Graham Maynard said:
Graham, I should never challenge you with something you cannot do.
If you can make simulations and read / write posts, you can as well take a digitezed music sample, input to a simulator in both versions with and without outpout coil, even use a model speaker load for that, and then substract the result and store in a digitized file. This you can play and listen.
I wish I had time to devote to this and many more experiments but it is not so.
Rodolfo
No Rodlfo I can't do that, and it is presumptious of you to suggest I can.
Cheers ....... Graham.
Cheers ....... Graham.
Graham Maynard said:
Sorry then Graham, but please bear in mind it is absolutely out of my character to challenge someone with something he cannot do. Apologies if I misinterpreted in this regards.
Rodolfo
Well Graham, we have just experienced 'The Night of the Living Coil Users' 
I think that is it pointless to go on. They will throw erasers at us next!
I think that is it pointless to go on. They will throw erasers at us next!
Workhorse said:Ok lets Sign a "TREATY OF NO OUTPUT COILS" in amplifiers from Present to Future...... 😀 😀 😀 😀 😀 😉
Who would be the first one to signature....
I would immediately sign if I can get output devices, PCB tracks, wires
and capacitors with zero (0.0nH) inductance. 😀 😀 😀
Cheers,
ingrast said:
Yes. This is a very easy and quick simulation one can run. My experiments showed max. error signal of about 100 mVpeak (during the time the output signal amplitude was approximately 20 Vpeak). The error signal amplitudes were highest with signal content of sharp transients, i.e. the very beginning of drum hits. The error signal also had very distinctive components originating from crossover distortion. Could I hear any of these differences while A/B:ing two signal sources? No way. I also tried the same experiment without capacitor-resistor Zobel in the output: It produced no noticeable effect. ...And with purely resistive load: This lead the error signal peak amplitude to grow substantially higher, i.e. max peak was now 360 mV, typical 40 - 70 mV peaks turned to 100 - 220 mV peaks etc.
In my test, the inductor filter consisted of 2uH in parallel with 10R. The load was an electrical model of a single driver instead of a complex crossover. (Re=3.88, Le=0.34m, Cmes=316.0011µ, Lces=10.0949m and Res=23.6803).
john curl said:
As a matter of fact I do not use output coils either, which does not mean I will endorse its deletion on the basis of not proved audibility effects.
Rodolfo
Graham Maynard said:
Bonsai stated to you that the inductance of the speaker cable, xover etc would swamp any 2uH output coil effect, to which you answered:
"Inductance within a closed NFB loop is not the same as having inductance between an output node and the LS system.
Inductance withing the loop can affect global loop performance and capabilities in the presence of reactive loads. It can trigger phantom non-linearity events which might arise at some particular instant of a music reproduction performance with a particular loudspeaker, but not others."
Sufficient?
Jan Didden
john curl said:Well Graham, we have just experienced 'The Night of the Living Coil Users'
I think that is it pointless to go on. They will throw erasers at us next!
Hi John,
This has actually been a very worthwhile thread, and I want to thank you for posting it. It has certainly made me think more deeply about a couple of issues surrounding the topic.
For example, even apart from the question about audibility of it, is the R-L network REALLY needed under reasonable conditions? And what ARE those reasonable conditions? Even if we think we will have to drive a 2 uF electrostatic loudspeaker, does a few feet of ANY speaker cable offer enough isolation to be safe? Maybe. Or are there speaker cables out there with so little leakage inductance and cable resistance that they offer essentially no inherent isolation between the amplifier and a 2 uF capacitive load 6 feet away?
And for similar reasons, is 2 uF of extremely low ESR capacitance right at the output terminals of the amplifier even remotely representative of the worst imaginable real-world load?
Regarding audibility, I'd like to know your opinion on a couple of the issues surrounding the output coil. Let's assume for the moment that we agree that the presence of a certain 2 uH output coil in a certain hypothetical amplifier is indeed audible. If it is audible, why do you think it is audible? Is it audible because of the mere presence of the 2 uH of inductance (even if the inductance were created by an ideal inductor)? Or is it audible because of nonlinear distortions created by the inductor that likely would be measurable (as, for example, by inductor nonlinearity due to proximity to a metal item)?
In other words, is the badness created by the inductance per se, or by the imperfections of the inductor device?
Further to the point, we know that virtually all amplifier outputs are a bit inductive even without the coil. This can be the result of a falling value of NFB at high frequencies or even can occur to some extent in an amplifier without global NFB as a result of the falling beta at high frequencies of an emitter follower output stage. Is such electronically-created inductance OK (not audible), or is it just as BAD as a similar amount of inductance created by an ideal coil?
As an example of inherent "electronic" inductance, if I look at the frequency response of the JC-1 in the Stereophile review for 8 ohm load versus 2-ohm load, it appears that the JC1 may be exhibiting an output inductance as high as 2 uH, even though it does not emply a coil. Do you agree with this estimate of the inherent output inductance of the JC-1?
Thanks,
Bob
Hi John Curl
There's been so much ringing on this topic that your answer got buried in the damped oscillations several posts ago. As Bob asked the questions more succinctly than I, I'll wait to see how you respond on that before asking any more.
On second thoughts, oh well hear goes... how do you hear ringing above 20 kHz unless it upsets an amplifier ...
or is this what you meant by "more complex than that.."
cheers
John
There's been so much ringing on this topic that your answer got buried in the damped oscillations several posts ago. As Bob asked the questions more succinctly than I, I'll wait to see how you respond on that before asking any more.
On second thoughts, oh well hear goes... how do you hear ringing above 20 kHz unless it upsets an amplifier ...
or is this what you meant by "more complex than that.."
cheers
John
john curl said:
John,
You do not know me, and I have only "met" you in this thread, which I found late and am still battling to get through before venturing a post.
So no instant judgement about you; I have heard many good things about you, but have not yet had the pleasure of experiencing any of your designs ...
But with all respect, the above is one sweeping statement that leaves you totally unprotected to misconceptions! There could be a myriad reasons why designs do "not need" series inductors, like too limited open loop bandwidth, too little NFB, etc....... that would also qualify as "flawed design skills", would it not ??
Certainly not suggesting that such is the case with your designs, but just to make a point. Flawed design is one thing, but in contrast "design skills" refer to a person, and I really doubt that anyone has the right to be presumptious in that regard?? (It is in fact these kind of occasional thorns in the way through this thread that is still retarding my progress in trying to figure out what exactly is the gripe here so that I can perhaps learn. Is this a prestige thing: "Look Ma! No coils!" or is there a scientific argument somewhere of real merit in audio? As said, I am still working forward to the present, but your statement just stuck out a bit.)
Sincere regards - from someone who is also a professional designer (... and might even have been, just possibly, at it for longer than you have - just for the record).
Just for the record, just try to find a contemporary power amp from any quality manufacturer that still uses a 2uH or more coil at the output of their power amps. I am curious to know. No mid-fi please.
john curl said:
Hi John,
Looking at the Bryston schematics on their site, it looks like they use 2 uH coils. Not sure if you'd consider these "mid-fi" though.
Not saying this is good or bad, but you were looking for an example...
Well, they've got lots of them 🙂, but the one I pulled out more or less at random was the 4B SST (schematic here). Which one were you looking at the data of?
Looks like they have schematics for just about everything they make at their site (bryston.ca).
They are using the slower output devices (21193 and 21194) so maybe that is why.
Looks like they have schematics for just about everything they make at their site (bryston.ca).
They are using the slower output devices (21193 and 21194) so maybe that is why.
Graham Maynard said:
Excuse me? Are you serious? I AM posturing?? And why the fudge should any onus be on me to painfully state opinions of mine which others don't have the right to pretentiously presume?
- Status
- Not open for further replies.