question

TylerA avatar image

Micromanaging solar?

I have a new Maine Cat 38' that came with 6 solar panels. 4x140W and 2x120W. The builder arranged these panels with (2x140W+1x120W) in series with two such banks in parallel.


They are placed flat on the bridgedeck roof.


(If you're curious: The 4x140W are all arranged bow-to-stern in a line side-by-side, and the 2x120W are arranged laterally amidships. I *believe* the three panels to port are in series together, and the three panels to starboard form the second series bank, the two banks combined in parallel. )


While I'm docked or at anchor, I keep the boom oriented for minimal shading (usually to one side, and maybe for 1-3 hours straight down the middle).


I've noticed things like moving the boom to remove shading from a 120W panel results in more than a 120W increase in wattage.


I've *heard* that mixing panels (wattage) can result in limiting the utility of the bigger panels, but at the moment, I'm playing the hand I'm dealt.


Points:

- At anchor, I have some control over shading. When the sun is positioned favorably, all 6 panels are in the sun (there might be slight/small shadows from the boat's rigging). Most of the day I can get 5 in the sun (with a 120W panel all or partially shaded).


- At sail, I have minimal control over the shading (aside from choosing a tack based on shading, but that's usually secondary to navigational reasons for choosing a tack).



It would seem to me if all the panels could be wired separately into a single box, that box could make the decision as to the best way to combine the panels for peak voltage. At present, however, the panels are wired via terminal blocks with a single pair of wires coming into the MPPT.


The micro-manager in me wonders:

Would it make sense to have the wires come into a bank of switches such that I could arrange the panels based on present need into the best arrangement of series/parallel/off (per-panel) before the wires enter the MPPT? (Leaving them in a default configuration when I leave the boat or do not feel like micro-managing)


The panels run into a Victron 150/35 MPPT. The output is a 48V Lithium Victron battery bank (comprised of 2x24V in series).


Is there anything I can do to improve what I get?


(I see 250-350W in the sun, the lower end when a single panel is covered, and closer to 140W when two panels are covered. When the sun is overhead and the panels appear to be lit, the wattage always seems disappointing.)





MPPT - Solar Charge ControllerSolar Panel
10 |3000 characters needed characters left characters exceeded

Up to 8 attachments (including images) can be used with a maximum of 190.8 MiB each and 286.6 MiB total.

2 Answers
Christyler avatar image
Christyler answered ·

As previously said, it would be hard to sort this with what you have. Yes your idea of switching panels on and off depending on shading would work, but do you really want to be pre occupied monitoring this and manually operating it ?

As far as I can see the way to manage this is to go with separate mppt controllers for each Pv panel, but that would require using high voltage Pv panels,maybe a good time to make them all the same as well. Or maybe put you battries in parallel and use a 24v invertor, and then use separate controllers on the panels you have

Share
10 |3000 characters needed characters left characters exceeded

Up to 8 attachments (including images) can be used with a maximum of 190.8 MiB each and 286.6 MiB total.

JohnC avatar image
JohnC answered ·

Hi Tyler. The main hurdle you're up against here is the 48V batts. Fine in themselves, but the panels must be strung at a V higher than that.

To allow for less than 'lab test conditions', with low morn & afternoon sun, cloud, boom, and even rigging wire (yes even that), your panels should be strung quite high in total V. Like maybe 100Voc or even more.

You don't mention panel specs, and if they're 'nailed down' might be hard to find unless the builder can help out. If you could measure the input V to that mppt (via Victron bluetooth or multimeter) it might confirm where your panels fit in there. Voc (that's open-circuit), is the figure to look for there, so maybe isolate the batts while testing.

There's even a fair chance your panels are '12V' models, with a Voc of say 22V, and in that case all 6x may be wired in a single string.

There may actually be little you can do to make improvements either way. Shade & short winter days can make life awful tough for we solar users.

5 comments Share
10 |3000 characters needed characters left characters exceeded

Up to 8 attachments (including images) can be used with a maximum of 190.8 MiB each and 286.6 MiB total.

I'm getting ~13h/day of solar (currently in Boston), and the voltage is 60+ (pulling a reading from now with expectedly low wattage and 7:30am sun), so I'm not concerned that I'm unable to feed the battery. I do acknowledge that shading and sun angle are the main factors in solar output. I manipulate the shade to the extent that I can, what I'm asking about is manipulating the solar configuration.


The CCGX reports that voltage range over the past week to be typically (Vmax) 60-77V, although there are a couple days where Vmax is over 100V (not sure if that's just an instantaneous reading/outlier).


My understanding of connecting in series is the christmas light problem with regards to shading. I'd like to find authoritative information on that.


If the voltage fluctuates (and I suppose the current does, too) with changes in shading/sun/etc, then what happens when taking two banks of panels that would be at two different voltages (again, due to shading based on positioning on the boat) and connecting them in parallel? Post-connection, the voltage would be the same on both sides. Is the bank with the (previously) lower voltage still contributing to the overall output? Does it weaken the whole system by being attached?


If the christmas light problem is true, I'd think the ability to micromanage (switch off individual panels that do more harm than good) would be a net win. (Improvements while I'm on the boat, but same behavior as before if I leave the boat in some default/reasonable configuration).


Thoughts?



Hi Tyler,

John is probably right about your panels being all in series.

Something useful to look at, https://www.pveducation.org/pvcdrom/modules-and-arrays/shading well worth having a good look at the site.

Good you have a CCGX, and I'd encourage you to link it to VRM and see the graphs. They will help.

I can assure you the 'christmas light' issue is for real. There's heaps about it on Google, and I experience it daily when a 25mm handrail shadow hits the end of 3 panel strings concurrently. Within 10 min I lose 70-80%. Some modern panels may be wired differently internally, but with mine I get the lot at once. Don't wait for my white-paper. :)

I did see a guy post somewhere once that this issue can be 'fixed' with external diodes (as distinct from the ones within the panel j-box), to bypass the shaded panel(s). Might work, dunno.

Then there's panel optimizers like Tigo you could look at too.

Your 60V early morning is typical of mine, but it winds up to like 100V+ with mppt in full swing, then settles at ~120V in Abs/Float (Voc). The more info you have on your own system though, the better chance of working something out.

For me the 100V+ is a rarity (once a week?), which is why I think it might be a sampling error. My peaks are usually in the 70-80V range.


If I just put all the panels in series, then a single panel shaded ruins the string, no?

Well yes, but you have to balance that against getting nothing at all if your string V isn't enough to kick away the mppt. Your panel Voc must be 5V more than Vbat for it to work at all..

And you must never exceed 150V in a series string, else the mppt is toast. Take care Pal.