array with classic and high voltage drop

[cardamon]

Hi All,
I am planning my system expansion for next summer and I have a plan that I wanted to run by the group as it is a bit unconventional, and some reassurance that my math and theory is sound and Im not missing another option.  I am in the woods right at the base of a 350' hill so getting a clear view of the sky for most of the day is tough.  This time of year I dont see the sun till 10:30 or 11:00.  Of course those are not premium solar hours anyway, but the same thing happens at the other side of the day due to trees.  Ive got a 45' utility pole that I thought about mounting them on and that would improve the situation but that pesky mountain is still there (actually the 10:30 figure is with them mounted on the pole).  So here is my idea.  Its 1300 feet to the top of the mountain, the wind turbine is already up there.  I could mount the pole and the panels up there and have the best sun possible.  I want to get about 2500 to 3000 watts out of the deal which can be done with 1 classic 150  if the input voltage is around 90 (I could get 2700 watts).  If I get 12 300 watt modules and connect them with 4 strings of 3 in series, that works out to 32 amps at 113 volts MPP.  If I use #2 aluminum I get about a 21.6%voltage drop so the classic would see about 89 volts which is right where it needs to be to max it out at 2700 watts.  Obviously I am throwing away some power to voltage drop but I think it is worth it because: 1) modules are cheaper than wire 2) I investigated using 2 classic 250's to crank up the transmission voltage as much as possible but its pretty much a wash buying a few less panels but buying a second controller, PLUS i like the idea of having more excess module wattage because it would give me more output on cloudy days.  Xantrex makes a 600 volt input charge controller which would cut my wire size down but it ends up being a wash because its real pricey plus is only good for about 2200 watts (my system is 24 volts).  My main question is if the classic will behave the way I am thinking it will with the high voltage drop and give me the power and voltage figures I have come up?

[Mary B]

calculator I use is showing a much lower voltage drop http://www.calculator.net/voltage-drop-calculator.html?material=aluminum&wiresize=0.5127&voltage=113&phase=dc&noofconductor=1&distance=350&distanceunit=feet&amperes=32&x=45&y=16 ran it thru another and came up close to the same.

[cardamon]

calculator I use is showing a much lower voltage drop http://www.calculator.net/voltage-drop-calculator.html?material=aluminum&wiresize=0.5127&voltage=113&phase=dc&noofconductor=1&distance=350&distanceunit=feet&amperes=32&x=45&y=16 ran it thru another and came up close to the same.

Thanks for the reply and checking me.  It looks like you used 350 feet instead of 1300?

[birdhouse]

wow-
that's quite the predicament.  morningstar also makes a 600Voc charge controller now (60A), but it's only for 48v banks.  ~$1300.00!

wow-  still thinking.  it's an odd time when panels are cheaper than wire (in your particular case). 

i like your plan of using #2 ALU and accepting the highesh voltage drop.  have you thought about mounting a few of the panels at your place, and the rest up the hill?  i understand the ones at your place only get sun for a short period, but they wouldn't have the voltage drop issue. 

as time/money permits, you could parallel wires runs (add a #12 CU here and there) and eventually get all the panels at the top of the mountain?? 

if you were to use a 2" conduit for pos and another for the neg, you'd have plenty of room to add conductors.  if you were to put in a few junction boxes along the way at less that 500' increments, you could easily add 500' spools as you find them on the cheap. 

i just recently ran across a #10 THHN CU roll of 500' for $32.  it had a broken spool and would require some untangling.  if you were to run across that same deal SIX times, you could drop your voltage drop quite a bit by running them alongside your #2 ALU. 

if you've got the money, i'd say go ahead with your plan.  my best guess is that the classic would react as though you had a lower Voc than actual due to the V drop, and treat the array as such.  you can always make it better over time with parallel runs. 

the only other option i can think of is using transformers, but that's over my head. 

adam

[birdhouse]

forgot to ask, how does your wind turbine do at 1300' away??


adam

[halfcrazy]

Yes I agree at the cost of PV today it sure does change the outlook on installations. If this was mine I would likely do something similar. I would however go with say 2/0 aluminum trailer cable maybe depending on the price difference in the cable. In reality if a single 250 watt panel makes up for the drop than that is a pretty cheap fix.

The 600 volt controller would help but again with the cost of the controller and all the associated over current protection and disconnecting devices would run the cost up above the cost of a couple extra PV modules. Sure makes designing interesting doesn't it.

$1300 for a morning star vs $600 for a Classic now add say another couple hundred bucks for 600 volt rated over current and disconnecting devices and we have a differential of $900 that buys more than a 1KW of pv.....

And the Morning star is 60 amps vs the Classic being 90 amps so if you figure you would need two Morning Stars it gets crazy quick..

Ryan

[southpaw]

What are the hours of sun today at the top of the pole and the top of the mountain?
What figure are you using for cost of 1300 ft of two conductor aluminum wire?
What is your cost/watt for panels?
Hard for me to form an opinion without these numbers.

[birdhouse]

$1300 for a morning star vs $600 for a Classic

$1300 is for a 600Voc morningstar.  they also do a 150Voc 60A unit for $500. 

600VDC is getting to the scary point!  not something you want to accidental moments with!

adam

[Mary B]

Oops I saw 350 foot hill and used that

[hydrosun]

Remember the voltage drop is only for maximum output. The rest of the time the loss is less. So in the morning and evening and slight overcast the amperage will be less. So you don't lose the calculated percentage on average. If you point part of the array in a different direction you will never hit that max peak but the output will be spread out over more hours of the day. It's also possible you might be able to put more panels on the same controller that way. You wouldn't get the maximum production from each panel but it would be designed to move the most power down your wire. If you already have wires in place for your wind generator on top of the hill you might be able to use the negative leg for the solar panels also. Or at least parralel the negative wires as long as the amperage isn't to much for the existing wire.
Chris

[cardamon]

Thanks for the responses and ideas.  To answer some of the questions:  #2 aluminum is really cheap because it is a very common size (100 amp dwelling unit services)  I can get it for about .33 a foot (one conductor, but a little trick from the electrician here, often URD which is 3 or 4 conductor with reduced neutral USE and RHH twisted together is often cheaper than buying single conductor XHHW)  Big orange sells 1000' spool of 2-2-2-4 URD for a grand so there is my two 1500 runs and keep the 4 for something else.  For comparison, #6 URD which would be suitable for the 600v charge controller is only $300 less for the 1000' spool.  I havent really tried to weigh out the energy production comparison for going up on the hill vs near the house.  I dont really know how to do that.  Can anyone point me to some charts that would break down energy production say per each hour of the day?  I could get more accurate figures but on the hill I would basically get sunrise to sunset and down here I would get from 10:30 to 2:30 ish The increase in cost for going up on the hill would be the extra wire, and the extra pv that I throw away to VD, but of course that still helps me on cloudy days or low production times like hydrosun said, so its far from a waste.  I dont figure in the labor of running the wire down because I need to go back and  bury the wind turbine line anyway as rodents chew on it and short it out sometimes (its 1kw sent down at 240 three phase on #10 UF.  as far as panels go I would probably buy a pallet from sun electronics so it would be .80 to .90 a watt

[southpaw]

Hold on to your hats boys this is going to be a wild ride.
I found a graoh in a Scientific American article showing panel output in % of panel rating vs. time of day. [ Specified attachment is not available ] [ Specified attachment is not available ]
This is for a 12 hr. day where the sun strikes the panel at 0 deg. at sunrise, perpendicular at solar noon and again 0 deg. at sunset. Because the sun will strike your panels at an angle greater than 0 deg. during any other length of day.
 I have used this graph to show the panel output for your location with 10:30 sunrise and 2:30 sunset. The black area represents the power your panels will gather during this period. [ Specified attachment is not available ]
This picture represents the power your panel would gather on top of the mountain, assuming 8 hours of sun. (I have no idea how many hours you actually get as you haven't provided this or your location)
[ Specified attachment is not available ]
Again the shaded area represents power gathered, in this case 8am to 4pm.
The maximum power gathered in all cases is about 70% of the panel rating, this take into consideration many variables
but is close to real world values.
Picture 1 shows an average of 68% of panel rating for 4 hrs. this times your panel rating of 3600 watts
gives you 9792 whrs of production in the valley.
Picture 2 shows an average of about 55% of panel rating for 8 hrs. giving you 44%hrs. this times 3600 watts
 gives you 15840 whrs. of production on the mountain top.
Now taking into consideration voltage drop in the 2 cases using 100 ft of 2ga. aluminum in the valley and 1300ft on the mountain top and because  voltage drop is linear with both amperage and conductor length I used an average of 68% of 32 amps for the valley and an average of 55% of 32 amps on the mountain top
 the calculator shows a .89% drop for the valley so 99.11% of 9792 watt hrs would give you 9705 watthrs.
The calculator shows a 9.4% drop on the mountain top so 90.6 % of 15840 whrs would give you 14350 watthrs.
This shows about 50% increase in output by moving the panels to the mountain top.
Moving the panels would cost $1000 more for wire.
Adding 1800 watts of panels in the valley to get the equivalent output would cost 1800w X $.85/w =$1530 plus an extra or larger charge controller.
There are hundreds of variables which would change these numbers but this gives you a rough idea of  what you can expect.

[Frank S]

 I have read horror stories about partial string shading. I don't know if these are true or have any real world basis.
 This has me wondering of the possibility of burnout in panels when a string has part of the panels shaded during the day
 I realize you would have shading from the trees in the early hours and the late afternoon hours when the sloar strength is less but would this possibly be a factor to consider?
 Just asking

[southpaw]

I have read horror stories about partial string shading. I don't know if these are true or have any real world basis.
 This has me wondering of the possibility of burnout in panels when a string has part of the panels shaded during the day
 I realize you would have shading from the trees in the early hours and the late afternoon hours when the sloar strength is less but would this possibly be a factor to consider?
 Just asking

My opinion for what it is worth is,  partial string shading causes the power output of the string to fall reducing total module output. Panel burnout is almost always caused by weak points in the strings, usually defects from manufacturing or weathering corrosion.  Panel shading from nearby objects will move across an array as the sun moves across the sky.
If partial string shading caused burnouts eventually all the panels would fail as the shadow moved across the strings.
All of my panels have partial string shading at one time or another, continuous spot shading from leaves, bird crap and snow
 and traveling shade from trees, flagpole and deck railings etc. and have not had any damage in 5 years of continuous use.
 Apparently partial string shading can cause burnout but I have never seen it and don't know how anyone could attribute
it to anything other than defects. Burnouts are extremely rare and usually occur in dump load systems where panels are
almost always at maximum possible current output.
My 2 cents (Canadian)
Kurt   

[cardamon]

Wow, thanks Southpaw for the detailed analysis.  It seems that going on the mountain is the way to go.  There are two other advantages to going up there beyond what was concluded in your analysis.  First is that it is advantageous for me to have  power distributed over  a longer portion of the day because it reduces cycling my batteries thus extending their life and reducing losses.  Second it would be easier to erect this mess on the hill because I could use the existing wind tower anchors for guying - well two of them anyway. Its interesting that, averaged out over the whole day, the losses are only 10% vs 20% during peak output.  i hadnt thought about that.