John Curl's Blowtorch preamplifier part II

[scott wurcer]

Scott, I am afraid that cheating is always possible in case of web based test, depends on level of technology used. Cheating is not the main goal - at least I hope.

If we exclude everything that could matter, we end with completely same files

I'll give them a listen when I can use a good external DAC and my Sennheisers and when it warms up enough to turn off the HVAC.

 

[PMA]

It's not the goal, it's an unfortunate means to achieve a goal for some people. The ease of cheating can be minimized with a good experimental design (I gave some suggestions), but it does make more work for the experimenter.

SY, tell me, namely, what is wrong with these files and what is wrong with foobar ABX comparator. I am not interested in a vague discussion regarding older files which had mistakes. You stopped following then. I ask you for real and defined objections.

https://www.dropbox.com/s/b6q9le2kw5ls5i2/kiwimike.zip

 

[jneutron]

Why can't you tune your setup to Billie Jean like a normal person.

Billie Jean??? Shirley you jest.

PYT. Or, for scott...Money.

Just to throw a wrench in the works let's reconsider the cable issue with a separate Kelvin connection for feedback to the amp.

I recall building a 60 wpc amp back in 74 setup as that. I don't remember what magazine it was in. Just that the front end used constant current diodes instead of discretes. The diodes were 30 bucks apiece at the time, more than the bulk of the semi's in the project.

As I recall, they didn't make provisions for a disconnect scenario..if the fb wires disconnected, something gave. Now I'd put a pair of 100 ohmers between the outs and the fb within the chassis, so even if the fb disconnects, it still works albeit degraded.

More importantly, do not twist the fb with it's comparable wire, the mag coupling will kill it. Go with two independent twisted pairs.

Since we seem to be back to cables again I have a question. Is there any advantage to using a twisted pair plus ground wire if needed in a power cable? Is there really any reason to worry about this before the power supply as the wires in the walls are either parallel wires in a metal conduit or parallel wires in Romex type cable? Just wondering if this helps in anyway before the power supply and any PSRR?

Worry about it from the duplex outlet to the amp. That's where the ground reference ends.

To expand a little on the Bateman issue, there is no way to make a bridge that will give the right answer for complex impedance in the transmission line sense where the real and imaginary parts are orders of magnitude apart due to there being .001 wavelenghts of line. His bridge could no more work at 10kHz than an RF bridge could at 10MHz with 1cm of line. There is simply not enough energy lost in the loss mechanisms to resolve it from the energy stored.

The assumption is that the bridge is looking at 10Khz. It's looking at a high speed composite reflection.

Jn you have a competitor
My coaxial cvr jig for impedance measurements

George

Nice work...Excellent.

One question. The scope ground to ground, are they connected? If so, the scope frame will short the resistor.

Speaking of longitude, there's yachting.

Aha...Harrison..

Skipping back a couple of days.

"Ground Loops: The Rest of the Story"

Bill Whitlock, AES Fellow and Jamie Fox, P.E.

This paper was presented at the AES 129th Convention, 4-7 November 2010, San Francisco, CA, USA

ABSTRACT

The mechanisms that enable so-called ground loops to cause well-known hum, buzz, and other audio system
noise problems are well known. But what causes power-line related currents to flow in signal cables in the first
place? This paper explains how magnetic induction in ordinary premises AC wiring creates the small voltage
differences normally found among system ground connections, even if “isolated” or “technical” grounding is
used. The theoretical basis is explored, experimental data shown, and an actual case history related. Little
has been written about this “elephant in the room” topic in engineering literature and apparently none in the
context of audio or video systems. It is shown that simply twisting L-N pairs in the premises wiring can
profoundly reduce system noise problems.

http://xa.yimg.com/kq/groups/20963848/268252969/name/Whitlock-Fox+-+Ground+Loops+.pdf

Thanks for the link.

jn

 

[Barleywater]

Depends on what you're testing. For example, if I believe that a Teflon dielectric capacitor in a coupling position gives a clearer and more incisive sound than a polycarbonate capacitor on certain source material, then there's no "reference" needed- I just need to be able to distinguish one from the other using only my ears (no peeking). If I can do that, then the choice becomes a matter of taste, assuming that both versions are audible in a bypass test. If one version can be heard when switched in and out and the other not, the "inaudible" version has higher fidelity. If that's my goal, I'm done.

If I can't distinguish one from the other by ear alone, then the difference was not in the sound, it was in the way my "processor" handled the sensory input with the knowledge of which was which and all the preconceptions and unconscious biases associated with my "processor." Any explanations of the difference in sound then need to be psychological, not electronic, at least if I'm not selling something or religiously invested in a belief in those differences.

edit: for the case of boxes of gain not intended to alter the signal, the reference is the input signal.

So in this case, clear and incisive becomes no coupling capacitor, and rest becomes personal taste, or no taste at all.

Input signal is always the reference signal. Residue is loss of fidelity, and becomes characteristic of signal chain.

Same applies to intended filter v actual filter. Once again, the coupling capacitor. Intention may be just blocking DC, but baggage is high pass filter and dielectric behavior.

Yes, experimental design is very important, as in Scott's comments on Pavel's lead in noise.

Testing via internet becomes no holds barred. If intent is listening only, then honesty is involved. Otherwise sharpest tool wins.

I have yet to see test where audible difference did not involve a measurable difference. Somehow, problem seems to be correlation of measurable difference to mountains of subjective terms used in describing sound.

 

[SY]

SY, tell me, namely, what is wrong with these files and what is wrong with foobar ABX comparator. I am not interested in a vague discussion regarding older files which had mistakes. You stopped following then. I ask you for real and defined objections.

https://www.dropbox.com/s/b6q9le2kw5ls5i2/kiwimike.zip

The problem is test structure, not the files themselves. You have to start by defining the specific question you want to ask. See my Linear Audio article for details on that step. IF the question is, "Can anyone hear the difference between A and B," the test setup is quite different than, "Can I hear the difference between A and B?" and will of necessity involve more control to prevent cheating if the evaluation is done out of your observation.

For the former question, the controls are pretty basic.

 

[SY]

I have yet to see test where audible difference did not involve a measurable difference. Somehow, problem seems to be correlation of measurable difference to mountains of subjective terms used in describing sound.

Neither have I, but trying to correlate good data with the random noise emanating from internet cowboys and magazine chimp reviewers claiming all sorts of magic auditory skills and "observations" is an exercise in futility.

 

[a.wayne]

I wonder how many who claim super hearing actually have it, and how many who think there hearing is average actually have golden ears....
Now I know the GEB's will claim they know they have the best instrumentation known to man flapping on either side of there head, but...
A few years ago I was involved with some subjective testing of some comms equipment, all double blind. The interesting thing was that all the participants had to have a full and thorough hearing test, including frequency response. No one was excluded as they were after as wide a range of hearing abilities as possible plus as many victims as possible to get the best statistical data out of what was a subjective set of listening tests. One thing I did learn from one of the audiologists was that those quite a few who do believe they have super human hearing, don't have and those normal people often have hearing that is far better than they expected. What the GEB often have though is belief in there hearing abilities, that precludes the necessity of any further investigation as to what is changing the sound. Gosh some can even determine where the interference is coming from in there system without the tools the poor objectives have to cart around.

It's not only about hearing , it's about perception, most people who hear very well, cant perceive differences, like everything in life , there are always others better..

Marce, the key might be in abilities to concentrate on sound/music and to find differences, rather than frequency limit and absolute sensitivity. Do not forget that hearing is a sensor + processor, engineering approach usually evaluates the sensor only in terms like frequency range and threshold of hearing at defined frequency, and forgets completely about the processor - brains.

+10

Trust your ears, not that lying brain

Is this why you need DBT to tell differences .......?

Maybe so but as I said the noise floor unblinds the test unless the cart dominates the noise. We have to stick with line stages only, less interesting.

Build it Scotty, Blowchip needs to be done and by the Chipman himself, imagine if you get John to shut up on this , incentive enuff ..

 

[SY]

Is this why you need DBT to tell differences .......?

No, precisely the opposite. It's why I (and you and all other humans) need to use DBT to determine if the differences we think we hear are real.

 

[a.wayne]

Blind dates scare me even if the reactors feels the same blindfolded, unless you're Willie Nelson...

 

[Turbon]

No, precisely the opposite. It's why I (and you and all other humans) need to use DBT to determine if the differences we think we hear are real.

One can also give proof of beeing the most trueworthy when it comes to hearing. I would prefer this. But then again it would demand audiologics receipees to be shown in public...

I trust my own and tweak my eq to my ears - it would be nice to have a switch - my ears or specs...

Regards

 

[hhoyt]

Initiative

What figures , I'm not against DBT, go ahead and build it ....

A.Wayne, I question neither your intent nor experience, but if you have an idea of how to prove a point, doesn't it make it your project? I consider it bad manners to rest your point on others doing the work for you.

Unless of course, you are the boss of the company, and in that case you are paying others to do the work.

Just my $0.02,

Howie

Howard Hoyt
CE - WXYC-FM 89.3
UNC Chapel Hill, NC
www.wxyc.org

 

[a.wayne]

You must be new here Howie .... , Blowchip vs Blowtorch, Ali vs Frazier, PPV ? enjoy ...........

 

[hhoyt]

Yup!

You must be new here Howie ....

I know...what was I thinking...

But seriously, my comment to you was more of a general comment of what many people do in internet blogs, they put the burden of proof on other people when it is their point they are trying to establish. There are others in this list more knowledgeable about the science of testing and logic systems (SY?) than me who have a term for this.

And also seriously, I agree with the need for some sort of controlled testing, but I would argue that I've designed and performed more DBX testing than most folks (Richard Clark's Amplifier Challenge, CD vs. Cassette at AES Shows) and I have seen some problems with it:

The transition type is critical, sudden transitions can cause an odd mental confusion, and if the transitions take longer than 5 seconds or so, your mental reference is lost.

Synchronizing sources allows for (conceivably) seamless transitions, but then it is not really an AB test, since 'B' gets a different part of music.

And I'm sure those more knowledgeable in the art know many more .

Maybe the greatest contribution this group of intellects could make would be defining a testing protocol which addressed some of the issues? Or maybe it is already out there?

Howie

Howard Hoyt
CE - WXYC-FM 89.3
UNC Chapel Hill, NC
www.wxyc.org

 

[scott wurcer]

Build it Scotty, Blowchip needs to be done and by the Chipman himself, imagine if you get John to shut up on this , incentive enuff ..

That will never happen, anyway the BT is old news. I've seen at least 2 of the few made on the used market so the owner could chase the next best thing from JC's "competitors". Face it there a dozen or more right now for far more money and every year those go and new ones appear just like on the runway. Take the Vitus at $60,000 gets plenty of Fremer hubris.

 

[a.wayne]

@ Howie,

DBT, DBX testing are just additional tools, there are many other facets involve in determining good audio. I'm not against them, but you can game it , by becoming familiar with certain note structures , not only that, what sounds good listening for a few mins, becomes unlivable after a few hrs..

The best way to go testing, IMO, would be to have those involve listen to different pre-amps in the same looking case, not knowing which is which, then have them give their evaluation, their answer can then be re-evaluated after via DBT, DBX, not before...

If it was my choice .........

 

[scott wurcer]

The transition type is critical, sudden transitions can cause an odd mental confusion

Not just that the click is the key after a few minutes it was easy to score 100%.

The click was not needed for the Carver Digital Time Lens.

 

[john curl]

I would like to say a few things about ABX or related tests. We have tried them, and found them too forgiving of the sorts of things that we like to improve on. Also, at least initially, the designers of these tests were SLOPPY about eliminating type 2 errors, (or not hearing a sound difference) and it was shown by a mathematician that the 95% level was also biased toward NOT hearing differences. I have put my findings in writing 35 years ago. I have spoken at length with Dr's Lipshitz, Vandercoi, Toole, and a number of other professors on the subject.
My best, and least biased reference was Dr. Rod Rees, who wrote a couple of articles on perception bias, starting with 'The Saga of the Golden Ear and the Meter Reader' or Psychological distortions in the human as a measuring device' way back 'The Audio Amateur' as early as October 1979 or 4/1979. Dr. Rees also gave me a personal visit within a year of his first article, and we spent several days going over this topic.
THE LAST PIECE OF ADVICE that he gave to me was: "Never participate in a double-blind test that you did not devise". AND I am sticking with that advice. No intimidation, insults, or goading will turn me. If you like what I say and the products that I design, then fine. If not, you are free to ignore them, and in return, me, as I have nothing else to offer.

 

[a.wayne]

That will never happen, anyway the BT is old news. I've seen at least 2 of the few made on the used market so the owner could chase the next best thing from JC's "competitors". Face it there a dozen or more right now for far more money and every year those go and new ones appear just like on the runway.

With all due respect Scott, i dont deem it fair for you guys to be digging into JC's assertions without some skin in the game, He has proven his position by putting up in real world application.

Is their a chip pre on the market, done the way you would like ....?

 

[john curl]

Scott, of course people sell off their hi fi equipment, sometimes. However, the unit in question that you noted for sale was our very last one, and we took a chance on an 'unreliable' customer, who disappointed us once again. He even BROKE the CTC Blowtorch and I had to fix it for free. Previously, he broke 2 stereo power amps, one of which is still in my lab needing fixing. He IS fickle and he also sold of the Vendetta phono stages that he had. So what? I think the second sale was the same unit.
For me, it is the end of the line, and it now appears to be quite a bargain compared to much of the competition, including Pass Labs and Ayre.

 

[john curl]

OK guys, I want you to all consider buying 'Constellation' audio gear. It will only cost you $100,000 or so to get a useful electronics package up to the speakers. '-)