damping factor is the ability for the amp to not "flywheel" after a particular component of the music ceases, as I understand it...not sure if that is the accepted definition though...in my tube amp I use no nfb except local. Doesn't that help, along with the lack of output trans...? optiimationman...not acquainted with the amp or your setup, sounds interesting though. Your amp has a damping control.?..vintage stuff, eh...what kind of amp is it.?..I can't imagine any (simple) control can nuetralize cable or transformer losses...that seems like salesmanship..wasn't the damping control the "flab control" for those boomy early corner cabinets ? edd
This may not apply to your circlatron if it has substantial feedback, but I built a more conventional Pushpull x-former coupled output tube amp with 0 - 12dB of negatory feedback, and my output impedance varied between 1 ohm and 8 ohms.
One thing this means is that the amp won't dampen the woofer resonance that usually exists around the 60HZ region very well. Another maybe more important thing this means is that this higher source Z will throw off the accuracy of a passive crossover immensely. The reactive components of the crossover will see a much bigger Z, and will do their tricks at much different frequencies.
One thing this means is that the amp won't dampen the woofer resonance that usually exists around the 60HZ region very well. Another maybe more important thing this means is that this higher source Z will throw off the accuracy of a passive crossover immensely. The reactive components of the crossover will see a much bigger Z, and will do their tricks at much different frequencies.
^^ you might want to read this, you are correct that the output impedance of the amp can change how the crossover behaves. The speaker designer has to take the impedance of the amp into account, depending on how he does the design.
http://www.atma-sphere.com/Resources/Paradigms_in_Amplifier_Design.php
Also, its unlikely that any speaker needs more than a damping factor of 20:1. This was shown decades ago by the head engineer of Electro-Voice.
http://www.atma-sphere.com/Resources/Paradigms_in_Amplifier_Design.php
Also, its unlikely that any speaker needs more than a damping factor of 20:1. This was shown decades ago by the head engineer of Electro-Voice.
Midmoe, the amps are mono block vintage design and construction(early 1960's). It is a fairly complicated(for a tube amp) design with a non trivial positive feedback control which defeats a portion of the negative feedback to make up for internal amp losses as the current output goes up(read I2R losses). This in fact reduces the amount of negative feedback. I guess this would give and excellent damping factor without using excessive amounts of negative feedback.
Bob, " higher source Z will throw off the accuracy of a passive crossover immensely" is not a factor as I am zeroing out the source Z or at least the real(resistive) portion of it.
Maybe the damping factor effect is a little overblown with better modern mechanically dampened woofer designs.
Bob, " higher source Z will throw off the accuracy of a passive crossover immensely" is not a factor as I am zeroing out the source Z or at least the real(resistive) portion of it.
Maybe the damping factor effect is a little overblown with better modern mechanically dampened woofer designs.
That's a very rare amplifier, how does it sound? Anyway, I think DF96 gave a good suggestion in post #35 - to get the best damping factor, null the amp's output impedance at the frequency where the speaker has its lowest impedance, which usually occurs in the mid-frequencies, since DF = Zload/Zsource.
Actually the link is right but the underlying html is malformed.
I don't know why, but it could be something to so with the DIY site, try it now:
Atma-Sphere music systems, inc.
Rather OT, but as the term damping factor showed up again lately, perhaps a revisit to what that is supposed to mean - and doesn't. More illustrative explanations can be googled under 'damping factor' (e.g. Rod Elliott's 'Westhost' site), but briefly:
It is common knowledge that the term came into being as an effort to define what degree of control an amplifier can have on continued loudspeaker cone 'vibration' after cessation of signal. (Being mechanical and suspended by some form of elastic means, it will exhibit decreasing vibration after e.g. having been tapped by a finger.) The electromagnetic principle comes into play here: If one can short-circuit the voice coil immediately after a signal, back-emf current generated in the coil will stop unwanted vibrations in their tracks - well almost.
Thus far old hat. The problem now relating degree of damping (so-called 'damping factor') directly to the shorting effect of a low amplifier output impedance, is that the latter is not the only 'brake-current' limiting factor in the basic circuit. In fact it is the least significant, making classic damping factor figures useless as an indication of anything unless they are as poor as <0 to some 3 - 5.
Point: The total (current-limiting) resistance in the simple equivalent amplifier-loudspeaker electrical circuit includes cable resistance, amplifier output resistance, and voice coil resistance; the latter does not suddenly become zero! In fact the d.c. resistance of a typical 8 ohm voice coil is of the order of some 5,5 ohms in practice. Thus a typical amplifier quoting a d.f. of 20 will have an output resistance (impedance) of 0,4 ohm. Thus the real 'braking factor' (to coin a new term) will be 8/(0,4 + 5,5) = 1,36, not 20. For an amplifier boasting a d.f. of 100 (output impedance of 0,08 ohm), such a factor will 'improve' to 8/5.58 = 1.433, not 100. (I have not bothered to include series cable resistance, which would normally make even less difference.)
This rather lengthy explanation (apology to those in the know) will indicate why d.f.s of much 'better' than say 10 have little if any effect on matters (as borne out in practice) and makes arguments regarding such rather academic.
(As mentioned before this excludes active measures such as motional feedback systems, where the amplifier impedance can actually be made negative, etc. - out-of-context here.)
It is common knowledge that the term came into being as an effort to define what degree of control an amplifier can have on continued loudspeaker cone 'vibration' after cessation of signal. (Being mechanical and suspended by some form of elastic means, it will exhibit decreasing vibration after e.g. having been tapped by a finger.) The electromagnetic principle comes into play here: If one can short-circuit the voice coil immediately after a signal, back-emf current generated in the coil will stop unwanted vibrations in their tracks - well almost.
Thus far old hat. The problem now relating degree of damping (so-called 'damping factor') directly to the shorting effect of a low amplifier output impedance, is that the latter is not the only 'brake-current' limiting factor in the basic circuit. In fact it is the least significant, making classic damping factor figures useless as an indication of anything unless they are as poor as <0 to some 3 - 5.
Point: The total (current-limiting) resistance in the simple equivalent amplifier-loudspeaker electrical circuit includes cable resistance, amplifier output resistance, and voice coil resistance; the latter does not suddenly become zero! In fact the d.c. resistance of a typical 8 ohm voice coil is of the order of some 5,5 ohms in practice. Thus a typical amplifier quoting a d.f. of 20 will have an output resistance (impedance) of 0,4 ohm. Thus the real 'braking factor' (to coin a new term) will be 8/(0,4 + 5,5) = 1,36, not 20. For an amplifier boasting a d.f. of 100 (output impedance of 0,08 ohm), such a factor will 'improve' to 8/5.58 = 1.433, not 100. (I have not bothered to include series cable resistance, which would normally make even less difference.)
This rather lengthy explanation (apology to those in the know) will indicate why d.f.s of much 'better' than say 10 have little if any effect on matters (as borne out in practice) and makes arguments regarding such rather academic.
(As mentioned before this excludes active measures such as motional feedback systems, where the amplifier impedance can actually be made negative, etc. - out-of-context here.)
Last edited:
I read the link and agree that using your hearing to guide you to what sound you want is the whole idea. However, I don't understand the following statement,"It is very easy to tell how an amplifier will sound using measurements based on the Power Paradigm as the measurements are made with regards to understanding and working with the rules of human hearing" as I don't know what the Power Paradigm is.
As for the "going negative" (which is actually adding positive feedback) to correct output impedance, that was the purpose of the adjustment on these amps. They were design/used as instrumentation amps and the process was to nullify the effects of the driving system, which includes the cable resistance. Since the speaker coil resistance and or series inductors resistance are also in series with these components, why would they not also be dial out? Crazy......
As for the "going negative" (which is actually adding positive feedback) to correct output impedance, that was the purpose of the adjustment on these amps. They were design/used as instrumentation amps and the process was to nullify the effects of the driving system, which includes the cable resistance. Since the speaker coil resistance and or series inductors resistance are also in series with these components, why would they not also be dial out? Crazy......
The problem is that feedback is also always arriving late to do the job it was intended to do because of propagation delay in the amplifier circuit.
The article attempted to introduce the Power Paradigm, which is the idea that the amplifier should make constant or nearly constant power into a load regardless of impedance, rather than constant voltage.
The article attempted to introduce the Power Paradigm, which is the idea that the amplifier should make constant or nearly constant power into a load regardless of impedance, rather than constant voltage.
Inherently Flawed Paradigm?
http://www.diyaudio.com/forums/soli...onstant-power-amplifiers-similar-schemes.html
^^ its apparent that you have not studied the history of audio. This is probably because you are younger than me and can't be faulted for that.
So here is an exercise: Google the term 'Fisher A-55 amplifier' and the first hit will be a YouTube image of the damping control on the amp. It is labeled 'constant voltage' at one extreme, 'constant power' at 12 on the dial and 'constant current' at the other extreme. 'Daft' is not a correct expression.
In the old days constant power was really the only game in town, we all know that tubes don't double power as the load is cut in half, and that many tube amps were (and still are) built without negative feedback.
Regarding feedback, I am correct and this is easy to see on rather mundane test equipment. Feedback will only work right if you are using repetitive signals like sine waves. Even in large amounts on amplifiers with low propagation delays, feedback contributes to odd ordered harmonic distortion, a fact which has been known for the last 50 years, I refer you to the writings of Norman Crowhurst, a well-known authority.
No- this seems to be something different. No amplifier is capable of truly constant power, just as no amplifier is capable of truly constant voltage. The linearity of a good number of Power Paradigm amplifiers can actually be quite high- does IM of 0.05% at full power sound OK?
So here is an exercise: Google the term 'Fisher A-55 amplifier' and the first hit will be a YouTube image of the damping control on the amp. It is labeled 'constant voltage' at one extreme, 'constant power' at 12 on the dial and 'constant current' at the other extreme. 'Daft' is not a correct expression.
In the old days constant power was really the only game in town, we all know that tubes don't double power as the load is cut in half, and that many tube amps were (and still are) built without negative feedback.
Regarding feedback, I am correct and this is easy to see on rather mundane test equipment. Feedback will only work right if you are using repetitive signals like sine waves. Even in large amounts on amplifiers with low propagation delays, feedback contributes to odd ordered harmonic distortion, a fact which has been known for the last 50 years, I refer you to the writings of Norman Crowhurst, a well-known authority.
No- this seems to be something different. No amplifier is capable of truly constant power, just as no amplifier is capable of truly constant voltage. The linearity of a good number of Power Paradigm amplifiers can actually be quite high- does IM of 0.05% at full power sound OK?
This will come as a surprise to people who design fly-by-wire systems. Likewise industrial controls like PID.
The Fisher was not alone in using that technique. You might want to read an article by the head engineer of Electro Voice:
Missing Link in Speaker Operation
I see you edited out Norman Crowhurst's name. You might want to read his materials rather than just trolling this thread.
Sy, those applications are pretty different but if examine them you will find that they have ways of dealing with it. The Technics SP-10 is one of the better examples of servo theory applied to audio without actually being in the signal path, more of a motion-control exercise.
Missing Link in Speaker Operation
I see you edited out Norman Crowhurst's name. You might want to read his materials rather than just trolling this thread.
Sy, those applications are pretty different but if examine them you will find that they have ways of dealing with it. The Technics SP-10 is one of the better examples of servo theory applied to audio without actually being in the signal path, more of a motion-control exercise.
Last edited:
I've read everything Crowhurst has written. Most of it is correct and educational. Much has been passed by with a lot of excellent follow on work- the speaker stuff, for example, was written in the days long before Thiele/Small and Klippel, not to mention Toole and Olive. Such is progress! 
Try a simple experiment: input a pulse to an amplifier and measure the output waveform. Now apply 10dB of feedback around the amp and measure the pulse again. Despite being non-repetitive, it is reduced by 10dB. You can do the same experiment using MLS, another non-repetitive signal. Do the same for measuring source impedance, before and after feedback.
Try a simple experiment: input a pulse to an amplifier and measure the output waveform. Now apply 10dB of feedback around the amp and measure the pulse again. Despite being non-repetitive, it is reduced by 10dB. You can do the same experiment using MLS, another non-repetitive signal. Do the same for measuring source impedance, before and after feedback.
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.