Long 240 V run

[junkhound]

OK, screwed up last summer (both me and customer).

background
Need service to new house site 870 ft up 10% grade hillsite. POCO HV line and transformer pad and one intermediate pull vault quoted at $xx last year (almost 4 figures), but was not ready to do the; digging. Customer responsible for trenching and sched 40 conduit. Luckily had bought mos conduit already. That price expired in Sept last year. NEW quote is TWICE the old one due to inflation (well OVER 4 figures). Would be even worse for customer if had not already gotten conduit for the job.

Got a 1000 ft spool of 3C 250 MCM with 1 AWG EGCs at an auction for about 12% new price, so looking at option of long run of 240 Vac.

So, the calculation question: Customer has past electric bills which run about 2500 kWhrs a month in winter, only aobut 1000 kW-hrs in summer (PNW). Is it a valid calculation to use the yearly total kW-hrs to calculate 'average' amps to determine line loss?
I think I can get the daily peaks to use, but would be easy to just use total kW-hrs to calculate line lossed and power cost.

Looked at it one way of assuming 20,000 kWhrs at 160 A, comes to line loss of 520 hours at 160 A = 1064 kW-hrs line loss (using 80 mohm line)
20 MW-hrs over 8759 hours is average of 2.3 kW; about 10 A, or 64 kW-hrs line loss

So, do I tell customer that his yearly power bill for losses would be $106 or $7? Big difference in ROI of outlay cost. Am thinking hte $7 number is more appropriate due to i^2*R ratio.

Or do I need to get customers old electric bills by the day?

 

[wwhitney]

Is it a valid calculation to use the yearly total kW-hrs to calculate 'average' amps to determine line loss?

No, because it's quadratic in current (I²R losses), not linear.

Simple example: I use 24 kWhrs over one day at 240V. So that's 100 A-hrs. If that's 100 A for 1 hr, the losses are 100²*R*1 hr. While if it's 4.2A for 24 hrs, the losses are 4.2²*R*24 hrs. So 10,000 * R vs 416 * R. I.e. 1/24 less for the slow steady load profile, in this example.

Cheers, Wayne

 

[Canton]

I’d ask the Utility to run the numbers/price to extend the MV line/transformer closer to the house.

Almost and just over (4) figures is cheap around here for a new service

 

[electrofelon]

I'll bet you'll never going to see the current above 60 amps, and 99% of the time it'll be below 40. I would probably use one set of 350 or 500 aluminum for that. But even 250 is probably fine honestly. Run it up as service conductors then only needs to be a three wire run. 250 urd, plexed, will fit in 2-in pipe and not be a bad pull.

 

[kwired]

Where is the meter? You won't see the resistive line losses in the kWhrs registered if metering is at the load end of the conductors you are concerned about.

You apparently are using more energy in winter than summer but with information given we still don't know if it is used in higher peak demands or if it is just longer run times at otherwise similar current levels.

 

[electrofelon]

Where is the meter? You won't see the resistive line losses in the kWhrs registered if metering is at the load end of the conductors you are concerned about.

You apparently are using more energy in winter than summer but with information given we still don't know if it is used in higher peak demands or if it is just longer run times at otherwise similar current levels.

Yeah here you can run service conductors as long as you want, but the power company doesn't want more than about 200 ft of conductors to the line side of the meter, but you can do whatever you want after the meter.

 

[winnie]

Looked at it one way of assuming 20,000 kWhrs at 160 A, comes to line loss of 520 hours at 160 A = 1064 kW-hrs line loss (using 80 mohm line)
20 MW-hrs over 8759 hours is average of 2.3 kW; about 10 A, or 64 kW-hrs line loss

Your calculation here demonstrates the quadratic scaling which Wayne mentioned in post #2. You are comparing the delivery of 20,000 kWh given the possible extremes of the situation, either slamming the 20 MWh at 160A in only 520 hours, or spreading it out perfectly evenly over the entire year.

With the same total energy delivered, losses are linear with the current. (Not quadratic, if you double current then losses per unit time scale by 4x (quadratic), but duration of operation scales by 1/2, net result is that delivering the same energy at 2x the rate means 2x the losses.)

The only thing you can say with certainty is that the losses are between the minimum and maximum value. But as a guess based on the load variability in my own home I'd say losses of 3-6x the minimum will be in the ballpark.

The big issue with the long run will be transient losses when large loads kick in. Not because of the cost of electricity, but because of things like dimming lights and systems resetting on low voltage. If you elect to use the straight 240V run, then make sure that all large loads are designed to ramp up slowly, possibly with things like VFD control. You don't want a 30A well pump taking the voltage down 20% for moment when it cycles on, or every light flickering like mad when someone boils a pot of water.

-Jon