynamic range or look at the weighted harmonic structure and you are just at what could be a problem. Power supply noise during what should be silence. Now could a trained listener pick that out. Possibly.
We could look at other amplifier sensitivities (or gain structures) and of course we would normally use lower 60 Hz ripple. However in many cases lower 60 Hz results in significant higher frequency content where the PSSR is reduced. Now how this folds back into the frequency range of interest is an open issue.
Of course some folks run their equipment on much higher unregulated voltages and have less PSSR. (We could get into the interesting difference between semiconductor and Edison effect rectifiers, but that might even make it into my article, if I ever get it finished.)
ES
Huh? Unless you are redefining the "ripple" concept, the 50p 7815 (hope this qualifies as "a lowly 15 volt IC supply") has 70-80dB ripple rejection @120Hz.
Yes, but that's a red herring- no one but you mentioned THD, John specifically claimed superiority for 7th and 9th harmonic and as it turns out, apparently has nothing.
I used the same concept for my protection circuit and as a dynamic distortion magnifier to look and listen to them.
I think this idea could help to design an OPA, with a reduced open loop gain and a high open loop bandwidth, in order to reduce further the distortion.
But i don't know if it is possible to fine tune enough resistances values on the wafer.
There is several known tracks for error corrections, i wonder why they are underused.
http://www.tubecad.com/2010/04/04/Tringlotron.png
http://www.diyaudio.com/forums/head...ased-tringlotron-amplifier-2.html#post2154368
John asked how it sound.
My thoughts:
First, you can only remove around 20db of distortion this way (but it is not so bad). you are limited by stability.
Second, the good one, the OPA i used (if you use some) does not add its signature.
Third, as i tested it on a yet very good amp (while mid-fi
, i am not sure of an obvious listening change, while measurements where better. It would be interesting to test this on a amp with more previous THD than mine.
Last edited:
- Status
- Not open for further replies.