reducing the skin effect

[chopficaro2]

take a look at the sketch of this wire:

http://www.mahanson.com/images/Hendrix Cable.jpg

those copper wires wrapping around the outside must be for reducing the skin effect right?

i have been reading up on the skin effect:

http://en.wikipedia.org/wiki/Skin_effect

i know the equation now, and its also telling me i should look up something called the proximity effect, which may or may not have any effect in this situation

my question is, that if i know the dimensions of this wire, the thickness, degrees per feet of how the outer wires wrap around etc.
can i calculate the skin effect for this oddly dimensioned cable?
or is the only way to know how different wrapping techniques will effect the skin effect to try them in an experiment?

 

[davidbeach]

They are shield wires and have nothing to do with skin effect.

 

[jhonsom]

As you strand the conductors the skin effect reduces, you find the formula in textbooks.

 

[chopficaro2]

its not on wikipedia, im having trouble finding it. does anyone here know where i can find the formula online?
also, i would think that stranding the wires would cause some inductance, do i have to account for that when calculating voltage drop across distances?

 

[7anoter4]

Allied Wire & Cable notes:
Insulation Shield:
A semi-conducting thermosetting layer, applied in a triple extrusive process, plus a concentric serving of solid copper wires acting as both a drain wire shield and as a grounding (bonding) conductor.
See:

http://www.awcwire.com/ProductSpec.aspx?id=HVTECK-1C-Shielded-Armored-Power-Cable-15kV-220mils.

If you'll take the concentric neutral in contact with insulation shield as a drain-wire then as per IEEE "The role of the drain wire in modern foil cable shields":
"Theoretical and experimental analysis of a wide range of shields that combine an electrically thin cylindrical shield with a drain wire/current diverter shows that the surface transfer impedance can be characterized as a solid foil shield in parallel with the impedance of the drain wire. The drain wire/current diverter should be modeled as an inductor and a frequency-dependent resistor in series. The resistance is frequency-dependent because of skin depth effects. Neglecting high-frequency aperture coupling, the surface transfer impedance of a foil shield with drain wire can be characterized in three frequency regimes. At low frequencies, the transfer impedance is determined by the resistance of the drain wire. At intermediate frequencies, the resistance of the drain wire increases due to skin depth and inductive effects. For most cable shields, the skin depth effect will dominate. At higher frequencies, for cable samples that use good (circumferential) shield termination techniques, the surface transfer impedance is determined by the transfer impedance of the foil shield"
See:

http://ieeexplore.ieee.org/Xplore/l...0473789.pdf?arnumber=473789&authDecision=-203

So, for a 15 kV concentric wire underground cable - where the current frequency will be 50 or 60 Hz [low frequency] no skin effect or proximity effect has to be taken into consideration .

 

[ScottyUK]

jhonsom,

If you strand the conductors AND INSULATE THEM FROM EACH OTHER the skin effect reduces. That's how Litz wire is made. Ordinary stranded cable experiences skin effect just like solid stuff of the same cross-section, except stranded wire is fractionally larger in diameter because of the imperfect fill factor and thus has a slightly lower AC resistance.


----------------------------------

If we learn from our mistakes I'm getting a great education!

 

[alehman]

http://en.wikipedia.org/wiki/Skin_effect

Alan
“The engineer's first problem in any design situation is to discover what the problem really is.” Unk.

 

[ausphil]

this is a pretty good document, which doesn't rely on wikipedia.



ausphil

 

[ScottyUK]

Aside: the entire Cahier Technique series is worth taking the time to download. Kudos to Schneider for making it available.


----------------------------------

If we learn from our mistakes I'm getting a great education!

 

[waross]

Hi Scotty;
Over the years I have had a number of discussions with audio types Re: skin effect and stranded wire. Sort of like discussing grounding with a data type.
But back to first principles. I subscribed to the position that skin effect was a result of magnetic fields which do not respect insulation and pass freely through insulation.
Stranding intruduces air gaps and the magnetic field reduces as the inverse square of the distance.
Am I off base in suspecting that the Litz effect is obtained by the spacing provided by the insulation rather than by breaking up the conduction between strands.
I suspect that if a Litz type wire was insulated with a material that allowed current to pass freely from strand to strand but blocked longitudinal currents that the Litz effect would still be seen, as a result of the greater spacing between the conductors.
I have long suspected that plastic jacketted 1/4 inch copper pipe would deliver current to speakers with less distortion from skin effect than the best available speaker wire of the same cross sectional area.
The few audio techs that I susgested this to were horrified at the idea.
Comments are encouraged.

Bill
--------------------
"Why not the best?"
Jimmy Carter

 

[waross]

PS: Pipe to reduce skin effect? Tubular IPS bus? Hollow core power cables?


Bill
--------------------
"Why not the best?"
Jimmy Carter

 

[ScottyUK]

I love the idea of using microbore pipe to connect up your speakers. Must be quite an installation you have there Bill...

With a Litz wire the individual conductors are transposed from the centre to the outside of the winding and back so no one conductor is the 'centre' conductor for the whole length. After a few transpositions the impedance of each insulated strand is effectively normalised and each strand carries its share of the current. In conventional stranded cores where the conductors are in contact with each other, those in the centre carry less current than those at the outside due to skin effect.

Tubular bars are certainly used in very high current applications where skin depth becomes significant compared to the conductor cross-section. I think most hollow core cables are usually to allow passage of pressurised oil rather than to reduce skin effect but I could be wrong. Won't be the first time today.


----------------------------------

If we learn from our mistakes I'm getting a great education!

 

[DRWeig]

Bill,

In another thread, you asked about measuring L for solid versus stranded wire -- sorry I never got back to you.

I had 10 AWG of both types (THHN building wire), about a 2-meter length, and swept them both from 60 Hz to 2 MHz at a constant current of 120 mA RMS (really pure sine wave). Layout was a simple U shape, about 10 cm between leaving and returning wire to source.

No measureable difference (to 1 mV resolution) below about 900 Hz. From there to 200 MHz, the stranded conductor actually exhibited about 30 percent more voltage drop as the solid one, and of course the voltage drops increased with increasing frequency. Worked out to somewhere around 5 microH for solid, 6.5 microH for stranded. (sorry, couldn't remember the alt-code for small mu).

I was surprised, as I had always believed that stranding would reduce inductance. At 2 MHz, the drops were on the order of 0.75 VRMS solid versus 0.98 VRMS stranded.

I'm fairly confident in my measurement equipment (Tektronix analog scope) and technique -- but I would really appreciate it if anyone else would repeat this test.

I couldn't make measurements with the wires straight -- the impedance of my power leads and measuring leads became a real issue.

Good on ya,

Goober Dave

 

[DRWeig]

Oops, pardon my typo in post immediately above. In one instance I put 200 MHz -- should be 2 MHz.

Shame on me,

Goober Dave

 

[fsck]

I have a section of hollow copper~0.75" in OD that came from the original Hoover dam feed to LA. Seems they rerouted it recently and made souvenirs out of the few miles.

It's of interesting construction. Think tongue and groove wood; there are ~12 sections interlocked around the circumference.

 

[davidbeach]

An aside: in TGML, any Greek letter can be entered as [ignore][&letter;][/ignore]. So a [μ] is entered as [ignore][μ][/ignore] while a [Μ] is entered as [ignore][Μ][/ignore]. Likewise [Ω] is [ignore][Ω][/ignore] and [ω] is [ignore][ω][/ignore].

 

[DRWeig]

Thanks David!

Goober Dave

 

[waross]

Thanks Scotty and Dave.
I wasn't aware of the transpositions in Litz wire Scotty.
The severity of inductive effects are often the products of current and frequency. The electronic folk tend to think of the effects as frequency related due to the small currents that they work with.
Power guys tend to think of the effects as current related because of our fixed frequencies and high currents.
Hence hollow bus, high voltage conductor bundles which improve reduce skin effect as well as providing better cooling and reduced corona.
I agree with you that most often hollow conductors were made that way to allow the circulation of the insulating oil, but reduced skin effect would be a bonus.
Re the transpositions in Litz wire. This has also been done with power cables. For a time, a system named "Flexi-Bus" was installed for feeders. It was a system of insulators and supports for cable.
Mulitple single core insulated cables would be supported and transposed about every meter. There may be 12 cables, 15 cables or more. The transpositions were such that each core occupied every position in the bundle for a distance. Installation was fairly labour intensive. These installations would be used for 480 Volt or 600 Volt three phase circuits.

Bill
--------------------
"Why not the best?"
Jimmy Carter

 

[waross]

Sorry I got hauled away to spend quality time with my 4 1/2 year old son before I finished. Thanks Dave for the information.
Re the speakers. I didn't install the speakers. The suggestion originated when I saw a large sum being spent on high grade speaker wire for a public address system at a school on a very tight budget. The proposed wire was just fancy extra flexible stranded wire, not Litz. I first suggested that using solid or seven strand building wire to power the speakers would be a fraction of the cost and work just as well. Even if the building wire was a couple of sizes larger to reduce the current density and skin effect.
I also suggested that an even better way to improve fidelity and avoid skin effect would be to use insulated 1/4 inch OD copper pipe for the installation. This would still be several times cheaper than the cable that was used.
But, when the guy on the sound system knows that youi have to use "special" speaker wire for speakers, you're going to see very expensive wire with "SPEAKER" printed all over it.

Bill
--------------------
"Why not the best?"
Jimmy Carter