Solar Not Keeping Up

[Datenight]

Greetings from Vero Beach, Florida! We have been traveling down the east coast since our departure from Groton, Connecticut on 20 October. I have noticed in the last few stops of two or more days that our solar is nor recharging the batteries back to 100%. This summer we would be back to 100% by about 11:00 am. I realize the days are shorter and the angle of the sun is not ideal. I checked my panels yesterday and discovered that the glass of one of my six 100 watt panels is cracked. I have two arrays of 300 watts each which go to one MPPT controller.

I have had to run the generator every evening to get the monitor to read above 75%. This morning after making coffee we left at about 8:30 with the charge monitor reading around 30%. We got back around 4:00 and the monitor shows about 40%. Attached are photos of the controller and the monitor reading at about 4:00. Not sure why we can't get to 100% as we did all summer. I will try to get the batteries to 100% with the generator tomorrow and do a load test on the house. Any suggestions would be welcome.

Thanks,

Rob

 

[Graeme Walker]

Suggest putting each three panel array on separate controllers; meaning at least three panels would be working, the single broken one is likely taking down the entire array.
Additionally we used twin 30amp DC-DC chargers from start battery to house so it could charge while underway.

 

[Delta Riverat]

I'd just disconnect one 300 watt array for now. Also think it's dragging the whole system down. Take it out of the circuit and be back to half.

 

[Datenight]

I'd just disconnect one 300 watt array for now. Also think it's dragging the whole system down. Take it out of the circuit and be back to half.

Good idea. I will have to trace the wire but should do that anyway.

Rob

 

[Delta Riverat]

If the 2 arrays are 3 in series and then parallel at the controller you can just break a series connection anywhere in the defective string and the whole string is disconnected.

 

[DavidM]

Between a failed panel and winter sun angle I think you have your cause.

Not familiar with the Epever MT-50. Is it a shunt based battery monitor? If not, I would change it out for a QWorks- about $30, a Renogy- about $80 or a Victron- about $150.

Any of these will work, just more features for more $.

David

 

[Datenight]

Thanks David and Delta. Here is a shot of the wiring at the controller. It looks to me like a single + and - come into the controller then cables take the power out. If this is the case I think I can just take what appears to be the bad panel out of the circuit. Will also be looking for someone local with expertise tomorrow.

Rob

 

[Delta Riverat]

Yes, the fat wires are the output to the batteries and the thin ones are the PV.

What I mean is with 6 panels going to one controller there are a number of possibilities;
1) I don't think it's 6 in parallel or series.
2) It could be 3 parallel in series with 3 parallel.
3) It could be 3 series and then paralleled.

If it's number 2 you can just disconnect the broken panel and be running at 5/6.
If it's number 3 than you have to take down the whole 3 panel string.

Since there is only on wire going into the controller these connection would be made elsewhere and just the output of the network brought into the controller.

 

[cafesport]

You can go online to calculate the values exactly, but i think the problem you have, in addition to a perhaps broken panel, is that we are about 3 weeks from the winter solstice.

The declination of the sun varies from from +/- 23.5 degrees due to the tilt of the earth from vertical. Today's declination angle is approximately -21 degrees and your latitude at Vero is around 27 degrees. Ideally you would want your panels to be tilted to your latitude for max yield, which would put the sun perpendicular to your panels to get max yield, but this would only occur at local noon. With the shorter winter days your output is also reduced.

Bottom line you can expect to get at most ½ of what you got on the 21st of June in Ct.

 

[mvweebles]

On the monitor display picture, shows 69.9v at 0.8A (55w production of 600w capacity). EDIT - checked a few 100w panels. The open circuit voltage is approx 19v. The OPs panels are almost certainly 3 panels in series, with the two strings paralleled (I had said they were all in series).

What time of day was this picture taken and was it clear skies? Getting 10% production means either the environmental sucked or there is a problem.

Any idea why or when your panel cracked? Since you left Connecticut in October, I would think solar production would be about the same this far south even with the late winter sun. Also, what are the voltage/amp specs of the solar controller (or what model is it - maybe this one rated 30A/100v Amazon.com Epever supposedly makes decent kit. It's been a while, but some time back I read that some of the Outback controllers were EPever under the Outback label.

If you can access the MC4 connectors beneath the panels, assuming they are in series, you can just disconnect the connectors to broken panel and connect the two active connectors therby bypassing (removing) the faulty panel and completing the series connection with two remaining panels.

EDIT - Slowgoesit makes a good point about checking batteries. If they charge normal via the generator, you can probably assume they are good. Given you have a visibly cracked panel, I would troubleshoot that first.

Peter

 

[slowgoesit]

I agree with the above regarding charge controllers. We have 8-360 watt panels, each with it's own dedicated controller. More $$, but we aren't dragged down in the event of shading, or one panel under performing due to damage, failure, etc. Personally, I would put no more than two panels on one charge controller. Our system has been in place for almost three years, and the only maintenance we do is keep the panels clean.

One item not addressed above is your batteries. I would check all your connections, and ideally do a static test (no load) and a load test on each battery separately. It is possible that you have an issue with one or more batteries, and that is preventing your system from charging properly. It would help to know:

Number, brand, AH of batteries, along with age of batteries.
Brand and output of charger used when generator is running.

Working on a friends boat, with FLA batteries, he had similar problems. Took forever to charge the bank. Didn't matter if you used generator or shore power (he didn't have solar). It turned out that 1 (one) battery in the 8 battery setup had failed. It read about 12.4v static, but when load tested using a simple (cheap) load tester, it dropped to less than 3 volts. It's a possibility.

Best of luck in figuring out the issue!

 

[mvweebles]

I agree with the above regarding charge controllers. We have 8-360 watt panels, each with it's own dedicated controller. More $$, but we aren't dragged down in the event of shading, or one panel under performing due to damage, failure, etc.

Good point about the battery condition

Personally, while your points about 1:1 ratio of controller to panel are well taken, in the absence of possible shading issues, I prefer series to keep the voltage high and decrease line loss. But a lot depends on the installation. Sailboats cannot avoid asymmetrical shading so 1:1 is almost a necessity. Powerboats are usually a bit different.

Peter

 

[Xlantic]

If you do need to run the generator to complement the solar panels it is better to do so in the morning when the batteries a less charged and accept more power.

For example, use the generator for "bulk" charging first thing in the day and use the panels for the "absorption" phase of charging after that.

 

[Datenight]

Thanks Xlantic, it seems so obvious when you point it out.

Peter, I noticed that panel was cracked this summer but we were still at full charge well before noon. No easy way to get at the back of the panels to disconnect bad ones.

I disconnected and checked each (total of 4 house) battery this morning all tested good. Then did a load test, all tested good. Thought that was a good thing.

In the photos posted you will see one of the monitor taken at about 11:30 am. Clear sunny skies. The photos of the pilot house show a shadow from the brow of the flybridge slightly shading all three panels. Turning the panels 90* would reduce the shading to one panel. I imagine that shadow is not helping. Wind has been out of the north so will be interesting to see what happens when the wind shifts later this week.

The photo of the hard top panels shows the cracked panel. It is also shaded by the radar. The other two are unshaded.

The final photo shows the wires from the panel coming out of a huge bundle. Those wires appear to come from below when I would expect them to come from above. Have not yet mustered the the will to cut the wire ties and separate the bundle to trace the solar wires.

Thank you for the help,

Rob

 

[Datenight]

Ed (cafesport),

Thanks for doing that homework. It may be with the cracked panel and the shading I am getting due to north winds this is the best I will do. I guess it will begin to improve after 21 December!

Rob

 

[H2O_Doc]

If you can get all panels in the sun, put a towel over one panel at a time, and see what the charge controller wattage is showing. If you cover one panel, and the wattage doesn't change, that's your bad panel.