Used Solar Panels from Santan Solar--Only 41% of the Rated Power?

Basspig

I bought forty 240W panels from Santan Solar in January. Have finally gotten around to deploying them and I'm noting dismal output levels when both 20 panel arrays are in full sunlight around noon-1pm. I'm at 42° latitude.

I'm getting about 2kW from the southeast facing array and 1.4kW from the southwest facing array at the time when both arrays are in full sun and their outputs peak.

The array that's producing 1.4kW has a bad cell. I found that last night while checking the junction box bypass diodes. One panel has a short and is reading 1.6V instead of 9.8V that the the other two cell groups read. I pulled that diode thinking it was the diode, but no, it's the panel itself. That panel will have to be replaced. So I disconnected that panel and it's series companion for now.

I have the cells wired in series parallel, with ten groups of 2 in series, for a 48V nominal output for each array of 20 panels. These outputs are connected directly across my 48V battery bank's BMS outputs, so the BMS can shut down incoming charge if the voltage exceeds a set value.

I read about MPPT controllers and wonder whether directly connecting the panels to the batteries is creating a mismatch (similar analogy would be like connecting a 4 ohm speaker to the 8 ohm output of a vacuum tube amplifier--power transfer efficiency would be cut by about half) in load resistance to source resistance of the panels. Is it possible that putting a charge controller could double my panel efficiency, or are these used panels degraded to the point where over half their capacity is gone?

BB.

From (I guess these are your panels?):

https://store.santansolar.com/product/santan-solar-t-series-240w/

Vmp = 30.4 volts
Imp = 7.89 amps
Isc = 8.37 amps

Say your batteries (Li Ion of some sort?) charge at around 52 volts. And your panels are outputting ~7.89 amps (or a bit more in your case) in full sun.

Two panels in series would produce (typical best case cool clear day around solar noon) on a MPPT controller:

• 2*240 Watts * 0.77 panel+controller derating = 739 Watts typical best case to battery charging for 2x panels in series
• 10 * 739 Watts = 7,390 Watts 20 panel array max output... Perhaps 1/2 that on less than ideal days.

Note: Vmp~30 volts is too low of voltage for MPPT controller charging a 48 volt battery bank... Typically, would suggest Vmp-array of ~72 volts minimum for proper MPPT controller operation on warm to hot days... A Vmp-array ~60 volts would be dramatically reduced as Vmp array falls toward 50 volts on typical warm sunny days.

A properly designed array (3 panels in series typically for mid-voltage MPPT controller) would be:

• 20 panels / 3 in series = 9 series parallel strings (3s * 8p)
• 240 Watt panels * 18 s/p array * 0.77 panel+controller derating = 3,326 typical "best case" (a few times a year fall/spring cool weather) with MPPT controller and properly configured array

For your array... Need to look at Vmp-array (for hot day):

• 30.4 volts Vmp * 2 series * 0.81 Hot panel Vmp derating = 49.2 volts Vmp-array-hot

If your battery bank is charging at ~52 volts--The Imp current and Pmp should not be greatly reduced.

However, if your bank is charging at ~58 volts--Then Pmp would be affected more. Possibly 10-20% additional loss of panel Wattage due to "mismatch" of panel Vmp-hot to battery bank charging voltage.

Looking at this chart (generic panel example):


I would be looking at both Array voltage (i.e., 52 volts vs 58 volts charging) and array current (50% to 100% of Imp rated on "sunny days" near solar noon). Array current should be pretty much tied to amount of sun per sq meter on panel, times working voltage (Vbatt charging). (P=I*V)

And Imp at noon would be derated by Sine (angle to sun)... 90 degrees to panel => sine (90) = 1.0 or full solar collection. If 60 degrees to panel, Sine (60) = 0.866 * Imp (or amount of sun falling on array)

Use a Current Clamp DMM and measure the current from each pair of panels. Ideally, each series string should have about identical current to its mates on the array. If you find one or more series panels with substantially lower current--Then you probably have Panel (or possibly wiring) problems.

-Bill

littleharbor2

If you received a couple underperforming panels from Santan I'm pretty sure they'll make it right. 

 As far as wiring up these panels and feeding it directly into your Lithium battery bank I'd say you are making a huge mistake. The BMS should be a last line of defense for your batteries. It is NOT a charge controller

Quote
 " I have the cells wired in series parallel, with ten groups of 2 in series, for a 48V nominal output for each array of 20 panels. These outputs are connected directly across my 48V battery bank's BMS outputs, so the BMS can shut down incoming charge if the voltage exceeds a set value".

Here you say IF the voltage exceeds the set value, Really?  Lithium batteries aren't cheap as I'm sure you know. You may not know how sensitive they are to overcharging. It can  and will ruin them. 

Basspig

My arrays output 69.7V DC open circuit in full sun. They only fall below 55V at the time approaching sunset. Even at sunset, there is still about an amp of current going into the batteries.

But my curiosity is about the MPPT aspect of these charge controllers. If I used one, would I really get more kWh out of the panels than not? I can only believe that the charge controller corrects what would be akin to an impedance mismatch (in amplifier loudspeaker terms) to allow for more power to be transferred.

I suspect your math is in error on the first example..  2 * 240W is 480W, so a number in the 700 range is not possible. If I were getting 739W from each pair, 7,390W from each array, that would be a miracle. But the reality is my east facing array tops out at 2.5kW at 9:45am and my south array with a defective panel tops out at 2kW at 1:30pm.

I contacted Santan a few times.. last month about a panel with a broken junction box, and day before yesterday about the shorted cell in one of the south panels. Have not heard back from them yet.

I don't like the idea of going direct into the battery bank either, but until the funds are available to do more, this feasibility test is what it is. If there is a real advantage in terms of increasing usable power from the panels, I will buy a charge controller. So far, when I look up these devices, they are only for 40amp max. Need 100A for each array to be safe.

With 42kWh of batteries, I never get my battery charge past 65% with base load in the house of several computers and lights and the well pump. Overvoltage just isn't happening with this small panel capacity.

BB.

Yea--That was some stupid Math there... I have fixed the two calculations in the above post.

Sorry,
-Bill

Photowhit

Ever hear the story about George's brother?

https://youtu.be/q0rRKdUZT6M

In the end you are using the BMS for something it was never intending to do...

A Midnite Classic 150 should handle the array's, they will limit the output to their maximum output and can be 'over paneled'. I think max output in a 48 volt system is around 80 amps. 

If you have "42kWh of batteries" which are lithium, Get some protection on them. Guessing they are in your house? Going cheap and building from old car batteries? (Only way you didn't plan to have charge controllers and think they are expensive)

I'd go along with the rest of the world and protect your investment. Don't be like your brother...

Basspig

I've done some research on this matter in the past couple of days and I've learned that PV cells are current sources. They don't transfer power effectively when their output is pulled down to battery levels. I could be leaving 25-30% of the power on the table without a MPPT controller. Even more, the MPPT, but optimizing the power point of the PV cells, can extend the useful hours  of output. That could add another 20% to the overall kWh captured.
As such, I am trying to get a hold of a company that has the Growatt SC4880 in stock. Signature Solar lists them at $269, but out of stock. It looks like two of these, one for each array, would yield the best harvesting of electricity. I'm looking for a dealer that has them in stock at similar price.

Basspig

I called Signature Solar and they are out of stock on the solar charge controllers. They also have no restock date.
Sango Solar has them for $299, but when I tried to order just now, their website timed  out.
Any other dealers carry SC4880?

Alibaba has it, only problem 45 days deliver: https://www.alibaba.com/product-detail/Super-Solar-Growatt-SC-4880-MPPT_1600260169433.html

Zeb Solar has them for $410 each, but shipping from TN to CT is $500!

Vic

Hi B..p,

Part of me, thinks that, perhaps you might want to slow down a bit, in throwing $$ at your system.  Perhaps you could do a complete, well-reasoned system DESIGN.  You know, what are the loads, when do they occur,  etc, etc.

No system is perfect,  but knowing where you need to go, and how is the best way to get there, could save your time and money, it seems to me, in reading a number of the past/present Posts.

Just sayin',  Good Luck, Vic

Basspig

We're a household with two businesses running, a video production studio/recording studio and an amp repair shop with a shop full of test equipment. It is not uncommon for us to use 80kWh a day. We've cut way back by turning off computers and lab equipment, hanging clothes out to dry instead of using electric dryer and curtailing use of my 20,000W multichannel sound system.
There are intermittent loads like a high capacity deep well pump that need high start current but run for only a few seconds may be ten times a day, but there is electric range, oven used intermittently, but the long term drain is air conditioning and computers and test equipment.
I'm building the system scaleable because I cannot afford the $400,000 solar system that would replace 200A service all at once. I'm doing it DiY, at smaller scale, testing it to see how we're doing and then making adjustments.
I've become convinced that the lack of MPPT is wasting perhaps 50% of my PV capacity to load mismatch. Correcting that one problem alone might make the 9600W of PV adequate for our needs if we're careful.

mike95490

Start with the deep well pump, find out it's starting requirements that the inverter must supply.  80kwh daily ?? you are looking at a VERY LARGE system to be able to run it on cloudy days, independent of the grid.
 Generic well pump requirement chart..  Your inverter must be able to supply the Start Amps ( Locked rotor amps ) for at least 1 second to spin the pump up.

BB.

Note that, overall, PWM and MPPT controllers have about the same harvest with an array "sized" to the battery bank. I.e., Vmp~17.5 volts for a 12 volt battery bank.

Because "hot solar panels/cells" will drop Vmp by upwards of 20%, that means the 17.5 volt cells * 0.81 "hot panel" Vmp = 14.2 volts -- Or pretty much 100% of available panel power (Pmp-hot=Vmp-hot*Imp) so PWM and MPPT are of equal "efficiency". (note 81% * Vmp is very hot day, no wind, ~40C temperature rise in panels--I.e., worst case).

One very cold days (i.e., subfreezing) an MPPT controller in the same setup may provide 10+% more power vs PWM.

One of the "nice reasons" to use MPPT controllers... You can run Vmp-array at a higher voltage. Better "hot weather" performance (i.e., charging >14.2 volts on very hot days) and with higher voltage arrays, you have lower array current (P=V*I) for the same level of power. The higher voltage / lower current array means you can use smaller diameter (less costly) wiring from the array to the charge controller. This can also allow you to have longer wire runs from array to solar charger when needed.

-Bill

Basspig

Our well pump pulls 86A at 240VAC startup or LRA and about 11.8A running. Our 6kW LF inverter handles it about as well as our 13000W Northstar generator. There is a blink in the lights when it kicks on, but so far, so good. But that won't be enough to run my huge sound system when I crank up loud, so I have an 18kW inverter on order for the permanent system. I find that what's using the most power are the things that are on 24/7. My computer workstations pull 1200W from the wall outlet. They are loaded with big GPUs for 3D imaging and video editing. Large banks of hard drives for mass storage, etc. So a water pump that runs for 18 seconds every few hours is not significant in terms of battery use. What surprised me is my oil fired hot water heater.. according to the load display on the inverter, it uses over 30% of inverter capacity. My two air conditioners when running, plus all the computers, use 25% of inverter capacity. The oil burner is 25 years old and uses the old transformer igniter which runs continuously, My furnace has a newer burner that has a solid state igniter and that shuts off after the flame is established. So I'm planning to replace the entire burner on the water heater with a newer more efficient model. May save some oil too.

Thanks for the comparison of PWM and MPPT controllers. I can see how MPPT helps to get more power in non optimal light conditions. And winter is usually our biggest energy draw because I leave everything on for the heat it produces. My computers and audio gear will heat my 1000sf studio without use of the heating system in January. It all keeps the studio at a nice 66°F when it's 0°F outside.

We're in a northern latitude, so, unlike AZ or other desert areas, we rarely see temps above the 80s. So panels up here won't get as hot as they do in AZ where it's 125°F. I'm sure there's some derating. I tested in April with one panel and a garden hose. I saw about 2% increase in power when I chilled the panel with water.

mike95490

I'd suggest dedicating the 18kw inverter for the sound system and leave it off when not needed, to save idle power

Basspig

I was thinking the same thing. 270W idle power vs 100W idle power is not insignificant. But I have to work out a rapid, glitch free transfer switching arrangement so that all of the computers in the house that are not on UPS don't reboot. That's the current problem when I switch to/from grid power.. I have to ask everyone to turn off their computers first.
I don't expect the extra 170W to be much of a problem, as long as my solar capacity outstrips my daily kWh use and I go into the evening with 90% or better battery level. It typically will go down to around 65% overnight with the present 6kW inverter.
I cleared more ground and had some high albedo fill brought in (looks like white beach sand) to spread around so the wife is wanting to order me some more panels on Alibaba, so I picked out some 700W panels, six of them, for the third array to fit in that new area I prepared yesterday. With almost 14kW of panels, I should make enough power to meet our full normal usage demands. With good MPPT charge controllers, I should harvest over 70kWh over the whole day.

Basspig

Bought a clamp on ammeter this week and used it to check the ten series strings in the array that was low on output and made a couple of interesting discoveries.
One was that one of the series strings had no current. The rest ranged from 2.7 to 3.3A. This string was dead. Zero. Popped the lid on the junction box and it was all corroded inside! Used panel from Santan Solar. While testing individual panels I used a 2 ohm 100W resistive dummy load. While testing individual strings, I was getting over 7A of current. But when I connected back to the system which goes directly to the batteries, the amperage was less than half. I now realize that this is operating far below MPP.
I have two Growatt SC4880 charge controllers on order with Signature Solar. They are backordered and estimate to be a month out, so I'll have to limp along for another month. Running 2S 5P for each of 2 arrays, there's not a lot of open circuit voltage, so when it's cloudy, total solar output drops from 4kW to 200W on rainy days. I think wiring them 5S 4P for 185V open circuit will help a lot on cloudy days with the MPPT controller. Presently, I get good power for 3-4 hours around high noon. I hope the controllers arrive soon! Going to need air conditioning!

littleharbor2

Have you used the clamp meter on both panels in that string? Shouldn't be too difficult to find the zero current problem with a continuity test  using a digital multi meter.