Best way to run AC power 500 ft from array to cabin

[AlaskanNoob]

So serious question. I am not the original poster.

I would like to put, say 5000w of solar panels 500 ft away from where I am currently at, on the other side of the mountain. It is exactly 500ft because my water line is 400ft and its flat another 100ft away. Let's say a 450V MPPT charge controller... with standard two parallel strings. One facing for winter another facing for summer. 400w 49VOC panels, let's make it simple 2p9s or 2p8s or whatever

Above ground conduit. Opening every 100ft. so I can drive my excavator around later or drag a tree over it at some point.

AC, or DC?
DC direct from the charge controller? would that work? To a battery bank.
Or would AC at 240v stepped up to 480v (?) and stepped down be my only option?

Some of you are just intent to entertain yourselves at the expense of the discussion here. Here come the bazillion posts about DC power that have nothing to do with this thread to pile on. So now as we try to figure out our system, we'll have more irrelevancies to confuse things. Well, it's just electricity. What's the worst that can happen. So long as you have a good time thread-shitting.

 

[AlaskanNoob]

You put the water heater right next to the house. Pump -> feed -> solar water heater -> electric water heater -> sink/shower. There's no electricity involved, it's just a passive heat absorbing system.

Yeah, yeah. Enjoy yourselves. I'm out.

 

[12VoltInstalls]

you're not demonstrating that you're a face value kind of guy to me in this discussion.

I guess ‘options’ is subjective; who knew?!
I’m not normally like this but the infantile expression is too tempting to not bait…

 

[Rednecktek]

Yeah, yeah. Enjoy yourselves. I'm out.

Sorry, I was still catching up on the 6 pages that showed up since last watch.

Yes, if you're looking to move high voltage AC a long distance, your choices are thick wire or honkin-huge transformers that can convert all 3 legs (hot, hot, neutral) to a higher voltage and then another one at the other end to kick it back down. The big question there being is the expense of 2 honkin transformers AND 500+ feet of smaller copper wire rated for over 600v saving you anything over just running really thick aluminum standard supply wire.

 

[Warpspeed]

So serious question. I am not the original poster.

I would like to put, say 5000w of solar panels 500 ft away from where I am currently at, on the other side of the mountain. It is exactly 500ft because my water line is 400ft and its flat another 100ft away. Let's say a 450V MPPT charge controller... with standard two parallel strings. One facing for winter another facing for summer. 400w 49VOC panels, let's make it simple 2p9s or 2p8s or whatever

Above ground conduit. Opening every 100ft. so I can drive my excavator around later or drag a tree over it at some point.

AC, or DC?
DC direct from the charge controller? would that work? To a battery bank.
Or would AC at 240v stepped up to 480v (?) and stepped down be my only option?

Edit: with this voltage drop calculator https://www.rapidtables.com/calc/wire/voltage-drop-calculator.html
I put in 12 AWG like romex I guess and got only 3% drop.
But what about sunrise and sunset when the voltage goes from 0v to 450v? Is there going to be a danger

Remember, that 3% is only at flat out full power with a clear blue sky.
When its cloudy in winter the available solar power will be less, and the losses much less too.
Another thing to think about, if you add 3% extra panels, your EFFECTIVE losses become zero.
So with a system like that you can afford to run much greater cable losses than 3% and save considerably on copper costs.

Where the magical 3% comes in, is when that 3% voltage drop is between the inverter and the load. You switch on a big load, and the lights dim down a bit. That is completely unacceptable. So the wiring rules in many countries say 3% drop max is what you design long wiring runs for.
For raw solar dc power you could run with a much larger voltage drop, and just add a few extra panels to get that lost power back.

The original poster has the inverter at the solar panel end, and that is completely different problem where voltage drop then becomes a very serious issue.

Work out the economics of installing a few extra panels against running a thinner cable. As long as the cable is not run beyond its nominal current rating that is quite o/k. Put the charge controller at the battery end of the long cable, and a proper mppt charge controller will work perfectly well charging your battery.

 

[noenegdod]

Remember, that 3% is only at flat out full power with a clear blue sky.
When its cloudy in winter the available solar power will be less, and the losses much less too.
Another thing to think about, if you add 3% extra panels, your EFFECTIVE losses become zero.
So with a system like that you can afford to run much greater cable losses than 3% and save considerably on copper costs.

Where the magical 3% comes in, is when that 3% voltage drop is between the inverter and the load. You switch on a big load, and the lights dim down a bit. That is completely unacceptable. So the wiring rules in many countries say 3% drop max is what you design long wiring runs for.
For raw solar dc power you could run with a much larger voltage drop, and just add a few extra panels to get that lost power back.

The original poster has the inverter at the solar panel end, and that is completely different problem where voltage drop then becomes a very serious issue.

Work out the economics of installing a few extra panels against running a thinner cable. As long as the cable is not run beyond its nominal current rating that is quite o/k. Put the charge controller at the battery end of the long cable, and a proper mppt charge controller will work perfectly well charging your battery.

As intended, the losses are when he is drawing full power, not producing full power. The intention as presented here is to have the batteries and inverter 500' from the location of the consumers and send the required AC power through the 500' conductor.

Be careful posting a suggestion regarding sending DC power over the 500'. The OP is a little sensitive about that subject being in this thread.

 

[offgriddave]

Yeah sorry I am not the original poster, him and his wife seemed really upset about various suggestions made here. I am sure they are just nervous about purchasing a bunch of stuff and having it not work. OP's in alaska I'm in oregon

I guess I am guilty of hijacking a thread by asking about going 500ft with DC

So I have about 500ft of ROMEX 12 awg the yellow stuff. Not rated for outdoor. On the side of it is says maximum voltage 600v. I calculated 450v at 11 amp is about 5000w.

Given I hook up 9 49v panels and get 441 volts. I run a 500 ft of romex. Then, on the other side of the hill, to a MPPT charge controller. And then that to a 48v battery and inverter.

Looks like Victron, Growatt and Morning Star make a 450v and 600v MPPT.

That would be safe?

What about in the sunrise, sunset and cloudy days when the panels aren't getting enough voltage? Will the voltage drop cause issues with the wire? Would it cause a fire? The lowest I've seen these 400w panels go to is I think 17v. 9 * 17v = 153v. The voltage drop on 153v is a lot vs 450v. But conversely, the amperage would be lower? No? Still dangerous? Would that be a problem?

This seems to me very dangerous... but I can't figure out why?

Maybe the OP can see this design is superior cost wise and just do this?

 

[justgary]

Yeah, yeah. Enjoy yourselves. I'm out.

Before you leave, will you answer a few honest questions?

1) Have you ever used a WSHP before?
2) Have you ever used a WSHP with a desuperheater for your hot water before?
3) Is it possible to move the cabin to the location that has better solar availability?

 

[noenegdod]

Yeah sorry I am not the original poster, him and his wife seemed really upset about various suggestions made here. I am sure they are just nervous about purchasing a bunch of stuff and having it not work. OP's in alaska I'm in oregon

I guess I am guilty of hijacking a thread by asking about going 500ft with DC

So I have about 500ft of ROMEX 12 awg the yellow stuff. Not rated for outdoor. On the side of it is says maximum voltage 600v. I calculated 450v at 11 amp is about 5000w.

Given I hook up 9 49v panels and get 441 volts. I run a 500 ft of romex. Then, on the other side of the hill, to a MPPT charge controller. And then that to a 48v battery and inverter.

Looks like Victron, Growatt and Morning Star make a 450v and 600v MPPT.

That would be safe?

What about in the sunrise, sunset and cloudy days when the panels aren't getting enough voltage? Will the voltage drop cause issues with the wire? Would it cause a fire? The lowest I've seen these 400w panels go to is I think 17v. 9 * 17v = 153v. The voltage drop on 153v is a lot vs 450v. But conversely, the amperage would be lower? No? Still dangerous? Would that be a problem?

This seems to me very dangerous... but I can't figure out why?

Maybe the OP can see this design is superior cost wise and just do this?

Just coat the romex with this and its good for outdoors. Problem solved.

 

[offgriddave]

Just coat the romex with this and its good for outdoors. Problem solved.

LOL I rented out my farm one season and the mexican workers laid down 1000's ft of romex on 3 acre field from one outlet. They used duct tape (?) to connect the wires. They would splice to the side of the romex with box cutters, and duct tape another one side by side, and keep going to the next greenhouse. The polarity was changed several times. You would turn on one thing, and the row next to it would turn off. At the end of this whole thing, they ran a 1.5hp pump. They kept blowing things out. Plugs would melt to each other. No ground what so ever. And several areas you would drive over it. I had to re-do the entire thing for them with GFI and stuff it cost me several hundred dollars. I tried explaining but they didn't speak english. Anyway they left it all that's how I ended up with a truckload of this stuff.

Another time, in my 20s, I was so poor I couldn't afford 300ft of well wire. So I posted on reddit electrician asked if I could use romex. There was one dude who said and I'm serious, That I would "KILL MY FAMILY" if I did this. I printed it out and framed it called all my buddies. LOL. I actually felt bad. Never did use romex in the well used real well wire.... But one time I think I was watching home stead rescue, they pulled a pump out and it was... romex haha

 

[Hedges]

The problem is finding where the cable has failed if its buried.
If you know within a yard where the problem is, digging a big hole and doing the repair may be only slightly annoying.
That may be possible with equipment and expertise you probably do not have or can get at a reasonable cost.

So if there is voltage at one end and nothing at the other end.
What do you do next ?

With a device known as a "TDR"




I don't like the generator and inverter at PV array idea.
You live in Alaska, and have an all-electric house, including heat. Kind of like the California of the future, when gas hookups are banned (think of someone in Tahoe during a power outage, if perhaps they weren't allowed propane either.) You'll have to walk 500' uphill in the snow to restore power, can't call the utility to do it.

I'd rather have a diesel generator at the house, and battery inverter. During a storm, generator can run occasionally and inverter keeps power going 24/7. "Combined Heat and Power", only 1/3 to 1/2 of energy from fuel generates electricity, so capture and use waste heat.
If electronics fails, you still have diesel for a heater.

This is one of the benefits of using AC coupling, GT PV inverters at the array (an AC power run from house to there.)

Yes, if you're looking to move high voltage AC a long distance, your choices are thick wire or honkin-huge transformers that can convert all 3 legs (hot, hot, neutral) to a higher voltage and then another one at the other end to kick it back down. The big question there being is the expense of 2 honkin transformers AND 500+ feet of smaller copper wire rated for over 600v saving you anything over just running really thick aluminum standard supply wire.

My math and internet shopping has said yes, transformers are cheaper than larger gauge wire.
We do have to consider whether transformers have a problem with the environment. Ventilated dry transformers would have windings and laminations fairly exposed. Cheaper ones are aluminum not copper. Are they OK there? Maybe OK inside the culvert walled earth-sheltered electrical vault?


High voltage "pig" transformers have been suggested. I think 960V will be sufficient for this moderate wattage system to make wire economical.
Possible issues with 12kV transformers:
While above ground can be air isolated, wires are subject to damage. Underground will require much better insulation than +/-480Vrms.
12kV will kill you more dead that 240V. or at least, the "scared straight" videos of our training feature amputees.
12kV, if it discharges in an arc, produces X-rays.
I think these transformers are oil-filled. That has its plusses and its minuses. Improves environmental sealing and power handling. But, oil degrades especially with electrical discharges. Big transformers get their oil lab tested and changed. Maybe the little ones are "sealed for life" and recycled when their time is up. Failures can be fires.


I don't know how much danger of damage to a buried line in conduit. I've had weak glued joints come apart. Cheap sprinkler pipes in shifting earth. PVC conduit with very loose tolerances, isn't mechanically tight when I try to glue them.
Is it possible to get a flexible reel of conduit with wire already in it?
If wire was backfilled with oil (check compatibility of insulation), that would exclude water.

 

[Warpspeed]

With a device known as a "TDR"

Yes indeed.
I could discuss transmission line theory with you for hours, its a really fascinating topic.
But a genuine Time Domain Reflectometer is not something most guys can immediately lay their hand on.
These days a good oscilloscope and pulse generator might be almost as good, but not many off gridders have those either.

 

[Warpspeed]

If I were doing this personally, 15Kw ac for 500 feet, its difficult to decide between big copper or high voltage.

If the power and distance were both less, copper run in an above ground duct would probably have more appeal.
If the power or the distance were both higher, high voltage overhead woulds be a slam dunk.

I would not be having custom transformers wound, even if I had just won first prize in the lottery.
I would look for a matched pair of fairly old disreputable looking pole pigs that I could buy for their scrap value.

 

[Hedges]

Yes indeed.
I could discuss transmission line theory with you for hours, its a really fascinating topic.
But a genuine Time Domain Reflectometer is not something most guys can immediately lay their hand on.
These days a good oscilloscope and pulse generator might be almost as good, but not many off gridders have those either.

I used to do that for MCMs and PCBs.

We bought a Tek 11801B and a couple pods for about $17K around 1990 (which was more money back then). (And Cascade Microtec coplanar probes.)
Decade or more later the mainframe died. Bought a replacement on eBay for $1000 (today available for half that.) (We still had the pods.)

We were trying to resolve impedance etc. of 1" traces on a substrate. From the data I tweaked extraction/simulation models.
To locate breaks in an underground powerline, only need 3' resolution. About 6"/ns propagation in plastic dielectric, so 6ns.
What's the rise time of the 1 kHz cal signal on a scope?

You can also do TDR with VNA and frequency domain/time domain conversion. But limited by the lowest frequency used.
Some off-gridders might have a scope. VNA? Well, I'm not really off grid.

Capacitance would also provide an estimate, but effect of water on capacitance might introduce too much error. (I think prop delay seen by TDR might not change as much.)

With the people of California calling for underground power lines to avoid wildfires (not with their own money, however), some people have said underground wiring fails 1/10th as often but takes 10x as long to repair.

If you depended on a DIY underground transmission line, DIY fault location is called for.

Some kind of TDR would seem appropriate. Measure from both ends, measure both wires. This should allow correction for variation caused by soil condition's impact on EM wave propagation (assuming uniform soil along cable.)

Low tech solution might be using the soil as a resistor, producing voltage drop. Assuming the failure involves broken insulation, not just broken wire still sealed in plastic.
With AC or DC bias applied to wire, uncoil a wire on the surface and measure current from it to a probe you stick in the earth. Poke around, to where current goes highest. Pretend you're looking for an avalanche victim.
Rather than long coil of wire, you might be able to just measure "step potential", voltage across a length of earth, e.g. 2 probes 3' apart. The leaking voltage (from a single wire fed with voltage relative to a ground rod at power station) should make contour lines, like a topo map. Closer to the leak (and closer to the ground rod), topo lines are more closely spaced.

 

[Hedges]

I would look for a matched pair of fairly old disreputable looking pole pigs that I could buy for their scrap value.

Did you not read my comments about 12kV killing you deader and then X-raying your corpse?

But I agree nothing custom. Previously I was seeing modest sized transformers around $1k new and $500 used. Then some closer to $500 new.
In the 8 kW to 25 kW range, I'm thinking 480V to 960V works pretty well. Should fit within standard 600V wire (center-tapped transformer) and gets it down to fairly cheap wire, like 12 awg or 10 awg. Make it 8 awg and it supports doubling system size.

I think the 12kV pole pigs would make sense in the 1/4 MW range. (maybe a bigger transformer needed for that wattage.)
Oh, some are 4kV or 7kV, which would make pretty "soft" X-rays. But I read 13kV to 14kV is more popular with utilities today. (if I remember correctly, it is 30kV and up where the X-rays generated are more hazardous.)
If you buy old enough ones (some still in use, I read), you might get lucky enough to find one filled with genuine, quality PCB oil.

 

[Hedges]

My personal preference for a system of your configuration would be SMA Sunny Boy inverters at the PV array. Or maybe others that have no fans, completely sealed. Step up to 960V to send back to house. Sunny Island battery inverters at the house. This minimizes what parts of the system are at remote location - no batteries, nothing that requires manual intervention. The 3x or 4x Sunny Boys at the site provide redundancy (reduced output of course if one fails.)

Biggest downside is price, because to suck down 22kW of AC into batteries would take 4x Sunny Island, even though 2x would have been sufficient for powering loads from battery. (I got a liquidation deal which is why I have 4x installed.)

I'm actually running a system pretty close to what OP needs, except much smaller battery. Also no transformer step-up or down.
It has 7 strings of PV panels about 380 Vmp. They run up to 200' in 12 awg.
AC from battery inverter goes about 200' in 6 awg to house, with 70A breaker. Some day I'll pull 2 awg and use 100A or 125A breakers.

In my case, GT PV inverters are adjacent to battery inverters, maybe 20' of wire tops. I preferred to have voltage drop in PV DC, where it doesn't eat into voltage range for UL-1741 grid connection and doesn't sag voltage to AC loads.
When in off-grid mode, UL-1741 is out the window, and much wider voltage (and frequency) tolerance is allowed; that tolerates deviation by a generator.

Just having a long run of wire and transformer step-up/step-down between GT PV and battery inverter would make my system what I'm suggesting.

 

[Warpspeed]

I did some cable testing a few years ago just before retirement using just an oscilloscope and a fairly fast voltage step. This was to measure the characteristic impedance of ordinary grey insulation displacement ribbon cable (104 ohms) The application was for wideband analog RGB up to about 250 Mhz.
Simple power cable testing need not be any more sophisticated than that. The problem is not knowing the velocity factor of an unknown dialectric, and even if the exact distance can be computed, buried cables out in the wild are rarely laid dead straight. Even measuring exact distance over rough ground is going to be a bit of a challenge.
So an open circuit cable fault might be some physical distance from where you feel it may possibly be.

The last custom made transformer I had wound was 5Kva using the more expensive grain oriented silicon laminations. Even with mates rates, it still cost me $2.5K. That was for the infamous "Warpverter" a rather unique non pwm true sine wave inverter of my own design that is slowly gaining some interest.
About sixteen of these home made "Warpverters" in the 5Kw to 7.5Kw size are now in successful operation around the world.

I suppose a couple of custom 15Kva transformers might come out to around $10K. That buys a lot of copper wire.
I am told electric chairs usually run at about 2.5 to 3Kv. Anything more just creates burning and charring and a bigger eventual mess, but you are still just as dead.

All the big commercial overhead high voltage transmission lines have earth fault protection, which trips power at the source very quickly. Something like that would not be too difficult to arrange. Power lines fall down all the time, the power goes off fairly instantly as soon as one hits the ground.

 

[A.Justice]

I wonder how long it is before OP starts a new thread "Best way to increase PV voltage for long distance". ?

Also, If OP is still reading this, and you do actually get the AC transmission lines installed with transformers, you absolutely have to post pictures. I've never even heard of a system put together like you plan to do.

 

[sunshine]

I wonder how long it is before OP starts a new thread

Now, now!
Just need to consider MrAlaskanNoobs endurance during all those Alaskan winters......and there is also the freezing temperatures as well!

 

[labeeman]

My power from the power company's pig transformer 220 split phase is over 500' run on 4/0, not copper wire, and have no problems with running my 1000 sq foot house.