John Curl's Blowtorch preamplifier part II

Status
Not open for further replies.
He's adding a physical resistance rather than creating a "virtual" one in the source impedance of the driving amp. The efficiency argument is, IMO, not totally honest.

Maybe the efficiency argument is not totally honest, but my mean intent was to show how increasing source resistance affects frequency response of the dynamic speaker near its resonance. The result is same in both cases, either I increase voice coil resistance (for simulation purposes) or add resistance to the source (generator, amplifier).
 
Pavel,

Maybe the efficiency argument is not totally honest, but my mean intent was to show how increasing source resistance affects frequency response of the dynamic speaker near its resonance. The result is same in both cases, either I increase voice coil resistance (for simulation purposes) or add resistance to the source (generator, amplifier).

Why not show something real instead of building straw-men to scare people with "look how bad current drive is"?

Create the same curve twice.

Once use voltage drive to your model.

Then derive the TS Parameters of the driver, set Qt = Qm and use these numbers to create a linkwitz transform to match the original systems Fs and Qt.

Add this into the signal before the current generator and drive your ersatz lautsprecher with current.

What you will find is that frequency response between both cases is essentially identical. If your model is sufficiently abstract but precise you will find no difference at all. You can also X(f) and others and you will find them identical.

Now for the next step.

Please add the necessary non-linearites for Le(i) to your model (which makes a little closer than before to a real speaker) and run distortion curves for both systems at 100W input to the tenth harmonic.

Then add Le(x) and Re(i) to the model and do the same thing again, including frequency response curves, where the voltage drive system will show interesting effects.

Next we can add BL(x) and Km(x) to your model to get it even a bit closer to reality and we can re-run the curves...

After this we can add all the more minor stuff to your model and eventually we may have something that begins to resemble a real speaker somewhat, instead of being an academic construct that bears scant relation to what it is supposed to model.

Ciao T
 
I think that the most important thing is that the 'efficiency' of the loudspeaker is NOT necessarily compromised by current driving the speaker. Why you guys want to go at each other is beyond me. I respect you both, and I have told you both this, but I respect Ed Simon and many others here too.
 
Last edited:
opamps for phono stage

Not really, I don't need to take anyone's secret find of the best sounding op amp. After all, many here think that all IC's sound the same anyway. If you have a technical secret, please keep it to yourself, but please don't ask anything extra out of me, either.

Walt Jung had some ideas about using an opamp for a phono stage using the AD745 designed by no less than........... Scott Wurcer. ;)
See page 6.22 in the link.
http://www.analog.com/library/analogDialogue/archives/39-05/Web_Ch6_final_I.pdf

:cool:
 
Last edited:
You can also put the RIAA in the feedback, but Walt thought that was going too far for an ADI app note.

Anyway Ed, Thorsten was talking about two speakers in essentially an acoustic bridge to null the difference. You mentioned using one speaker as a drive and another as a microphone to measure the absolute frequency response without a microphone. I sent that B&K link because it is a no holds barred math and apparatus description of the process. Spare the maths spoil the project.
 
Last edited:
Why you guys want to go at each other is beyond me.

I do not understand it as well, and I do not feel like that.

I only thought it was necessary to explain how current drive affects frequency response of the dynamic speaker (for those who might not be aware of it). It was purely technical explanation, nothing personal.
 
You can also put the RIAA in the feedback, but Walt thought that was going too far for an ADI app note.

Anyway Ed, Thorsten was talking about two speakers in essentially an acoustic bridge to null the difference. You mentioned using one speaker as a drive and another as a microphone to measure the absolute frequency response without a microphone. I sent that B&K link because it is a no holds barred math and apparatus description of the process. Spare the maths spoil the project.

Scott,

It is pretty simple arithmetic when you look at it. Some place there is the tale of how all the standards folks ship their microphones around for comparison and who has the most traveled and best calibrated one.

The actual first important use of the reciprocity principle was in making machine screws! One of the foundations of all modern equipment is the ability to position things accurately. Things such as a screw cutting lathe require a gear box and feed screw.

As you know even the ancient Greeks could file gears and build calculating machines. What is one of the modern wonders is the precision screw. To make the first screws required placing uniform guide marks on a rod and then using a triangle file to form the threads.

One fellow did this for a few pieces and then used a primitive version of the modern lathe to use one screw to trim up another, back and forth, front to back etc. He kept refining his screws. Virtually all modern screws were made from his standard. Now that is taking reciprocity to the extreme.

Around 1947 is when enough was known about interferometry to make more accurate screws! Of course that didn't really become practical until you had a laser light source in the 60's!
 
The original JC-1 was actually better, subjectively, than the JC-1AC or JC-1DC which were 'improved' versions of the JC-1. I did not know what input loading did at the time. I did an A-B test with a JC-1AC to prove the point, subjectively. It was unfortunate, and it shows how difficult it is to predict subjective audio quality. It measured OK.
 
Scott,

The actual first important use of the reciprocity principle was in making machine screws! One of the foundations of all modern equipment is the ability to position things accurately. Things such as a screw cutting lathe require a gear box and feed screw.

!

We might have a slightly semantic difference in the definition of reciprocity, better to let it go and await the physical results.
 
John,

Why you guys want to go at each other is beyond me. I respect you both, and I have told you both this, but I respect Ed Simon and many others here too.

What is going on is simple.Ther are some who will champion unorthoox ideas that have merit while others feel threatened by unorthodox ideas and feel the need to opose them strongly, to the point where they publicise falacious proofs of just how bad an idea it is.

Then they receive of course strong opposition as they basically posted balderdash, which they often take personal.

I have no beef with Pavel, but his reaction to current drive would be comical ifhewas not so serious that is the greatest evil in audio...

Ciao T
 
As far as i understand reciprocity needs 3 reversible transducers and devision of the last two transfer functions to extract the error component. A dynamic loudspeaker is not 100% reversible. It works for capacitive drivers-receivers though. An example of a non reversible speaker is a bending wave transducer. Over the coincident frequency sound that is send to the transducer is transferred into heat. It is really not that simple but i pass on that. The next 100 posts about nothing. A last question : who has hands on experience with reciprocity ?
 
As far as i understand reciprocity needs 3 reversible transducers and devision of the last two transfer functions to extract the error component. A dynamic loudspeaker is not 100% reversible. It works for capacitive drivers-receivers though. An example of a non reversible speaker is a bending wave transducer. Over the coincident frequency sound that is send to the transducer is transferred into heat. It is really not that simple but i pass on that. The next 100 posts about nothing. A last question : who has hands on experience with reciprocity ?

That is correct, I question the reversibility of loudspeakers to the degree of getting any useful results, and I don't see any use for doing it with two.
 
As far as i understand reciprocity needs 3 reversible transducers and devision of the last two transfer functions to extract the error component. A dynamic loudspeaker is not 100% reversible. It works for capacitive drivers-receivers though. An example of a non reversible speaker is a bending wave transducer. Over the coincident frequency sound that is send to the transducer is transferred into heat. It is really not that simple but i pass on that. The next 100 posts about nothing. A last question : who has hands on experience with reciprocity ?

What the hell are you talking about! :D
 
Status
Not open for further replies.