String length for boat installation

[mirage]

I am installing 8 Panasonic 240 watt panels feeding two Midnight Classic 150's on my boat. The panels are rated for 5.5 max amps at 43.7 max volts. Space is a premium, so I don't have much flexibility in where the equipment is located. The controllers are near the middle. Five panels are located a 40 ft wire run forward of the controllers and three are located a 25 ft run aft of the controllers. I plan on running two strings of two panels in series from the forward array to controller A. For controller B, I plan on running one string of two panels in series, both from the aft array, and one string of two panels in series, one from the forward array and one from the aft array. I was told that Midnight Classics don't like different lengths of strings. Does that mean that I need to lengthen the short leg of the odd string (the one with a panel forward and aft) by 15 ft to match the length of the longest leg. Also, does that mean that the other string coming from the aft array (the one with both coming from the aft array) needs to be lengthened by 15 ft. I haven't bought the wire yet, but I was planning on using 8 ga which puts the voltage drop at about .3 for the 80 ft round trip run.

[Westbranch]

For the MPPT function to work properly the output of the 2 strings should be very close. with the one array of 2 panels so far apart the output of these 2 strings will be too dissimilar for MPPT in the classic to give you its best performance...  I would look into how you can use  different panels on the area where you only have 3 planned and drop the 5 th panel on the other. 

Also is there a chance for /of shading on any of the string?  If so your output will be severely diminished...

[zoneblue]

while two classics seems like overkill for 8 panels, on a boat there are challenges for sure.
for mppt, kids might be better maybe.

personally i would have thought that wire length wasnt a major factor for mppt, so long as the voltage drops werent excessive, and the panels were identical. when in doubt using bigger wire. however i agree with wb, that on a boat, the less the number of panels in each string the better. in some ways what id be looking at is strings of one, with 36/72 cell panels with several small pwm controllers. however strings of one reduce the value of mppt somewhat.

however something is odd about your pv specs. a 240w panel usually has a vmp between 30-32v, imp around 8a.  such a panel can only work with strings of one with mmpt at 12v system voltage, something better off avoided on a 2kw rig,

[mirage]

Are you saying that the input will be too dissimilar even if I take care in making the lengths of the strings the same? I could bring all three of the panels from the aft array to controller B, but that would mean the fifth panel forward would need it's own controller on a single string. There is no way to increase the number of panels in the aft array. However there is plenty of room for the fifth panel in the forward array. I already have the panels, so it's a bit late to rearrange the configuration. Shading shouldn't be to much of an issue. It's not a sailboat, so I can control most of the shading issues.

They are Panasonic HIT 240S panels. The data sheet calls for 43.7 vpm and 5.5 lpm. If I must use the fifth panel in the forward array in a single string, what would be the best controller for such an application?

[Westbranch]

I just reread your description of the arrays. 

Panels in a Series arrays are connected using the supplied wiring, probably an MC4 connector (M/F)

The difference between the 1 Fwd, 1 Aft array and the 2 aft array is about 60 feet, not 15 as you estimated. 

Reason is that the 2 Aft are going to be connected with about 3 -4 feet of wire then the 15 ft, 2 wire run to the controller.

The 1A/1F array need the 2 panels connected (60ft) and a 45ft. run and a 15 ft run to the CC. 
Hope this makes sense.

[mirage]

Westbranch

My thought was to connect the 1A/1F array in series by running two 40 ft wires from each back to a junction box at controller B where the connection to make them in series could be made. Granted, this is wired differently than the 2 array aft, but the end result is the same and the wire length is the same. Maybe this is more complicated than it needs to be. Why can't I run four single strings to controller B each with a length equal to the longest run, 40 ft. What difference does it make which direction the wires run comes from? Other than using an extra 80 ft of wire, would this be a more reliable way to optimize the performance of the controller?

[zoneblue]

They are Panasonic HIT 240S panels. The data sheet calls for 43.7 vpm and 5.5 lpm. If I must use the fifth panel in the forward array in a single string, what would be the best controller for such an application?

Ok, unusual panel. Starting to see a few such variants around. Im still thinking strings of one gets around your odd number issue.  A string vmp of 43 volts is plenty to charge a 24V bank using mppt. A single classic can handle all that just fine. Unless you are aiming at a 12v bank, which i already suggested was a bad idea, but if you did it would be two classics both running hotter than they need to. Run the two groups of panels into mnpv6 combiners tehn make two home runs. All good.

[Westbranch]

near the bottom of this page there is a voltage loss calculator that will help you see what losses there are for varying wire lengths
http://www.powerstream.com/Wire_Size.htm
hth

[mike90045]

If you are charging 12V battery system, I would simply parallel all the panels and use heavy gauge wire.  If you series them, you become dependent on the the panels internal bypass diodes managing the shade, and dissipating the heat involved.  That's the job they are supposed to do, but it's like driving a car with your foot resting on the clutch, it works for a while, but it's not best for the clutch lifetime.

[Robin]

I think this system is getting overly complicated.
1. I do not think it is a good idea to be putting one front and one rear panel in series. The wire lengths discussed here have almost no effect, but the chance of shading on one or the other panel is great.
You will have a max power voltage of 34 volts or so with only one panel. That is sufficient to charge a 12 or a 24V battery. Put all eight panels in parallel. Use a MNPV12 combiner or something similar. This way when one panel gets some shading, it will barely affect the max power point voltage. Two panels in series on a boat is asking for shading issues.