14ga or 12ga...Which is more effiecient and why?

[iwire]

It comes into play a bit for higher currents. That's one of the reasons tubing is used in sub-stations for busbars; they weigh less than solid for a given ampacity.

They could be hollow for the heat dissipation aspect just as much as the skin effect.

I don't really know why they go hollow.

Sometimes it seems to me that people gravitate toward 'skin effect' because it sounds technical. I could be entirely wrong about it, it's just always struck me that way.

 

[gar]

090922-1838 EST

iwire:

Consider a 2500 mcmil solid copper wire, diameter is about 1.58", and a copper tube of twice this diameter and the same cross-sectional area of copper. My calculations indicate this would be a tube of OD 3.16" and an ID of 2.72". Thus, a wall thickness of 0.22". The weight per foot would be the same as solid wire.

According to Okonite's table the AC/DC resistance of the solid 2500 wire would be 1.326 at 60 Hz. Assume that going to this tube size reduces the ratio to near 1.000, then there is a reduction of 32.6 % in power dissipation for the same current in both conductors.

Next assume that because by doubling the OD of the conductor the heat transfer thermal resistance has been cut in half, then one could double the power dissipation in the conductor for the same temperature rise. Thus, current could be increased about 41%.

In this size range both skin effect reduction and increased cooling are somewhat equally effective. Much is to be gained by the tubular bus bar. The comparison is obviously quite dependent on my assumptions.

If one had long bus bars or conductors water cooling could be applied thru the tube for even greater benefit.

You are correct that cooling is a major factor, and should be more important relative to skin resistance in smaller wire sizes.

Someone should check my math to see if there are any mistakes.

.

 

[hurk27]

They could be hollow for the heat dissipation aspect just as much as the skin effect.

I don't really know why they go hollow.

Sometimes it seems to me that people gravitate toward 'skin effect' because it sounds technical. I could be entirely wrong about it, it's just always struck me that way.

Below 100hz skin effect wouldn't come into play, I would bet cooling/capacity is the real reason.


Here is food for thought:

Create a Graff, at the bottom across, put increases of frequency from DC up to 100 MHz
On the left side put a impedance table from the bottom to the top, calculations could find the correct scale, but this is just for the visual reference.

Now draw a line from the bottom left corner to the top right corner, this line represents skin effect, now notice that it is linearly, to frequency and impedance.

So the higher frequency the higher skin effect = higher impedance
Why? simple the more skin effect the less copper used in current carrying.

The main reason lightning don't see that little #6 going to that little ground rod
But will see a large flat conductor with a lot of surface area.

 

[jim dungar]

Sometimes it seems to me that people gravitate toward 'skin effect' because it sounds technical. I could be entirely wrong about it, it's just always struck me that way.

I agree. At 60Hz skin effect does not exceed 1.00 until 233kcmil. The NEC ampacities tables take skin effect into account.