Low light output on different PV panels

[TAH]

I am interested in what anyone else might have found with differnt types of panels in low light conditions. I have thin film panels and the lowest output I have had so far is 13% on a day that was very dark and snowed the whole day. That's 13% of KWH that the panels put out on a sunny day that same week.

[ghurd]

I'm not sure what you are looking for.  Too many variables, like a 'self-regulating' PV makes a lot less power under less than good conditions. If it is thin-film or crystalline makes no difference.

We sometimes have days here where any normal PVs don't reach charging voltage, or if they do, not for long and at a very low output.

But this is a dark crappy place in Dec & Jan. Average worst case 30 days is less than 0.5 hours/day.

G-

[TAH]

What I was looking for is what output dropped to under the worst conditions. I know it wil not be exact but different panel types supposedly work better at much lower light levels than others. Going by the percent change should lesson some of the location variable.

[craig110]

The problem with what you are trying to compare is that the "worst case" is nighttime and anything brighter than that is subject to our interpretation of what is a bad condition.  Many of the vendors have power output charts showing what their panels give in differing light intensity levels.  Looking at those charts would likely give you the best apples-to-apples comparison of how the panel types behave in low-light conditions.

[TAH]

I have looked at the charts. I would like to see if they are accurate. I know there are many veriables but if someone has a charge controller that tells KWH for the day I am sure they have looked at it if not recorded it. If it says 10 KWH one day and another day within a week or so only does 1 KWH then that is 10%. I would consider this to be a fairly accurate test of various panels and it would eliminate many of the location differences and other variables. It would be a min max of a set of panels at a fixed site.

[craig110]

Well, ok, since you insist...  Here is my info: Some cloudy days I get about 50% of a clear day's production and other cloudy days I get well under 10%.  With cloud thickness variability, how are my numbers helpful?  I might not be getting at what you're thinking, but I'm not sure how you'll get numbers that are sufficiently comparable to validate the manufacturer's charts.  If you question the validity of the charts, perhaps asking the manufacturers to describe how they derive the numbers would give you a better sense of whether to believe the charts or not.

Craig

[TAH]

But what type of panels do you have and is the less than 10% of a day of full sun with the same length day or the maximum ever?

[craig110]

Sorry, but knowing the inaccuracy of my readings (I wasn't there all day monitoring the cloud cover and didn't measure the light intensity) and the difficulty of comparing one person's subjective "bad cloud cover" to another's, I'd prefer not to say what type of panel I have so as to not accidentaly disparage or overly applaud that manufacturer based on unsound science.  I gave you the 10% / 50% numbers just to show how different cloudy days can be (and yes, those days and the comparative sunny day were just a few days apart so the day lengths were basically the same) so that hopefully you'd see that the numbers you want won't help you.  If you are trying to decide which type of panels to buy, let me be blunt and say that the inherent subjectivity behind this approach will do you a disservice.

Craig

[ghurd]

It is my understanding they must be accurate, or the PVs can't be UL Listed.

A non listed PV is more of a toy.  Or so small being UL listed was not worth the cost.

Factors other than the PV can change things too. Wire losses, crappy controllers...

G-

[SamoaPower]

I did a study 20 years ago on my then-new multi-crystal panels (11% efficiency) to determine what their utilization factor was for this location (small tropical Pacific island).

We may have one or two totally clear days per year. Mostly it's partly cloudly (or partly sunny if you're an optimist) and can go up to a week of overcast. Panel ambient temperature variation isn't much of an issue since year-round daytime temps don't vary by more than +/-5F degrees with a mean of 85F. Hours of daylight also varies little since we're at 14 degrees south latitude.

Collecting data for three years with a fixed array, showed a utilization factor of close to 20% of their rated energy.

Interestingly, the same panels (20 yrs old) are now delivering close to 35% with a tracking system and ridge concentrator.

[ghurd]

Extra Crappy today, even for here.

Tested a small PV for giggles.

High noon.  ~1.45% of the 'sunny amps' into a battery.

Sunny with this PV is about 275ma, today is 3.8 to 4.1ma.

At least nothing got too hot.  :-/

G-

[nothing to lose]

I agree with you, sunny day, cloudy day, percents, etc.. mean nothing here for any type usable basis.

I have one pannel. On a bright clear day it made 100% of the power it would make on a bright clear sunny day. High volts, good amps.

But 2 days later we had a storm nearby and it was DARK because of thunderstorm cloud cover all day. Those really nasty thick purple black clouds, about 2pm (afternoon/daytime) it looked like 10pm (late night/dark time), and all day was that way though sometimes it was not raining.

My pannel made 0% power that day. I know this because about 11AM when I should have been making the most power I could make that day I disconected the pannel and connected a test wind genny becuase I was gettting winds and it was turning well. At that best time of day my pannel read less than 12V open, It was darkout so I expectd nothing and that's what I was getting. I was below battery charging volts and that was about the best sun light of the day. This was about a 35-50watt pannel, unsure the brand.

So I went from 100% on a good day to 0% on a real cloudy day within about 2 days.

 I think this would be a good average for about any brand of pannel and pretty acurate testing of highs and lows I can expect based on cloud cover in my area now and then

Percentage of cloud cover will probably effect percentage of power, and I have come to the conclusion that the percentages pertain in reverse order to some very rough degree of which I am not yet certain.

It would seem at this time that 0% clouds equals 100% power,

But 100% clouds equals 0% power. I have not test the power cruve ratio like 25/75, 50/50, or 75/25 to see if the ratio holds steady, pretty shure it would not. But 0/100 and 100/0 is correct ratio for highs and lows.

However this is not totally acurate and could vary with voltage of battery bank being charged also. I was getting 0% into a 12V battery, however I did have some lower volts, perhaps I could have been charging a 6V battery at .5 or .25 amps??

Certainly at 5-7V pannel output I could have been charging a bit into my sub-C 1.3V cells.

In such case of lower volt battery bank, I may have been getting 0.2% or 0.05% power?

I did have a tiny 12V fan that starts up around 2-3V, just a few watts, and it would run very slowly at that time of 0% so there was some amount of power, though not usable to me right then other than a tiny fan or some leds maybe.