Panel String Question

[casaneil]

A question regarding panel strings please.

I have 6 x 200w Hyundai in a parallel string to a Midnite Classic 150. All good except some very small cable loss we are fixing. This array is OK.
The second string is into a second Classic 150. Both Classic's are networked and in Follow Me. We are on a 12v system at present with an Outback Fx 2012.

The second string is to comprise the following panels. An odd mix but the enclosed specs I am hoping are a very near match. The cable runs are around 8 to 10 meters (real distance) to the combiner using 10/12mm 7/6AWG single core cable.

From the combiner we have 25mm 3AWG cable running a further 5 meters to the Classic Breakers.

3 x Sharp 165w Mono.
Voc...43.1v.....Vmp...34.6v...Isc...5.46A...Ipm...4.77A.

4 x BP 585..85w
Voc...22v...Vmp...18v...Isc...5A...Ipm...4.7A.

We have tried them with the BP's in series for 36v and then in parallel with the 3 Sharps at 34.6v. Pretty dismal with 25 to 30% cable loss and/or a mismatch.

Or we can go 2 Sharps in series for 69.2v and then in parallel with the BP's array at 72v. Is this OK with this system? We used the Midnite sizing tool with 2 parallel and 2 series with min zero degs and max 40 degs C.

We have the cabling laid and all is in the ground from old times. An extremely mountainous location with no even levels but loads of light. We are aiming for 5 to 6kw a day produced. But tend to use around 3 to 4 at present.

Any advice would be appreciated. Or we have to rip it up and start again.

[zoneblue]

Id be trying 2 strings of 2 of the BPs, in parallel with the 3 sharps. 5 strings total.

That gives you:
- Isc  5.0- 5.46A,
- Vmp 34.6 - 36V

Not tooo bad. The bps will drag the sharps down to its Isc.

Or get a couple of kids, or flog the bps.

[casaneil]

Thanks for that and the input.

Trouble is we have tried that with our present cable infrastructure. BP 4 x 85w in series string with parallel 3 x Sharp 165. The loss is up to 30% as we said. Is our cable loss so great at 36v? The amount of copper must be dearer than PV Panels to improve?. We are using the OY Not Solar Cable Calculator. What is your opinion on cable lengths? Maybe time to renew the BP Panels and wiring as you say. Anyway our main dilemma was shall we go to 80v to the Classic and avoid our copper cable losses perhaps. Really not sure on the comparison. There are a lot of conductors in the ground and copper. Thanks.

[Westbranch]

Something is amiss, you should use a clamp meter on ALL wire looking for an 'out of range ' value or 2...  rearranging them should give comparable results.

possibly a bad panel? Loose connection, reversed on one array or??

Where are you taking the readings?

[casaneil]

So I gather that my cable lengths and type are acceptable and there is no need to up the voltage from what is used now.

I will when next on site with a clamp meter trace through the system for something missing or wrong as suggested and report back. Thanks.

[Westbranch]

Are those actual readings to determine loss or are they calculated losses?  You will not get the rated value from any panel, you need to derate to ~ 77% of name plate rating...

[zoneblue]

If i get your setup right, the cable is fine:

First leg:

http://www.calculator.net/voltage-drop-calculator.html?material=copper&wiresize=1.296&voltage=34.6&phase=dc&noofconductor=1&distance=10&distanceunit=meters&amperes=26&x=46&y=12

Home leg:
http://www.calculator.net/voltage-drop-calculator.html?material=copper&wiresize=0.2028&voltage=34.6&phase=dc&noofconductor=1&distance=5&distanceunit=meters&amperes=95&x=57&y=12

As WB said, id be more concerned with your methodology.

For panels in parrallel the key factor is Vmp. The difference is what 4%. Thats bearable. If you take one extreme, and say that the bp strings run at the sharps vmp of 34.6V, heres a iv curve similar to the bps. 1.4 volts is nothing as you can see.

But at 1600W are you sure you dont want to go up to 24V/48V. Classic 12v grow trap.

[casaneil]

This is great feedback and some new knowledge for me. I am on my way to the installation to test my methods. The big info drop for me WB is the need to derate the PV rated power to 77%. This explains our entire system and why on average the 6 x 200w Hyundai's give an average power on the Classic of 950w max with the occasional short peak above that. Assuming the array is working hard in Bulk Mppt in good sun.

What alerted me to my initial problem I started at was the 3 x Sharp 165w in parallel and on there own to the combiner and then to the second Classic were only ever showing average 350w. That is in Bulk Mppt with best sun and overhead. So, my long cable runs are working out and the system is only slightly below par. I did this test simply to learn if the Sharps were performing.

So, I will still double check with clamp meter for any issues as mentioned earlier. But taking your combined advice we are almost there on general system performance it seems. By simply factoring a 23% Array loss every time if planning a new array. And I begin to understand ZB comments and graph. 

Do I understand the comment of the 12 to 24v system 'crossover point' towards the wanted 2Kw. Is that to enable an increase in PV array voltage? Any other reasons for 24v?

Now with 5 hrs of mountains to drive to the location we have much to talk over.

[zoneblue]

Those array deratings are a guide only. Much depends on a range of system variables, like air mass, haze, ambient temp, wind, etc etc. So that figure is not fixed, its just a rule of thumb.

For system voltage, the rule of thumb would be something like 12V for systems up to 1kWp, 3kWp+ are 48V, in between can be 24V.

Thats not hard and fast you understand. But deciding factors include:

--> 80A charge controllers can take 1K,2K,4K at 12V,24V,48V resp.
--> keeping current/ampacity/cable size down, to reasonable/common/affordable sizes
--> the longest/most costly cable is generally the pv. Higher vmps needed to reduce the cable size are only efficient with a higher voltage charge controller.
--> generally the bigger the pv, the bigger the battery, and the bigger the inverter. Inverters a lot of current to provide peak support, and thats easier to deliver at higher voltages

Like i said its not hard and fast, some big systems have small inverters, some small systems have big inverters. But one thing is certain, with 12V, youve got nowhere to grow to. As youve found out.