|
Audio Asylum Thread Printer Get a view of an entire thread on one page |
For Sale Ads |
104.129.198.56
In Reply to: RE: Let's step out of audio for a minute! posted by morricab on August 20, 2015 at 01:41:16
"All of this is academic though because you cannot equate stable operation with changes in perception. Changes inconsequential to the basic OPERATION of the transformer could still have perceptible impact on an audio signal passing through the transformer. Never forget that in many cases a transformer in industry has a more or less constant load, whereas an audio transformer has an ever changing load, so things like permeability, hysterisis etc. take on different meanings beyond static values that most transformers are subjected to."
So you think audio is the only application that subjects a transformer to a varying load? The world has only two types of transformers, audio and power transformers?
Let's look at just my small slice of the electronics industry - broadcast engineering. Back in the analog days we had video isolation transformers that went from a 10hz to 10mhz* within 0.1db at a 1 volt max operating level! Think that silver wound tube output transformer is expensive! Inside broadcast grade analog VTRs we had many precision transformers both on the RF/video path as well as the servo systems that controlled the rotating heads and linear tape speed. Speaking of rotating heads, the head disk had four RF transformers that had to endure 14,400 rpm on the old 2in machines. Then there was the rotary transformer, yes a transformer that spins! Even you old VHS machines had one of those. Oh, yeah, the audio section of the VTRs also had a few crude transformers in comparison for the audio path.
I studied servo system engineering in college based on my fascination with VTR servo systems at the time. Back in the early 1980s analog servos still ruled. Differential transformers in sensor circuits were very common. The precision of these circuits made audio in comparison look like tin cans and a string.
And while I only have a cursory knowledge of related fields, I do know that many other areas of electronics have precision transformer requirements with voltages in the nanovolt range. A colleague of mine worked on gun targeting systems in the Navy. He once told me about precision transformers that putting a VOM across on the x1000 scale would vaporize the coil.
The point here is that transformer design has for many, many years, advanced far beyond the rather basic requirements for HiFi audio reproduction. Stability over time, temperature, and most other external influences are proven and well documented.
The issue is not about stability per se, the issue is about measuring it. And based on my experience, any parameter change that would result in a perceived difference in audio reproduction can certianly be measured quite easily with rather low cost test equipment to boot.
As for the audibility of fresh solder connections vs 100hour old solder connections, well that's just silly and has no room for discussion in formal electronics circles.
*to be fair these transformers did have an LC EQ network internally to meet this flatness spec. They were in fact lossy by a db or two which required a downstream amplifier to compensate as well as DC restoration.
Follow Ups:
"So you think audio is the only application that subjects a transformer to a varying load?"
No, I think it is the only application that subjects a transformer to a varying load where human perception is directly involved.
"And based on my experience, any parameter change that would result in a perceived difference in audio reproduction can certianly be measured quite easily with rather low cost test equipment to boot."
Maybe so, why don't you show us how easy it is? I have already mentioned several ways I think it COULD affect the sound and most are measurable. However, I don't think you or anyone else can PREDICT what the impact on sound will be because no one has done any kind of correlation study.
"As for the audibility of fresh solder connections vs 100hour old solder connections,"
Why? You don't think oxidation and other potential chemical/physical transformations couldn't impact the sound? A thin layer of oxide makes already a fairly good resistance.
The fact is that these devices are not perfect and are in some cases quite lossy and that lossy behavior will be most severe when SMALL changes are required of the transformer. Hysteresis and eddy currents will work against small changes due to resistance to change. This is why top audio designers like magnetic cores with high permeability for sensitive signals (like output of phonocartridges, for example). However, for a power amp it is more complicated because it has to handle power and the balance between responsiveness and saturation is just that...a balance.
How can you say things like this are not audible when in digital audio, for example, it has been shown that timing errors in the PICOseconds have audible consequences. Picoseconds!! No one would have imagined a prior that it would be possible to hear such a small timing error.
It seems that sensitive humans can also tell the insulation materials around wires. Malcolm Hawksford looked into this and it seems consistent with signal transmission theory given the way electromagnetic radiation doesn't really travel through a wire, it more travels around a wire. Even at audio frequencies there is some of this skin effect going on.
Linear distortion (like speaker frequency response) seems to be far easier for humans to adapt to than non-linear distortions...particularly when they change with frequency and/or amplitude. Transformer distortions are distinctly non-linear, once we get beyond the obvious FR.
You are reading too many audiophile magazines. Why not trying some electrical engineering texts if you really want to learn this stuff.
1) Yes, I have seen those jitter demonstrations from audiophile magazines. One I remember in particular showed jitter causing inter modulation components, which is what jitter does. However it was 140db down from reference level. Now if the magazine author and the audiophiles who soaked it up were to have studied electrical engineering, they would realize no human can hear -140db from reference. No speaker ever made is that sensitive either. And no power amplifier has a noise floor even close to that - especially a tube amp, and a SET is worse yet in terms of noise floor.
2) Malcolm Hawksford discovered nothing that has not been known since at least the 1920s. Skin effect is well known and does occur above DC, even at 0.1hz. However did the audiophile rag where you read that show any quantified numbers. If you do the math and understand electrical engineering basics, you will easily see how skin effect at 20khz is irrelevant.
"You are reading too many audiophile magazines. Why not trying some electrical engineering texts if you really want to learn this stuff.
"
I believe the jitter issue has been addressed in the AES so I will see what I can find...not just audiophile magazines.
Again, I will see if I can find the information because it was not in an audiophile "rag" as you put it.
Yes, it has been covered by AES and other standards bodies.The issue is in the numbers. The scale of magnitude. "Throwing deck chairs off the Titanic" or the more modern version, "throwing file folders out of the WTC" comes to mind.
Please do research the AES. You will find they stick to standard established electronic theory as a large majority.
Edits: 08/24/15
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: