John Curl's Blowtorch preamplifier

Status
Not open for further replies.
Jump to Latest
#19,244
john curl said:
Demian, you gave us quite a handful, and not enough REAL input to make intelligent discussion. We need to know about every part. Sorry, but I certainly can't help further.
Click to expand...

I'm not entirely clear on what you want to know. You have the schematic and a detailed photo of the parts and layout. I'll be happy to fill in more info. as best I can.

Listening to the output of the differential output is best described as uninvolving. Since changing the components would be a little risky (surface mount etc. and not mine) I connected to the output of the current to voltage converter. It was a little better.
 
#19,252
lineup said:


I think I have to check out that Charles Hansen investigation.
Subjective matters are a bit more difficult to measure than resistance or some different sorts of distortions, for example.

Sound impressions or what people can visualize in a painting, colors
are such things.

We often have rely on statistical analysis and so count result with probability factor.
100% sure = 1.0,
because from stastistical probabilty the possibility of just chance/guessing is zero.


Another member, 'Gedlee' have done one reserach, much to prove how good his spekers are.
He also suggest a special method a scale for how we experience distortion.
Gedlee is a nickname, short for Earl & Linda, who performed some investigations.

Member profile:
http://www.diyaudio.com/forums/member.php?s=&action=getinfo&userid=31488
Sound Quality
Distortion Perception
http://www.gedlee.com/distortion_perception.htm

Here are some results and plots / diagrams:
http://www.gedlee.com/results.htm

You can freely download his findings and description on the method he used, together with another researcher.

/regars lineup
Click to expand...

Thanks for the links, interesting. Though, creating a new metric for distortion that corresponds better to how we hear, is kind of a different subject than blind testing.

scott wurcer said:


One problem I have with these tests is the editorializing, phrases like "everyone in the room" etc. show to me a bias and lack of objectivity equal to the Audio Critic mindset.
Click to expand...

Well, I know people tend to put their research in the most postive light. Especially when presenting it to the DIY audio attack crowd. I'd like to see some well documented extended listening blind testing with an objective observer present. I can dream. But some things Charles Hansen said about the importance of experience and familiarity, definitely fit with my experience and what I know about how we hear. And I agree with his testing method for the most part.

I mean when you've dealt with the kind of sound senstivity I have for 10 years you learn some things. You have to know how to make progress and if you don't you go backwards. The ear is extremely senstive to an increase in sounds we have aversion to. Every sound is different. As to what is and isn't audible? I think blind testing can work if it takes into account that sounds are learned.
 
#19,255
open loop transfer function

john curl said:
[snip]
However, the open loop LINEAR TRANSFER FUNCTION is now often ignored, because feedback HIDES the open loop characteristic. Is this OK? What do your ears tell you? Ever listen for yourself?
Click to expand...

Hi John,

I don't think the open loop transfer function is simply ignored by most of us, neither by those who apply a lot of NFB. This is because you can't 'hide' all non-linearity by increasing NFB ad infinitum without violating stability criteria. So, at a certain point, it's much easier to linearize the open loop transfer function instead of all the hocus pocus to stabilize an ultra high gain NFB loop.
Nevertheless, I don't see any disadvantage by doing both, i.e. linearizing and applying moderate NFB.

BTW, if NFB really 'hides' the shortcomings of a nonlinear open loop x-fer function, what's the problem anyhow?
(In my vocabulary, 'hide' means not only you can't measure it, but also you can't hear it)

Cheers,
Edmond.
 
#19,257
What amazes me, how people ask the same questions, over and over. Most IC designs, for example, have fairly lousy open loop transfer functions.
What Charles Hansen and I concentrate on is the open loop transfer function, and how to improve it. This is most important to us, because we have found that to rely on negative feedback to hide it or convert it to another form of distortion isn't normally that effective, when we are making top designs, at the level where we compete with each other. It is like F1 racing vs stock cars. There is a difference, although most people are not aware of it. One important approach is TANH that Scott Wurcer put forth, through the work of Barrie Gilbert. Does anyone have ANY idea of what we are discussing? What are the upsides, and what appear to be the downsides. That is but one. Now, PMA, as well as many other first class designers, tries as hard as anyone, to generate a first class transfer function. However, many more try to just add more feedback. This is why it needs discussing.
 
Status
Not open for further replies.
Top Bottom