Pv configuration question

Started by BlackwaterPark, December 27, 2021, 11:18:34 AM

Previous topic - Next topic

BlackwaterPark

So, I just did a big overhaul of my system, and switched from 377ah@24v FLAs to a pair of Kilovault HAB 7.5kw batteries, to make a 15kw 48v system.  I've got 8 (well 9 now, as I went out and got another, but cannot position it in current conditions, ie; winter) Piemar 310w modules:
Vmp: 33.5v
Imp: 9.27a
Voc: 40.81v

These are going through the classic150, and I had them arranged in 4 parallel banks of 2 in series.  I was told that when I switched to the 48v system, this wouldn't work well, and the best option would be too make strings of 3 in series. The absorb setpoint is 56v for these batteries. Having only 8 panels at the time, this wasn't super ideal obviously, so I went and found another panel who's specs were within 10% of my existing ones, but as the ground is frozen, it would be very difficult to install that ninth module securely.  I've been experimenting with running them as 2 banks of 3, and now going back to my original configurations of 4 banks of 2, and I'm finding, at least today, that they are charging better with the old setup.  Under this setup, I've got about 65v coming in (little hazy for a winter day, add the bank voltage is at 53.5 and slowly climbing.  When I had it as 2 strings of 3, the bank voltage dropped to 53.3 and seemed to be hovering around that number. The batteries read about 65% soc (though I'm not certain how accurate that really is).  I feel like maybe I'll missing something here then, because I was told by someone at Midnite, can't remember who, and also from rep at AltE, that the incoming voltage running 4x2 would not be sufficient to adequately charge the bank.  Is this really to say that it will just be impossible too charge the bank up fully like this?, because honestly, it's no concern, and technically better anyway, if I can't charge past 90% anyway. 

mike90045

To charge a 48V bank, you need a higher voltage from the Controller, to push the power into the batteries - often 60V is enough.
  However a MPPT controller usually wants to see about 30% more voltage from the PV, so that it can work the MPPT magic, otherwise, it defaults to a $140 PWM mode.  80V is close to borderline - compare your input amps to output amps and if they are the same amps, you are running in PWM mode and need to increase the PV voltage.   The link below is for the Classic string sizer.

https://www.midnitesolar.com/sizingTool/index.php
http://tinyurl.com/LMR-Solar

Classic 200| 2Kw PV, 160Voc | Grundfos 10 SO5-9 with 3 wire Franklin Electric motor (1/2hp 240V 1ph )| Listeroid 6/1, st5 gen head | XW6048 inverter/chgr | midnight ePanel & 4 SPDs | 48V, 800A NiFe battery bank | MS-TS-MPPT60 w/3Kw PV

boB


2 strings of 3 in series should work great at 40 Voc.

How long is the wire run from the Classic to the PV array ?  If it is a ways away, higher PV voltage can work better than lower.

When you say works well today with the old setup, do you mean 24V battery or 48V battery ?

boB
K7IQ 🌛  He/She/Me

BlackwaterPark

Thanks for the replies gentlemen.  Just got back, and in the fading, obscured light, I'm getting 3.4a in and 4.1a out. This is using the 4 string by 2 series currently.  By what I've seen today, running the panels like that is yielding me ~25% more wattage in. It might be different if the sun was out on full force, so I'll have to wait to meter the two configurations under such conditions... Knowing Maine this time of year, I might be waiting awhile.
BoB, I meant testing both layouts today, and strictly on the new battery system.  The wire run from array to controller is roughly 60' of 6 gauge. It could be the classic itself, also, which has been acting a little strange the last couple months before the change to the new system.  Running that 2460 watt array with all 4 banks of panels on would typically only achieve around 1900 watts on the best, clearest late summer days, when one would be expecting cost y to, if not sightly over, the rated panel wattage.  During peak insolation, it would have the habit of running it's fans hard for a period of time and then suddenly the wattage would go from 1.8-1.9kw down to 11 to 12 watts, and would start that way until I flipped the cc breaker off and on again, which would restore incoming power back to what it was previously, for a short while anyway, into it would again throttle the wattage down to 11 watts.  After awhile, I figured it was just to much panel for the controller and was causing some overheating issues, so I got in the habit of just shutting a bank (2 panels) off on days that were clear and sunny, and that seemed to let the classic run without hindrance. 
So, I'm wondering here, seeing as I'm not likely going to install that 9th panel until spring, of it makes more sense to run 4x2 instead of 2x3, as the extra power from the additional 2 (8 total) panels would trump any efficiency gains of running 6 panels at higher voltages.

ClassicCrazy

I think that what both Bob and Mike said is that 2 panels in series is not a high enough voltage to charge a 48v battery.
Did you plug the panels and your system voltage into the Calculator that Mike linked in comments above ?
If you do that and attach a screenshot that would really be a big help to give advice on your setup.

Larry
system 1
Classic 150 , 5s3p  Kyocera 135watt , 12s Soneil 2v 540amp lead crystal for 24v pack , Outback 3524 inverter
system 2
 5s 135w Kyocero , 3s3p 270w Kyocera  to Classic 150 ,   8s Kyocera 225w to Hawkes Bay Jakiper 48v 15kwh LiFePO4 , Outback VFX 3648 inverter
system 3
KID / Brat portable

BlackwaterPark

Oh yeah, definitely looked at the calculator, long before posting on here.  Accrued all the basic info prior, and weighed it against what I was being told otherwise. In this case, what I was told about the previous configuration not being able to charge the bank is false, if I'm seeing this correctly... It's clearly charged the bank, and at a better clip than what was recommended (bearing in mind with the 4x2, I get an extra 2 panels). On the calculator, both configurations check out green across the board.  If this were a lead acid bank, things might be different and it's likely I'd opt for the fewer panels @ higher voltage to get to consistent absorb voltages(please correct me if I'm wrong about that), but the lifepo4 bank seems to take a good charge the way it is, and I'm wholly unconcerned about getting it to full, as they seem to prefer being in psoc anyway.  This is all temporary at any rate... Come spring I'm going to double the array with a second similar array, though facing a bit more west I think.  And maybe the Hawks Bay or Barcelona might be out by then...

boB


The lower input voltage will definitely work OK to at least bulk charge your batteries.

The only concern would be making sure you absorb the batteries for the necessary time, especially if there are large loads eating up the power at those higher voltages.  As long as the battery voltage keeps as high as it needs to be, it doesn't matter.

Now, if you have Lithium batteries, those don't need an absorb cycle to desulphate them so those should work even better with a semi-low PV voltage.

You should be fine as long as the panels don't get too hot.  It's winter so maybe that doesn't matter much unless you are south of the equator.

If the Vmp is 64 volts or greater, you should be just fine.    You have 60 feet of PV wire so that's not too bad.  Just be glad you aren't one of those with several hundred feet of PV wire !

boB

boB
K7IQ 🌛  He/She/Me