blah
EM theory??? That would be the exponential approximations...
The actual effect is about 1/6th that..but, the bessels are too complex for most. so they simplify..
It's not skinning that's the problem...it's the integral e/m field w/r to cross section..
For audio...20 hz to 20K..that is a wee bit more complicated..
It's too easy to say is doesn't matter...or, that it does..
Most have no clue..
Unfortunately, e/m theory as taught at grad and phd level is sadly lacking..
Sullyu
EM theory??? That would be the exponential approximations...
The actual effect is about 1/6th that..but, the bessels are too complex for most. so they simplify..
It's not skinning that's the problem...it's the integral e/m field w/r to cross section..
For audio...20 hz to 20K..that is a wee bit more complicated..
It's too easy to say is doesn't matter...or, that it does..
Most have no clue..
Unfortunately, e/m theory as taught at grad and phd level is sadly lacking..
Sullyu
piertr said:
I am sure that what you mentioned above has nothing to do with the material of the wire and geometry.
maybe you can educate all of us on what will happen to the electrons when you run 20khz signal over wires thinner than 18awg.
Hopefully those poor electrons don't have to walk in thin air outside of the wire.
millwood said:
I don't know that we don't get fooled in this case. There have been a number of rather big holes pointed out in Hawksford's Essex Echo articles.
What I find odd about this case is that for someone who is not at all shy about publishing in professional journals and has dozens of articles and papers published in the JAES and IEE on such esoterica as "fuzzy distortion" (see Malcolm's Publications) he's chosen not to publish on this issue and has relegated it to consumer audio magazines.
se
Steve: ""And don't get sully started on Hawksford""
Who...Me???
Steve: ""And for crying out loud, we're talking about reducing the diameter of the conductive portion of the wire by something on the order of less than a micron. Do you know of anyone producing wire to such tolerances?""
I do know of manu's that try..major pain in the A..they can't keep the cross section of #32 phosphor/bronze (low thermal conductivity) to less than 4% error..really makes production line acceptance a headache. Ribbon wire, red/blue/yellow/black..all with different resistances, makes measuring the Liquid helium level probes a PITA.
Piertr: ""EM theory will tell you that skin depth at 20kHz is roughly equal to the radius of an 18AWG wire""
Wrong equations..use bessel instead.. Since this is now a skin thread, I'll find some links..Scots guide is a start, but he negs out at less than 5 depths..there's some graphs of bessel vs exp...gotta find them..
Steve: ""How does a twist constitute a local loop when there's no conductive path between the two wires making up the twist? Last I looked, a loop constituted a complete circuit path. Where does such a loop exist within a single twist?""
He's talking about the twist pitch producing loops that alternate polarity..for low gradient flux fields, that will integrate out..It's only for the high gradient stuff that the high pitch angle stuff fails to cancel loop intercepts..
Sully
Who...Me???
Steve: ""And for crying out loud, we're talking about reducing the diameter of the conductive portion of the wire by something on the order of less than a micron. Do you know of anyone producing wire to such tolerances?""
I do know of manu's that try..major pain in the A..they can't keep the cross section of #32 phosphor/bronze (low thermal conductivity) to less than 4% error..really makes production line acceptance a headache. Ribbon wire, red/blue/yellow/black..all with different resistances, makes measuring the Liquid helium level probes a PITA.
Piertr: ""EM theory will tell you that skin depth at 20kHz is roughly equal to the radius of an 18AWG wire""
Wrong equations..use bessel instead.. Since this is now a skin thread, I'll find some links..Scots guide is a start, but he negs out at less than 5 depths..there's some graphs of bessel vs exp...gotta find them..
Steve: ""How does a twist constitute a local loop when there's no conductive path between the two wires making up the twist? Last I looked, a loop constituted a complete circuit path. Where does such a loop exist within a single twist?""
He's talking about the twist pitch producing loops that alternate polarity..for low gradient flux fields, that will integrate out..It's only for the high gradient stuff that the high pitch angle stuff fails to cancel loop intercepts..
Sully
sully said:
I only see one actual loop here. One whose segments alternate their geometric positions in a helical fashion. Let's look at the original quote:
The wires form a local antenna loop, but after the next twist the loop is inverted so a voltage with the opposite polarity is generated. This will cancel the voltage from the first loop.
If this were the case, then it seems that the induced voltage would be entirely dependent on the pitch of the twist and wholly irrespective of the overall length of the line. In other words, for a given pitch and a given magnetic field, the induced voltage would be the same regardless of how short or how long the line is.
What am I missing here?
se
Millwood: re:hawksford ""maybe he isn't 100% comfortable about his writting on this particular subject?""
He is a monster with the dsp and signal stuff..But I would agree..e/m theory and test is not his forte..Switching steel wire in the middle for test results really troubles me..
Sy: I've no idea what "deep link" is. Does that mean I'll have to bring in my ten year old as a consultant, telling me what to do?
I hate it when that happens..
Sully
He is a monster with the dsp and signal stuff..But I would agree..e/m theory and test is not his forte..Switching steel wire in the middle for test results really troubles me..
Sy: I've no idea what "deep link" is. Does that mean I'll have to bring in my ten year old as a consultant, telling me what to do?
I hate it when that happens..
Sully
Steve: ""The flux rate of change is "opposite" in what respect? Are you saying the flux changes direction?""
No..consecutive loops receive the flux in reverse..
Imagine one wire as pos..other as neg..
one twist: flux goes around pos only, between it and neg..
next twist: flux goes around neg only, between it and pos..
Each loop has it's generated emf due to flux rate of change..but consecutive loops have opposing voltage polarities..
A pic will help..but, I'm heading to the car..got a dehumidifier condensate pump to hook up before the bermuda trip..
Sully
No..consecutive loops receive the flux in reverse..
Imagine one wire as pos..other as neg..
one twist: flux goes around pos only, between it and neg..
next twist: flux goes around neg only, between it and pos..
Each loop has it's generated emf due to flux rate of change..but consecutive loops have opposing voltage polarities..
A pic will help..but, I'm heading to the car..got a dehumidifier condensate pump to hook up before the bermuda trip..
Sully
sully said:
Again, I only see one loop. Sure, each segment of the loop that makes up each twist has each wire alternating at different distances relative to the source so as they go back and forth the EMF across one segment of one wire will be either higher or lower than the EMF across one segment of the other wire producing a differential voltage between the two wires which will change polarity as the wires alternate back and forth.
But I don't see how you get any cancellation here.
Let's just follow the EMF across each segment of just one wire.
When one segment is closer to the source, the EMF across it will be higher. When it changes positions and is farther from the source, the EMF across it will be lower. However along the wire, the polarities of each segment will be (for example):
+[SEGMENT 1]- +[SEGMENT 2]- +[SEGMENT 3]- etc.
So as you proceed along the length of the wire, the EMF voltages of each segment add. Just as they would along the length of wire that's part of a straight, non-twisted parallel pair. So that the longer the length of the pair, the greater the voltage across each wire.
I don't see any cancellation going on in the cable itself. I only see twisting as reducing the differential voltage between the wires by keeping each wire on average the same distance from the source as the other wire so that on average, each wire has the same voltage across it so that it can then be rejected by a differential input.
This as opposed to a straight parallel pair where one wire is always either closer to or farther away from the source which would result in a greater differential voltage between the wires which would not be rejected by a differential input.
se
A Balanced View.........
........by keeping each wire on average the same distance from the same AC magnetic field source as the other wire so that on average, each wire has the same INDUCED EMF WRT THE earth of the system so that it (the common mode voltage) can then be rejected by a differential input.
Hi Steve,
With the above editing I agree with what you are trying to say - at least with respect to my reference books on the subject.
Of strong importance is the impedence to system ground of each of the balanced lines (at source and load ends) in order that any induced EMF's are equal, and then able to be eliminated in the balanced differential line recieving stage.
If these impedences are not equal, un-equal induced EMF's will result, and a valid output will present at the output of the recieving balanced differential stage.
Eric.
Steve Eddy said:
........by keeping each wire on average the same distance from the same AC magnetic field source as the other wire so that on average, each wire has the same INDUCED EMF WRT THE earth of the system so that it (the common mode voltage) can then be rejected by a differential input.
Hi Steve,
With the above editing I agree with what you are trying to say - at least with respect to my reference books on the subject.
Of strong importance is the impedence to system ground of each of the balanced lines (at source and load ends) in order that any induced EMF's are equal, and then able to be eliminated in the balanced differential line recieving stage.
If these impedences are not equal, un-equal induced EMF's will result, and a valid output will present at the output of the recieving balanced differential stage.
Eric.
sully said:
Yes. Go up to the top of this page, click on "user cp" and then "Edit Options." Then go down to the "Messaging & Notification" section and turn off "Use 'Email Notification' by Default?"
Go back a page and in the "Subscribed Threads" section and click on "View All Subscribed Threads." Then click on "Unsubscribe To All Threads."
That should do it.
se
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.