OT: amtrak wreck

a o

standstill. That's not enough of an argument to establish the dependence of rolling resistance on train mass.

subside by 0.06 inches per train. Assuming 10 trains per day, that's aroun d 2 feet per year. The tracks would be level with the roadbed in at most 6 months, which would lead to a fairly startling number of derailments.

Maybe so, but your bald statement doesn't prove it. Show me, don't just te ll me. A few well-supported numerical estimates on your part would be help ful, for instance. Otherwise, who's in the armchair?

Cheers

Phil Hobbs

Well, a bigger conductor still has more surface area. May as well be a pipe for all the good the inside copper does.

I guess the stuff strung between those big towers is the industrial equivalent of Litz wire.

--

John Larkin         Highland Technology, Inc 
picosecond timing   laser drivers and controllers 

jlarkin att highlandtechnology dott com 
http://www.highlandtechnology.com

Yes, and kin effect is only one reason. The inside copper isn't cooled, either. Some transmission cable is essentially a pipe on some supporting structure (steel cable usually but also structures like a helix). Copper and aluminum don't have very good tensile strength, either.

Look closely at them. You'll often see three cables per phase, separated by perhaps a foot.

On 16 May 2015 07:08:42 -0700, Winfield Hill Gave us:

"relatively little" does not equate to any actual numeric figure(s).

On Sat, 16 May 2015 13:52:07 -0400, rickman Gave us:

There are materials which can stop an arrow or knife.

A tub full of it can be walked on. Stand still and you sink.

Same principal. The crushed stone elements bind upon high velocity delivery of force. Press a backhoe shovel into it and it gives and fills the shovel head with far less resistance.

On Sat, 16 May 2015 13:59:42 -0400, "P E Schoen" Gave us:

AND they are also typically arrayed in a quad spread across several inches which reduces the skin issues on the single element 'strands'.

eroUno:

the resistance of a 32mm copper wire is ~23% more at 60Hz than at DC

-Lasse

Here ya go:

Copper, 50 Hz, about 10 mm.

--

John Larkin         Highland Technology, Inc 
picosecond timing   laser drivers and controllers 

jlarkin att highlandtechnology dott com 
http://www.highlandtechnology.com

Good grief AlwaysWrong, you've been given the numbers.

The copper diameter of my HV cable section is 5 to 10x that.

--
 Thanks, 
    - Win

andstill. That's not enough of an argument to establish the dependence of r olling resistance on train mass.

ubside by 0.06 inches per train. Assuming 10 trains per day, that's around

2 feet per year. The tracks would be level with the roadbed in at most 6 mo nths, which would lead to a fairly startling number of derailments.

See page 71 Railroad Engineering 1982 Hay in Google books- rail deflection is a MAJOR uphill battle.,.,

andstill. That's not enough of an argument to establish the dependence of r olling resistance on train mass.

ubside by 0.06 inches per train. Assuming 10 trains per day, that's around

2 feet per year. The tracks would be level with the roadbed in at most 6 mo nths, which would lead to a fairly startling number of derailments.

There IS a startling "nummer"[RR-talk] of derailments of heavy freight, onl y really toxic stuff like bakken oil gets reported, ALWAYS channeled rainfa ll erosion of track bed.

On 16 May 2015 17:40:26 -0700, Winfield Hill Gave us:

So you just shit all over both efficiency as well as established convention. Good job... not.

A thick skin effect IS a good thing. A thin one is not. That is why power transmission is NOT at 400Hz or other high frequency.

HV lines are also such that most of the conductors in question DO get most of their cross sectional area utilized.

It can be inelastic and satisfy those criteria, consider spring and dashpot.

still 6 inches sounds unrealistic.

--
umop apisdn

** Well, rather thick walled pipe of up to about 4 inches dia.

** Bundling spaced cables in groups of 2 to 5 is the go at seriously high voltages. Primarily done to reduce corona discharge but also mitigates against skin effect and inductance losses plus runs cooler.

Based on DC current, 100mm dia Al plus steel cable has about 10 micro ohms of resistance per metre at room temp. I reckon skin effect and self heating must increase this number by many times.

With 200kms of line and thousands of amps flowing, suprising to have any power emerging at the far end at all.

... Phil

Oh, there'll be some inelasticity, for sure. I'm also sure that there are published curves of rolling resistance vs weight and speed for various lines.

Cheers

Phil Hobbs

Thanks for the info, forgot about the 'practicalities' of electrical distribution where the weather can be a bit daunting to survive, especially where it is not possible to make the system 'grounded' like here in the U.S. [I am assuming Finland soil is a lot like Norwegian soil.]

Learned a lot about skin effect after working with a MEMS motor design. The motor had over 1000 poles, so the drive was 1MHz, and desired to be

3MHz. You want skin effect? Work at those frequencies!

Learned two things:

1. Turned out 'leaving' unused conducting metal created a lot more loss. It wasn't just a matter of hollow aluminum tubes in a TV antenna to save money, but truly the unused metal in the centers 'ate' some energy. Skin effect is usually described as 'attenuation vs depth' and the phase shift, a 'rotation' as you go down in depth is left out of those papers. Currents actually go the WRONG way, subtracting from what you want and still consume energy.
2. Trying to 'simulate' Litz effect by using multiple strands made little difference. Envision a bus bar where the carriers all go to the two furthest separations of the rectangle. So I naively tried placing a multitude of wires to simulate the rectangular cross section BUT having the wires separated I THOUGHT would produce better results. NO! The second you connect the conductors in parallel at each end, the distributed inductance/resistance forced the carriers right out to the outside, almost as if everything was still a solid rectangle. ...very frustrating. However, I have seen some IEEE papers describing a 'pseudo' Litz effect from using stranded cabling, where the cross conductance is decidedly different than the longitudinal conductance. Just because its a paper doesn't make it so, but I never pursued checking data OR refining my analyses tools, and their premise does seems like a reasonable effect.

By the way, that MEMS motor has some incredible characteristics, like when the rotor and stator of this 1.5 inch diameter motor were put together, its overall thickness was around 80 microns, thinner than a sheet of paper, had NO wear surfaces, all air bearing, and had several oz-inch of torque and small mass so was up to 6,000 rpm in about 9 mS. and we routinely/indefinitely ran current densities through the copper that were equivalent to 100 Amps through a 36 gauge wire! Only if we went to 200 Amps equivalent densities did we lose a copper trace. As a lark, I once applied the design concepts to making a 'stacked' motor with diameter of

4-6 inches and lots of disks to make a 6 inch cube size [didn't check the heat] the idea was to make a small efficient electric car motor. Amazingly, using a standard multi-battery pack, the predicted output torque/power of such a 'block' was an electric car motor that had something like the equivalent of 360 HP gasoline engine, be like a corvetter engine, but only a six inch cube. And drain your batteries REALLY fast.

Quoting from your reply to my posting an actual numeric figure...

"I cannot imagine that it is worse at power line frequencies, so a third of an inch sounds ridiculous." Goose/gander here.

I highly recommend downloading a FREE PC Tool for finite element analyses called, femm 4.2 It's easy to use and with a bit of experimentation you can explore 'actual numeric' quantities to develop some intuitive undertanding of just how insidious skin effect is.

Oops, I forgot, also another useful FREE PC Tool is octave, a Matlab clone. It's easy to use and that 1/3 inch value came from a function I wrote where

skin effect = sqrt( 2/(w*permeability*conducivity) ), where everything is in MKS units: skin effect is in meters, etc.

It's an easy formula to remember BUT one must keep in mind that this standard formula only applies to a 'planar' wave and an infinitely thick flat block of material, with the wave hitting perpendicular to its surface. That's why for more complex structures AND non-planar waves finite element analyses can describe much better what's going on.

I don't know about current situation in Norway, but a long time ago they used 220 V delta and no grounding, 220 V loads were connected between two of the three phase.

In Finland the 230/400 V wye/delta is used and the star point is grounded both at the transformer and should be grounded at each building, to protect against hazardous voltages due to loss of the Neutral wire. The experimental 1500 Vdc feed is actually +/-750 Vdc. Even with a grounding loop is installed under the basement of the building, the grounding resistance can be several ohms, especially if the house is built on sand.