LV cable sizing for efficiency 1

[DaveScott]

I was wondering if anyone had ever come across this?

The idea is that: in addition to volt drop; short circuit capacity; and conductor temperature restrictions, the cable is selected with long term costs in mind.
The long term costs of a cable are somehow related to investment interest, and ohmic heating costs. The cable is heated during use, and that heating represents an energy cost that can be substantial.

A cable that just satisfies the normal requirements, but is upgraded to a cable that is the next size up may seem like an extravagance, but it is easy to calculate the kWh costs in each case.

I first made some calculations some 15 years ago, but I seem to remember that for longer cable runs, the potential energy saving was comparable to an energy efficient motor.

I heard mention that a standard was being considered, but that was a long time ago.

Any comments?

 

[bacon4life]

I don't know about LV applications, but that is considered in transmission applications. For example ACSS (aluminum conductor steel supported) can be run in excess of 200C, but is going to have much higher losses than a somewhat larger standard conductor run at 50C to 100C.

 

[dpc]

Utilities do this type of evaluation for transmission lines. With an analysis program, it would be fairly easy to do what-if calcs for various size cabling.

For smaller runs (600A and below) my hunch is that the NEC is already so conservative that going larger than NEC requires might not buy you too much. For heavy runs (2000A, 3000A), it might be worth looking at. The skin effect somewhat diminishes the impact of larger wire, especially in the larger sizes.

Most industrial clients that I deal with would not be receptive to oversizing conductors now to save money later. They have a very limited time horizon.

You'd probably get more for your money by going after motor efficiency and transformer core losses. Very few industrial buyers even bother evaluating transformer losses.

 

[jghrist]

Economic sizing should definitely be considered for bare overhead conductor. Studies I have done on MV systems have indicated that initial peak loading of 30% of the thermal rating is economical.

The economics are considerably different for insulated conductor. I haven't looked at it in quite a while, but the economic loading for MV cable was at its thermal limit (at least if load growth is not a factor).

For cable in conduit, you also have to consider increased conduit size.

 

[DaveScott]

Thanks for the comments. Of course, for HV and MV, transmission losses are considered. Bread and butter stuff.

I find it interesting that, as far as my experience goes, it doesn't happen in industrial distribution.

 

[dpc]

Cable runs in industrial facilties are generally pretty short compared to utility circuits. The cable losses just aren't that significant when compared with other losses - especially process-related losses. When 40% of your energy is going up a stack, the impact of cable resistance losses is pretty small by comparison.

 

[jghrist]

Length is normally not a factor in economic conductor sizing because both the losses and the cost are directly proportional to the length. The cost of determining the economical conductor size is a fixed cost, so this may not be worth the effort for short lengths of cable.

 

[DaveScott]

Termination is also a cost which is fixed with length, but variable with cable size...

 

[DOCWin]

Try this:
IEC 60287-3-2
Title:
Electric Cables - Calculation of the Current Rating - Part 3: Sections on Operating Conditions - Section 2: Economic Optimization of Power Cable Size

And check also publication IEC 1059.

Here a tool for calcs:

http://www.leonardo-energy.org/drupal/node/459

Problem I faced in the past whit these calculation is: how to find reasonable values for the next 20/30 years for the economic parameters (discount rate, rate of increase of energy cost...)

 

[DaveScott]

DOCWin:

Wow, I'm impressed. Thank you very much. I tried out the tool and was quite surprised at the payback times (of course, using the default rates etc.)
Best regards.
Dave