Voltage Loss - Impact of >3%

mjp24coho wrote: »

I'm working on some voltage loss calcs for a new array I'm contemplating, and the distance will be quite large. I'm looking to do 6 panels in series for ~220v, and then two series in parallel. The calculated total system voltage loss would be 3.75% with 6 AWG THHN wire.

What are the panel specs, how far is the distance, and what controller will you use? You must have some cold winters in the Utah mountains... if your Vmp is 220 volts, the cold temp Voc will be too high for most controllers.

As Zoneblue points out, its all just a matter of how many watts you lose in the cable.... and it may be more cost effective to buy an extra panel than to invest in more copper.

You might even be better off configuring the panels four in series per string (three strings in parallel)... sure your cable losses will be higher, but your controller may be much more efficient (what controller are you planning to use?).

By the way, the wattage losses you calculate go as the square of the current in the cable, so at typical power levels your losses are much less than what you calculate at full power production.

vtMaps

mjp24coho wrote: »

400 feet each way with 8 awg wire (not 6 as I misstated above), and Midnite Classic CCs.

Doing what you propose will be a disaster with any Midnite controller.

First of all is the issue of cold Voc. Your Voc at standard conditions is 225 volts. On a 24 volt system that may damage a Midnite Classic 200 without even taking into consideration cold temperatures. Even with a 48 volt system on a Classic 250 you are pushing the envelope in cold weather.

Secondly, for a 24 volt system you are over the power limits of either a Classic 200 or a Classic 250. Midnite's power charts for a Classic 200 only go up to a PV input voltage of 160 volts. At that voltage (160 volts) the maximum capacity of a Classic 200 is 2066 watts (and even less for a Classic 250).

At higher voltages the controller is less efficient. That means that it makes more heat. Heat is the enemy of electronics. To run the controller at maximum capacity AND maximum inefficiency will shorten the life of the controller. Of course, with your proposed configuration the controller will probably die first from exceeding hyperVoc.

Here is a quote from Robin Gudgel (cofounder of both Outback and Midnite):

Max power for the Classic 250 is somewhere between 3000 and 3500 watts depending on your system configuration.
<snip>
I would not put more than 3000 watts on the Classic 250 myself. It will just run hot. That is never a good idea to run the controller flat out at its maximum day after day.

In this quote Mr Gudgel was discussing a 48 volt system on a Classic 250 which has a maximum capacity of 3212 watts with an input voltage of 180 volts. The point is that it not good design to run a system at max power continuously.

--vtMaps



Edit: you might want to read this also:
http://midnitesolar.com/Forum/index.php?topic=2008.msg18796#msg18796

mjp24coho wrote: »

I'm working on some voltage loss calcs for a new array I'm contemplating, and the distance will be quite large. I'm looking to do 6 panels in series for ~220v, and then two series in parallel. The calculated total system voltage loss would be 3.75% with 6 AWG THHN wire. I understand that is over the magic 3% rule, but I wanted to understand what the impact is of being over 3%. What am I risking/sacrificing by being between 3-4%?

I will offer some opinions and suggestions:

1. Use aluminum conductors
2. As other said, consider the watts lost in PV cost - I have an array with 10 percent loss but adding a panel was cheaper than adding more wire.
3. I think you are probably fine with 2 classic 250's
4. take a look at the schneider mppt-600-80 CC. Costs the same as two classics. The schneider will only give you about 2300 watts but your 3000 watt array will work pretty well for that consideringwith losses, less than perfect conditions, mismatch panels, VD, etc. you will be around low to mid 2000's often anyway. You can usually do 13 - 250 Watt panels all in series with that controller - I would load it all the way,series is a free lunch in terms of losses and more output on cloudy days.

I have a 980' run to my water system and used #6 aluminum wire. It is light and pulled real easy. You can order spools with the footage you need, so there is no need to splice. Just be sure the gear has aluminum rated connectors (nearly everything does)

Re Array Voltage. Schneider or Morningstar 600V charge controller. { I think this is the way I would go, and I'd use the Morningstar fanless & nice ethernet 16V min turn on voltage, can be programed via computer and MSView software for 24V bank}
http://www.morningstarcorp.com/wp-content/uploads/2014/02/600V-TS-MPPT-Operators-Manual.pdf page 35 lists switch 3 as being able to config with MSView for 24V

Or I suggest you re-think the array. 250V into a 24V battery will be a heavy loading on the classic. My 170V array on a Classic 200 @48V, sits and cooks all day long and you are considering an even worse PV-Bat volt mismatch, Consider 2 or 3 panels in series, #4 alum wire and combiner box.

I have been happy with the online calculator from Southwire, at http://www.southwire.com/support/vol...calculator.htm
It take a little getting used to, and you have to choose between inputting wire size and %VD and getting run length or %VD and length and getting min wire size.
It does not give you %VD for a particular wire size and length AFAIK. But it is easy to work that out yourself from the output it gives.

Aluminum will need to be between 2 and 3 AWG numbers larger than copper for the same VD.

The rule for commercial solar is design at 2% loss. A good article this month in solar pro talks about the trade offs and costing for large projects to bend the rule upwards. They basically say the cost of wire on the solar side of the controller is cheaper to size correctly and skimp on the output side of the controller.

That does not really relate to you as you should just buy the Schneider hi voltage controller as the Morningstar will not work at 24VDC. XW-80 will give you room to expand later and its cost is the price you pay to run an array that far.

Mike, I was hoping someone like you might come back and correct me. The older 600vdc Mstar did not have a 24vdc option. So this new model does? --Dave

mjp24coho wrote: »

As mentioned above, I plan for this to be a 48V system, not 24V. I'm a big fan of Midnite's CCs, so I plan to use those instead of a high voltage Schneider.

I really think in this situation, one of the 600V CC's is a much more logical choice.

mike95490 wrote: »

Well, then you are all set. The Morningstar View SW will let you tweak the settings beyond what the dip switch can select.

And when the user adds more solar it is really easy to change wiring at the array with a 600 volt controller. With low voltage controllers adding solar will just add more current and a higher percentage of loss on long distance runs. The high voltage wiring is simple with most applications just needing a junction box at the array and a disconnect outside the house/garage.