 
|
Audio Asylum Thread Printer Get a view of an entire thread on one page |
For Sale Ads |
|
|
Audio Asylum Thread Printer Get a view of an entire thread on one page |
For Sale Ads |
202.49.80.100
In Reply to: RE: Too many variables ... posted by Tre' on August 18, 2016 at 17:25:19
or the essence of what you say as being typically the main attributes of Class A. But taking a line here or there out of a handbook is out of context in that they are not the lines that define Class A, they are merely attributes typical of the class and its configuration as applied to SE power amps in most of your examples.
But Class A is a definition applied to all amplifiers and (for example) there is no way that it's necessary to set an operating point anywhere near "one half of the cutoff voltage in a low signal application to achieve performance worthy of Class A even by your high standards:)
The best way I can get my point across is to ask you this question. If I design an amplifier that under normal operating conditions remains in conduction with no grid current, but I do not set the operating point anywhere near the "midpoint" or "the most linear part of the curve" yet achieve the main goal (even using your argument) of low distortion, how would YOU classify it?
Dave also made a valid point about clipping. Does this means that any amplifier capable of overload cannot be classified as Class A?
Naz
Follow Ups:
"The best way I can get my point across is to ask you this question. If I design an amplifier that under normal operating conditions remains in conduction with no grid current, but I do not set the operating point anywhere near the "midpoint" or "the most linear part of the curve" yet achieve the main goal (even using your argument) of low distortion, how would YOU classify it?"
I think I would just ask you why you didn't set the operating point at the midpoint.
I answered Dave.
A Class A amplifier driven beyond the linear part of the curve is a Class A amplifier that is being abused. Why would anyone want to do that?
I think this horse is dead.
Tre'
Have Fun and Enjoy the Music
"Still Working the Problem"
What Class is it in, it must be in something?
I can give you many reasons why I may not bias for the midpoint in a low signal amp and I'll bet you may too from what I've read.
I also don't appear to have been clear enough in making my points re plate curves and their "apparent" most linear part or CCS loading, which renders the point almost superfluous anyway ... sigh!
Anyway, it seems that we can't all agree on this and I respect everyones right to hold their own beliefs.
Naz
Don't take this personal but I would call it a poor attempt at Class A.
"I also don't appear to have been clear enough in making my points re plate curves and their "apparent" most linear part [see graph above, it's pretty clear to me what part is linear and if you move too far from the center it's not so linear anymore] or CCS loading, which renders the point almost superfluous anyway ... sigh!"
A horizontal load line won't help this tube much. Draw a horizontal load line and then derive a dynamic curve. It's going to be pretty curved with no straight part. (maybe a little between -14 and -18) The spacing starts large on the left and gets skinnier as you go to the right.
"Anyway, it seems that we can't all agree on this [that's clear] and I respect everyones right to hold their own beliefs."
I see some of those beliefs as flying in the face of what is explained in the technical books.
Oh well.
Take care. Nothing personal......
Tre'
Have Fun and Enjoy the Music
"Still Working the Problem"
> > Don't take this personal but I would call it a poor attempt at Class A < <
Class A none the less?
> > A horizontal load line won't help this tube much < <
Well, actually it will but that's not the point. Are you saying that this tube can't ever be be run in Class A?
Have to be very careful with curves as they can look much worse depending on the scale. Believe it or not the curves look much the same if we expand the region 300-400V up to a max of only 5mA.
Cheers!
Yes, still Class A but not one that follows any of the examples in the books.Is there something wrong with my eyes?
For a 8 volt peak grid swing I get a plate swing of 142.5 one way and a plate swing of 122.5 the other and then there's the problem of the plate resistance increasing as we go to the right.
Let's try this instead.
I get 120 and 125.5 and the plate resistance hasn't changed much at all.But in the end you win. Your example is Class A but bad Class A.
Why in the deuce would we want to spend time talking about a bad Class A?
What would be the point in that?
I pledge from now on I will try to preface my talk of Class A with the word "good" if you will try to remember to preface your talk of what I call bad Class A with the word "bad". OK? :-) (with the understanding the "bad" just means that it doesn't follow the examples in the books and "good" just means that it does)
Or how about when you see me talking about Class A just remember that when Tre' talks about Class A he's always meaning Class A done the way the books show it (silly Tre').
So now we've come full circle. The book Micheal posted the link to says, on page 68 that a Class A amplifier is "An amplifier operated as shown in fig. 306 in which the output waveshape is a faithful reproduction of the input waveshape, is known as a Class A amplifier."
Fig 306 is for all intents and purposes the same as the figure below.
You'll just have to excuse me, I choose to stand on that.
Tre'
Have Fun and Enjoy the Music
"Still Working the Problem"
Edits: 08/19/16
I accept that linearity varies from one tube to another. My only points are that linearity will always be better by varying degrees with a more horizontal load line and that to the eye, scale can be deceiving when viewing plate curves.
On the latter, here's a link to a graph of plate curve for a 6SN7 at low voltage and low plate current taken from Stephie's Site. It's not what I was looking for but it will tell the story. This region on a larger scale would normally look non linear. You can Sim curves to explore the effect yourself.
It's clear that good linearity can be gained at many different operating points where power is not a prerequisite.
Naz
"It's clear that good linearity can be gained at many different operating points where power is not a prerequisite."
I think I've left myself misunderstood.
While I think choosing the most linear operating condition is in keeping with the intent of Class A that is a different thing than choosing not to bias in the middle of whatever dynamic curve your chosen load impedance provides.
The sim tool linked below can show you what I mean.
The outside left arrow will change the load impedance and idle current.
The inside left arrow will change the load impedance only and the top white arrow will change the bias point (quiescent plate voltage and current) along that load line.
Note, keep the yellow arrow (grid swing) somewhat minimized to start with or it will limit what you can do with the other controls.
Watch the HD in the readout below as you make your adjustments.
Tre'
Have Fun and Enjoy the Music
"Still Working the Problem"
that there can be many operating points that will provide Class A operation unless max power or swing is a prerequisite.
Whilst I agree there will always be an operating area of lowest distortion, equally you could pick an area that would normally provide slightly higher distortion but achieve better overall results by making the load line more horizontal as your Sim will show. Great tool BTW!
So for clarity I'm saying that the operating point cannot be singled out in the definition of Class A because several factors dynamically interact, meaning that similar distortion characteristics can be obtained at various operating points, within the limits of swing and power we are designing for of course.
To include subjective statements like lowest distortion into the definition would be absurd. Eg, in an SE amp one of the biggest factors determining distortion is the load line so how would one determine whether an SE amp met the criteria of Class A by your expanded description?
Anyway, one thing we both agree on is that best practice should always be used in order to achieve the very best result. So I'll leave it at that.
Thanks for a civil discussion that I'm sure will be of some benefit to some others, particularly newbies.
Cheers
Post a Followup:
| FAQ |
Post a Message! |
Forgot Password? |
|
||||||||||||||
|
||||||||||||||
This post is made possible by the generous support of people like you and our sponsors:
Top of Page ]
[ Contact Us ]
[ Support/Wish List ]
[ Copyright © 2024, Audio Asylum, all rights reserved ]
