sully said:
Is that really you, John? Doesn't sound like you. You've been going on about Hawksford and the errors regarding skin effect for months. And you recommend closing this thread after a week?
Why not address the errors? How will anyone know what's error and what's not otherwise?
Again, this doesn't sound like you at all. Sure someone didn't hack into your account?
se
Steve Eddy said:
At least the skin thread is doin some neat pics..
Here?? words like "efficacious"...pleeeeeaaaassseeee...talk english, guys....
Hey...Hawksford's wrong...plain and simple..He's got guys believing there's 2.93 meter per second 50 hz signals runnin through their speaker wire, for gods sake.. (that's like "ache", not the japanese wine).. It's gonna take a while to show that...and I'm patient..
And his scientific method.
........I'll be nice.....I have nothing to say regarding that....I am hitting the silver bird saturday morning...goind to bermuda for a long deserved one..with a very nice friend...
Do you really expect me to address the technical issues??? Pleeeeeeese...as of tomorrow at 5???the internet will cease to exist...till the 27'th.
BTW, can't ya tell...I posted my pic, and nobody noticed...crap...I feel so used...so cheap...
Oh well..
Later dude..
Cheers, John
Ps..wanna keep me interested?? post some pics...baby, baby, love doze e/m field pics...the personalities?? I can go to prophead if I want to see that stuff..
sully said:Here?? words like "efficacious"...pleeeeeaaaassseeee...talk english, guys....
I'll send you a dictionary for Christmas.
Ah, ok. I see what this is about.
Actually I did see your pic. And I asked SY if he could move that to another thread so people wouldn't complain about it having nothing to do with the original question. I hadn't heard back from him and then got sidetracked by Eric's and Frank's posts.
Sorry 'bout that.
Anyway, I haven't any problem with your pic and post. I'm simply saying that it seems at odds with other references that have been posted.
One says that twisting works by way of keeping induced voltages as common-mode as possible by way of equidistance to the source of the two conductors, the other says that twisting works by way of local cancellation within the cable itself.
The reason I find this dichotomy intriguing is that if the latter is true, it would seem to obviate any need for a balanced interface as the induced noise would be a function of the cable, not the interface.
se
Re: ok...ok...one comment..
Then that seems to be in contradiction to the Whitlock reference, which says:
AC magnetic fields induce voltages in wires. The magnitude of the voltage depends on the strength of the field, which depends on the distance between wire and field source. Therefore, equal voltages will be induced in the two wires of a balanced cable only if they are equidistant from the field source. Any voltage difference is added to the signal as noise.
Twisting of the two wires is a first-order technique to make induced voltages identical by averaging the physical positions of the wires.
In other words, you seem to be saying that the distance of each conductor from the source is wholly irrelevant. That it's only the effective loop area and that twisting reduces the effective loop area to the area of a single twist.
se
sully said:
Then that seems to be in contradiction to the Whitlock reference, which says:
AC magnetic fields induce voltages in wires. The magnitude of the voltage depends on the strength of the field, which depends on the distance between wire and field source. Therefore, equal voltages will be induced in the two wires of a balanced cable only if they are equidistant from the field source. Any voltage difference is added to the signal as noise.
Twisting of the two wires is a first-order technique to make induced voltages identical by averaging the physical positions of the wires.
In other words, you seem to be saying that the distance of each conductor from the source is wholly irrelevant. That it's only the effective loop area and that twisting reduces the effective loop area to the area of a single twist.
se
Your Intellectual Challenge For Today.......
"Do you really expect me to address the technical issues??? Pleeeeeeese...as of tomorrow at 5???the internet will cease to exist...till the 27'th."
That's a good one John - run away and leave us other guys with the mess eh !!!.
I/we do sincerely hope you have a very nice time with your "very nice friend."
Steve, you clearly have problems with the information in the references given.
I suggest you go off and find all that you can about the subject, even go to the local university library, even go and talk in person to some expert lecturers, then collate and condense the newfound information that you aquire, and then post your own web page about it.
After then you can tell all of us about it, and YOU can be the new expert regarding balanced systems.
Until then as far as I am concerned your part of the internet does not exist.
Fellow members here try to help you, and you have been given gems of information, but it is clearly not working.
Information is being given but is not being recieved - if you do your own research this may change.
For your convenience, here is a long list of transformer manufacturers to irritate - http://www.transformers-ez.com/transformers/0035114_0035162_1.html
Eric.
"Do you really expect me to address the technical issues??? Pleeeeeeese...as of tomorrow at 5???the internet will cease to exist...till the 27'th."
That's a good one John - run away and leave us other guys with the mess eh !!!.
I/we do sincerely hope you have a very nice time with your "very nice friend."
Steve, you clearly have problems with the information in the references given.
I suggest you go off and find all that you can about the subject, even go to the local university library, even go and talk in person to some expert lecturers, then collate and condense the newfound information that you aquire, and then post your own web page about it.
After then you can tell all of us about it, and YOU can be the new expert regarding balanced systems.
Until then as far as I am concerned your part of the internet does not exist.
Fellow members here try to help you, and you have been given gems of information, but it is clearly not working.
Information is being given but is not being recieved - if you do your own research this may change.
For your convenience, here is a long list of transformer manufacturers to irritate - http://www.transformers-ez.com/transformers/0035114_0035162_1.html
Eric.
SY said:Sorry, I got sidetracked myself.
In any case, the pic seemed to cause no complaints, so out of sheer laziness, I'll leave it where it is unless there's a groundswell of opinion.
I was talking about the complaints that discussion regarding balanced interfaces did nothing to answer the original poster's question.
se
Re: Your Intellectual Challenge For Today.......
Yes, I do, as some references seem to be at odds with others, as I've explained several times already.
Finding all one can about a subject often simply raises more questions. You seem to presume that all information on a given subject is accurate.
Because some gems seem to be at odds with other gems.
You again presume that all the information given is factual.
se
mrfeedback said:
Yes, I do, as some references seem to be at odds with others, as I've explained several times already.
Finding all one can about a subject often simply raises more questions. You seem to presume that all information on a given subject is accurate.
Because some gems seem to be at odds with other gems.
You again presume that all the information given is factual.
se
Let me explain perhaps a bit better what I've been trying to say here.
Both the Drude model and the quantum mechanical (QM) model will both give you accurate predictions with regard to electrical conduction under most circumstances and can often be used interchagably.
Similarly both the Whitlock and the Capgo references tell you that twisted pairs result in less induced noise compared to parallel pairs.
However in terms of explaining exactly what's going on with regard to electrical conduction, the Drude model is incorrect whereas the QM model is much closer to the truth.
From my point of view, I see the same conflict here with regard to the Whitlock reference versus the Capgo reference. And just as the Drude model and the QM model can't both be correct, I don't see how the Whitlock and Capgo references can both be correct with regard to explaining what's actually going on with regard to twisted pairs.
I'm simply trying to sort out which is correct as each has different implications with regard to addressing noise issues.
For example, the Whitlock reference states that that as long as the conductors are equidistant from the source, it doesn't matter what the spacing between them is, the induced noise will be common-mode and rejected by differential input, that twisting simply makes the two conductors on average equidistant from the source keeping the induce noise more common-mode than differential.
The Capgo reference implies (and which is stated more directly by sully) that it's all about loop area, regardless of any equidistance from the source and that twisting alternately flips the loop relative source so that the voltage induced at each twist is opposite that of the next twist, canceling each other so that your effective loop area is defined by the pitch of the twist.
After giving it some thought, it's looking more to me like the Whitlock reference is the Drude model and the Capgo reference is the QM model.
Which brings things full circle to the "statement in question" of:
Noise induced by varying magnetic fields will be differential and not rejected by a balanced circuit, which rejects common mode noise.
This statement, while inconsistent with the Whitlock reference, is consistent with the Capgo reference. And if noise induced by varying magnetic fields is ALWAYS induced differentially, then keeping conductors equidistant from the source cannot help. Only keeping the loop area as small as possible can help. And twisting does that by alternately flipping the loop with respect to the source.
se
Both the Drude model and the quantum mechanical (QM) model will both give you accurate predictions with regard to electrical conduction under most circumstances and can often be used interchagably.
Similarly both the Whitlock and the Capgo references tell you that twisted pairs result in less induced noise compared to parallel pairs.
However in terms of explaining exactly what's going on with regard to electrical conduction, the Drude model is incorrect whereas the QM model is much closer to the truth.
From my point of view, I see the same conflict here with regard to the Whitlock reference versus the Capgo reference. And just as the Drude model and the QM model can't both be correct, I don't see how the Whitlock and Capgo references can both be correct with regard to explaining what's actually going on with regard to twisted pairs.
I'm simply trying to sort out which is correct as each has different implications with regard to addressing noise issues.
For example, the Whitlock reference states that that as long as the conductors are equidistant from the source, it doesn't matter what the spacing between them is, the induced noise will be common-mode and rejected by differential input, that twisting simply makes the two conductors on average equidistant from the source keeping the induce noise more common-mode than differential.
The Capgo reference implies (and which is stated more directly by sully) that it's all about loop area, regardless of any equidistance from the source and that twisting alternately flips the loop relative source so that the voltage induced at each twist is opposite that of the next twist, canceling each other so that your effective loop area is defined by the pitch of the twist.
After giving it some thought, it's looking more to me like the Whitlock reference is the Drude model and the Capgo reference is the QM model.
Which brings things full circle to the "statement in question" of:
Noise induced by varying magnetic fields will be differential and not rejected by a balanced circuit, which rejects common mode noise.
This statement, while inconsistent with the Whitlock reference, is consistent with the Capgo reference. And if noise induced by varying magnetic fields is ALWAYS induced differentially, then keeping conductors equidistant from the source cannot help. Only keeping the loop area as small as possible can help. And twisting does that by alternately flipping the loop with respect to the source.
se
So you'd predict, then, that a wire with very few twists would, in the presence of relatively homogeneous noise fields, show the same noise pickup as a wire with many twists? (I'm assuming in both cases an odd number of twists, i.e., an even number of loops)
Side note: "QM Model" doesn't say much. Which QM model? There's a million of 'em, and they're all approximations of one form or another. Not that it's on topic here...
Side note: "QM Model" doesn't say much. Which QM model? There's a million of 'em, and they're all approximations of one form or another. Not that it's on topic here...
SY said:
In terms of twists per unit length, I'd predict that the more twists you'd get less noise pickup because as you have more twists per unit length, you reduce further the effective loop area.
Bottom line, I'm finding myself in agreement with the "statement in question" and that the voltage induced into any loop, balanced, unbalanced or otherwise by way of magnetic fields will be differential. That there will be no induced common-mode voltage as the Whitlock reference states. That twisting reduces the effective loop area by way of local cancellation of those induced differential voltages.
In other words, I'm now finding myself in agreement with the Capgo reference and in disagreement with the Whitlock reference.
I mean the QM model that doesn't violate Pauli as the Drude model does.
se
JOE DIRT® said:
My sentiments exactly.
I tried a simulation of an ideal transmission line having a delay of 45.2 nsec, with a source having resistive impedances ranging from 5 Ohms to 5 KOhms, with a resistive load of 20k. Next to this I put a simple voltage divider having the same source and load. I did a plot which takes the ratio of the output voltage of the transmission line with that of the voltage divider in order to keep everything to 0 dB at DC. The 45.2 nsec assumes a coax length of 30 ft, with a dielectric constant of 2.2 (teflon). That's a nice long cable. LTSpice does not have a lossy line model, so I'm stuck with a lossless line model. However, the popular Belden 89259 has an attenuation per 100 ft of 0.9 dB at 10 MHz, which is not too bad, even at this length. The circuit is shown below.
Attachments
...and here's the plot. The trace with the peaking is 5 Ohms source. The flat trace occurs when the source is matched to the 75 Ohm characteristic impedance of the line. The one with the rolloff is the 5k source - meant to simulate a 20k pot at 6 dB down from maximum. These are surprisingly good results, even though the line has almost a purely reflecting load.
So my opinion in all of this is: use an active preamp with a 75 Ohm resistor right at the op-amp output. Make every attempt to preserve the 75 Ohm impedance up to as high a frequency as possible. So you might use the microstrip formulas to determine the correct trace width for your board thickness and dielectric constant to give a 75 Ohm microstrip. Use a good 75 Ohm cable, such as Belden 89259. Make the interconnect cable long and place your monoblock amplifiers as close to your speakers as possible.
So my opinion in all of this is: use an active preamp with a 75 Ohm resistor right at the op-amp output. Make every attempt to preserve the 75 Ohm impedance up to as high a frequency as possible. So you might use the microstrip formulas to determine the correct trace width for your board thickness and dielectric constant to give a 75 Ohm microstrip. Use a good 75 Ohm cable, such as Belden 89259. Make the interconnect cable long and place your monoblock amplifiers as close to your speakers as possible.
Attachments
Steve Eddy said:And while you're installing that 75 ohm resistor, better ditch those RCAs and XLRs and replace 'em with BNCs.
Well, many RCAs have 75 Ohm characteristic impedance. But probably not most of the "audiophile grade" ones
Didn't know there were 75 Ohm BNCs. Thought they were all 50 Ohms.
I think both Canare and ADC make "true" 75 ohm BNC.
(you know its late when you stare at the keyboard wondering where you left the ohmega)
Why not an "F" connector? w/ Quad sheild RG-6U. Its not as tactile as the Belden RG-59 (few things are) but for long runs the quand sheilding would have to be better than the RG-59 single copper braid.
(you know its late when you stare at the keyboard wondering where you left the ohmega)
Why not an "F" connector? w/ Quad sheild RG-6U. Its not as tactile as the Belden RG-59 (few things are) but for long runs the quand sheilding would have to be better than the RG-59 single copper braid.
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.