Problems getting our system producing expected output

[Paul Cahill]

Hi everyone,
After a long and arduous period getting installed on our land I have finally got the solar installed. Now we have the situation of poor PV output and this slow battery charging. The system is as follows: GW5048EM inverter, 12 X Telsun 265w panels connected in 2 serial strings of 6, 2 X Pylon tech 2.4 Li batteries. The panels are currently housed in a temporary frame which is pointing 20 degrees West of South tiled at 45 degrees. I know more directly south would be ideal and probably a steeper angle would help too but we have limitations until the final placement is constructed. Given all this the production figures seem very low. This morning we have brighter weather and the batteries are charging better but there are strange things happening the upper string is producing lower v but higher amps than the lower one, the complete inverse to previous days, no shading so I'm somewhat flummoxed!
Any suggestions would be very welcomed.
Paul

[Rob Beckers]

Paul, moved your post to its own (more appropriate) thread and forum. More people will see it this way.

What charge controller do you have? Strings of 6 panels are somewhat unusual; most off-grid charge controllers have 150V or 200V limits (resp. 3 or 4 series panels per string).

I don't have a magic bullet to offer. The cases where I was called in to troubleshoot systems that weren't performing it was just going through the grunt work of systematically testing and measuring everything. Right down to disconnecting panels and testing them one by one for open Voltage and short-circuit current. The one tool that served me very well for this is a DC current clamp meter. Those are pricey and I wouldn't expect you have one in your toolbox, maybe you can beg/borrow/steal one from a local PV installer? Short of that you can use the charge controller to test string by string, compare Voltage and current with similar light exposure, and possibly rig a few test wires with MC4 connectors to test panels with a regular multimeter for Voltage and current.

There's one case that I remember, where the customer suspected the panels weren't performing well. His batteries weren't charging properly and he wasn't seeing much current coming from the solar array. After I tested every single panel (all normal) it turned out to be the charge controller setting with a much too low absorb Voltage. The controller 'thought' the batteries were full (and tapered down the PV current), while they really were not.

I wish you clarity!

-RoB-

[Paul Cahill]

Hi folks and thank you Rob for your, as always, excellent advice.
Today has seen a dramatic change, due entirely to few hours of brighter sunshine. I was surprised to see that by 2pm the batteries had shot up to 56%, this continued to rise steadily and we're 100% by the time the sun faded. A couple other things were different today string 1 is on the top row, 2 in the lower. String one has always.regiarered the higher of the two outputs, usually within 1 - 5v, today it was lower 120/200 this was variable but pretty much 40+v difference, then a Little later it jumped back to 'normal', above all the voltages were consistently lower but amps very much higher, I assumed this was due to brighter, almost full sun for a couple hours. This is the time the batteries charged exponentially faster than the three previous days, the light was abysmal though.

In answer to your questions Rob, The inverter is a 5 kva hybrid, recently upgraded to use as an off grid with Pykon technology 2.4kWh Lithium batteries, we have 2. I'mno a expert but the kit looks, feels very good quality, the battery cables have Amphenol connectors, super quality. Batteries have built in BMS and are CANbus connected. BMS of the fist battery is the master. Charge parameters are battery specific and preloaded.

Tonight we are using our solar PV's for the first time and whilst it's no major load it has transformed our lives, the pain and cost of the whole thing has proved worth it. Consequence of needing to make every penny count! I know the bank of NiFe i had hoped to use can now be look forward at more closely. Improvements to efficiencies can be made to the current system, meanwhile I can start gathering parts for a secondary system to add 250ah at 24v into the mix. Because we haven't started building yet we have the ideal opportunity to incorporate a second circuit to priority usage, fridge, freezer and lighting circuits. This way we have a better chance of keeping food from spoilng and minimum lighting as part of an emergency standby system.

I'm also going to check all the connections as you suggested, adding dialectic grease when reconnwcting, I will also solder the end tails that I made up to connect to the inverter from the ends of each string. I'm old school and know solder is the only sure-fire way to be confident! I have checked each panel already, no problems and all were very similar. Voltage input to the inverter is rated 100-500v across the two MPPT's. Charging is integrated, as I said the batteries and inverter are designed to work together, all the charging levels, Max DoD, the whole shooting match is automatic once the preferences are loaded into EZManitsel. It's been dark for 4 hours, the Batteries are down to 93% we'll be in bed in 2 hours at which point the load will be 0 except for the Inverter itself, I have no idea what the inverter will eat up during the following 10 hours before we have sufficient light to start producing/charging again, I will post a quick update tomorrow. All in all it seems the weather was simply darker than we realised.

I am truly grateful for your input. regards, Paul

[Michael Scott]

Hi Paul. I am also preparing for a DIY solar system and I still cannot decide which panels are the best in my case. From reading about the different efficiency ratings (https://greentumble.com/most-efficient-solar-panels-on-the-market/) and types of panels, I think I will go with polycrystalline panels (because of the price). Which setup do you have? Maybe your solar panels' efficiency is not good enough? Since efficiency (conversion of sunlight to electricity) can differ a lot. Also I noticed that solar panels have maximum 20-22% eff. ratings.

[Rob Beckers]

Paul, good to hear it's charging properly! A big Voltage difference between strings could be due to shading: When part of a panel is shaded it gets bypassed (panels are generally divided in 3 sections, with bypass diodes), that drops the Voltage of that panel and with multiple shaded panels it can have a large effect on sting Voltage.

Michael, panel efficiency is a non-issue. The way panels get rated is independent of efficiency; a 300 Watt panel of brand A will produce the same power as a 300W rated panel of brand B, if it receives the same amount of sunlight. There are small differences, in particular once you get down to very low light levels, but by and large any A-list brand is fine. The higher efficiency panels are more expensive because they pack a higher rated Wattage in the same form factor (ie. we can get 65" x 39" panels from 220 Watt all the way through 340 Watt, and that 220W panel will of course be much cheaper on a per/Watt basis). To put it another way, 6 panels of 250 Watt work just as well as 5 panels of 300 Watt each, what their efficiency is doesn't make a difference.

-RoB-

[Paul Cahill]

Thanks again Rob,
I'm happy to say we are finally using some power form our system, the weather once again is poor with associated poor light conditions, however we are managing to keep circa 65% charge at the start of the day. We're using less power than we'd like but this definitely seems to be light level dependant, on a bright day we'll have little or no issue even in the panels current less than ideal location. So onto better things!

To recap, I have 20 x 1.2v 250ah Changhong NiFe batteries that were intended to be our primary system. It was pointed out to me that this would be insufficient and I needed to get some power quickly for our current situation. So I bought and installed the system eluded to previously. Now I would like to get back to the NIFe setup and get ready to install it as a completely separate and largely secondary system to augment our overall use which will come on line in a year or two once we finish building. For sure we will be able to gobble most of any excess power in the 12-15000 ltr hot water heat sink I will install under the back porch of the house. This will also be fed with solar water heaters, the idea being that entering the winter months we should have underfloor heating and domestic hot water that will be topped up by a wood burning stove with a back boiler if required, but the heat sink should be up to max, basic calculations say this should last at least 6-8 weeks with little or no additional heat other than the solar HW panels.

Back to the issue. I need to find a way to install this second system as economically as possible with at least moderate reliability. So how do I do that. Research tells me this isn't going to be easy as all the control gear seems quite expensive, mainly for the required charging capability. I will only need 2 - 3 kw output. Do we go for a separate inverter and charge controller or a combined unit and if so which one/s? Housing the batteries will be important to maintain a reasonable temperature, our climate is largely -10 to 40C so an insulated compartment with efficient ventilation is my aim.

Any suggestions would be greatly appreciated.

I came upon this article just this morning on the Outback site, it may be of interest to those running or planning to run NiFe systems although if you're like me you've already seen it, irrespective here's the link:
http://www.outbackpower.com/forum/viewtopic.php?f=24&t=9111

Kind regards to all, Paul

[Rob Beckers]

Hi Paul,

The trade-off is (as usual) between cost and reliability. The PV module should be the easiest part; panels became cheap, and they are very reliable. Any A-brand is fine, whatever you can find cheapest locally. Even used panels would/could be a good deal if the price is right.

With 20x 1.2V batteries I assume you want to wire them for a 24V battery bank? At 3kW of solar PV that requires a 125A charge current, which is beyond most regular charge controllers. If the 3kW of solar you mention is for winter, when there's less sun and it's coming in at a lower angle, then a smaller controller may be all you need. If you want versatility, it's hard to beat a MidNite Solar Classic controller! They have build-in modes for diversion (ie. hot water heating), when the batteries are full, and all the charge Voltage are customizable over a wide range, something you may need for NiFe batteries (don't know enough about them to tell you what charge parameters they need). The Classic 150 goes up to 100A in charge current. Downside is that they are not cheap.

There are cheap pure-sine inverters, but one thing to watch for is that their Voltage range matches that of NiFe batteries. Of the better brands I personally like Magnum most, as giving the best bang for the buck.

FWIW,

-RoB-