Feeder efficiency

[ron]

Often we think about efficiency of transformers, motor,etc.

When is it worth it to upsize to the next size feeder? Is anyone familiar with the losses of a feeder of varied sizes and applications?

 

[iceworm]

...When is it worth it to upsize to the next size feeder? Is anyone familiar with the losses of a feeder of varied sizes and applications?

Interesting question. I've rarely taken the time to calculate copper losses. Probably should some day - if for no other reason than to increase my understanding.

But I do have some excuses as to why I haven't:
For sizing feeders, I'm looking at voltage drop at continuous load, voltage drop under motor starting. Generally, for inside process limits, my customers are not concerned with copper costs. They just want it to run, flatout and forever.

However, once the application gets outside of process limits, say 1000ft to 20,000ft, I have done a couple of copper cost calcs - but only far enough to see the answer was, "Buy the transformers". I have designed only a few (I hate design work - but I like design engineers**), but I have worked several, 600V, 2400V (single phase), 4160V (1ph and 3ph). It depends on the distance and the load.

And , of course you know about these guys:
http://www.copper.org/publications/pub_list/energy_efficiency.html
But I suspect about 99% of their literature is advertising hyperbole

If you do the research and calcs I'm interested in what you find.

ice

 

[Smart $]

Assuming you are referring to I?R losses, I've never been in a position to even do an assessment that would matter. If I were, and being a feeder, I probably wouldn't even consider assessing unless the load was constant. The I?R reference itself infers losses will be exponential with load current. As such, you'd likely assess based on the ROI, and you could even take it to the degree of including the impact on HVAC cost.

 

[ron]

Assuming you are referring to I?R losses, I've never been in a position to even do an assessment that would matter. If I were, and being a feeder, I probably wouldn't even consider assessing unless the load was constant. The I?R reference itself infers losses will be exponential with load current. As such, you'd likely assess based on the ROI, and you could even take it to the degree of including the impact on HVAC cost.

As pointed out, the Copper.org documentation "onesizeup" might be biased, but seems pretty believable. I will try to check their calculations but they have ROI / payback in a year or two.

I work on facilities that have pretty stead loads, sometimes 50-80% of the full amapacity .... essentially for ever, so if payback is that good, maybe I need to consider onesizeup more often.

 

[Jraef]

I've done it a couple of times officially, only because someone was paying me for my time to do it, and I think I did it once in here (or maybe another forum) not too long ago, mainly to prove to someone that it's hardly worth the time it takes to do it. In that last case the question was about (IIRC) there being enough energy savings to warrant using #12 instead of #14 on a given circuit, and as I recall, it saved something paltry like $0.60 per year, I can't quite recall exactly. But the point was, it would take something like 30 years or so just to recoup the difference in the wire cost, which follows the findings I came out with in the two cases where I was paid to evaluate it, even with larger conductors.

In all of those cases, I didn't even factor in an issue that someone else raised in that 12ga discussion, being the potential thermal issues in terms of reducing heating costs by the amount of radiant heating the wires give off into a conditioned space which might be lost by decreasing the I²R losses. My point is, you are dealing with misiscule numbers to start with, so including other minuscule numbers based on comparitive relevance becomes an exercise in the ridiculous very quickly.