Found this interesting post while researching the issue at hand:
http://www.diyaudio.com/forums/showthread.php?postid=676488#post676488
http://www.diyaudio.com/forums/showthread.php?postid=676488#post676488
jlsem said:
Don't know what you find particularly interesting about it. When I said "sounds best," I didn't mean it to be synonymous with improved objective performance.
se
Do you feel that distortion quantity should be subjectively or objectively measured? I mean, it's scientific to put a number to it but my ears are not a distortion analyser.
planet10 said:
Mmmm. But if you look at these curves, which show permeability down to 10 Gauss (which is quite small and is typically the point at which initial permeability is measured) for varying levels of DC magnetization, permeability decreases. So I'm not seeing how DC sidesteps the problem of low initial permeability.
se
lndm said:
Right. So when you're talking about distortion in an objective sense, you would measure it objectiely. When you're simply talking about what sounds good to you (or bad as the case may be), then you're in the subjective realm and unless you've some desire to try and determine whether it sounds good (or bad) to you due to some objective performance parameter, that should be left in the subjective realm.
se
This part I found interesting:
...that pure 3rd can be amoeliorated with just a wee bit of DC current in the primary or secondary.
This text on magnetic amplifiers shows how the ac input wave becomes distorted as the DC control voltage moves the operating point up and down the magnetization curve. A simple explanation with illustrations begins on page 28. What it doesn't show is an operating point with no DC which would be at the intersection of the B and H axes.
http://www.pmillett.com/Books/mag_amp.pdf
John
http://www.pmillett.com/Books/mag_amp.pdf
John
hysteresis distortion
from what i see, the error in interpreting all these hysteresis graphs to support low level improvements when dc bias is in the primary (as in se amps) is the initial magnetic polarization curve (the one at the B-H origin) is being used to justify operation after the ouside hysteresis loop is established due to an ac mmf being passed.
apples don't equal oranges... as I see it, once the core is magnetized/demagnetized by any ac signal. it will exhibit hysteresis and follow some version of that loop and will not respond in a positive way to lower operating distortion by applying a dc bias; the latest graph shows in fact that dc bias moves the core closer to nonlinear saturation on the half cycle that corresponds to the polarity applied.
But what do I know...I guess this is all out of my "domain" of knowledge 😎
John L.
from what i see, the error in interpreting all these hysteresis graphs to support low level improvements when dc bias is in the primary (as in se amps) is the initial magnetic polarization curve (the one at the B-H origin) is being used to justify operation after the ouside hysteresis loop is established due to an ac mmf being passed.
apples don't equal oranges... as I see it, once the core is magnetized/demagnetized by any ac signal. it will exhibit hysteresis and follow some version of that loop and will not respond in a positive way to lower operating distortion by applying a dc bias; the latest graph shows in fact that dc bias moves the core closer to nonlinear saturation on the half cycle that corresponds to the polarity applied.
But what do I know...I guess this is all out of my "domain" of knowledge 😎
John L.
Re: hysteresis distortion
That's rather the way I'm seeing it as well. So for now, I'll just wait until someone can actually show an actual transformer behaving more linearly at low levels with a DC bias on it than without.
se
auplater said:
That's rather the way I'm seeing it as well. So for now, I'll just wait until someone can actually show an actual transformer behaving more linearly at low levels with a DC bias on it than without.
se
http://www.diyaudio.com/forums/showthread.php?postid=1137000#post1137000
You asked for a reference so I thought your word was as good as any.
John
You asked for a reference so I thought your word was as good as any.
John
Please read my lips: The magnetic field never collapses in an SET OPT. There is no "outside B-H loop". The only reason there are charts that show a DC hysteresis loop is because the guy testing the sample switched the DC off. A saturable reactor with a DC control voltage is as close a textbook example as you will find to an SET OPT.
Can't you see that at the origin of the B-H axes the magnetization curve is nonlinear and the application of DC bias shifts the input to the middle (linear) section of the curve?
John
jlsem said:
So what mechanism, exactly, are you proposing accounts for the modulation of the non-collapsing dc field from the applied electrical ac signal?? Perhaps the domains are wiggling around/expanding/condensing, moving away from the easy direction as they must in order to diminish the mmf as the desired ac signal forces the non-collapsing field in the negative direction (and back again)???
Maybe, in doing so, they exhibit the same reluctance (and hence hysteresis) wiggling/expanding/condensing in response to the ac modulation of the DC field you so strongly believe extinguishes this effect?
I don't need to "read your lips"... I'll just keep on listening to non-distorted ss music, and focus on speaker improvements, which are orders of magnitude more rewarding than this discussion 😉
John L.
Jlsem, I have removed your offensive post. Any more like that and you'll be spending time in the bin.
- Status
- Not open for further replies.