Measuring the Isc of small solar panel

[Peabody]

I bought this solar panel:

https://www.ebay.com/itm/184566218058

The open circuit voitage under full sun is about 5.7V, but when I connnect my ammeter across it to measure the short circuit current, I get about 40mA, which doesn't seem like "1W" to me.  Am I doing the measurement wrong, or is it just an example of Ebay BS?  I get same result with my old analog meter.

If anyone knows of a US domestic source of an inexpensive  5V panel with Isc of 200mA or more, please provide a link.

[TheMG]

While it is entirely possible that the panel is overrated, keep in mind that there are many factors at play when it comes to the output from a PV panel.

First of all, they are rated under IDEAL conditions, which you will almost never get. The solar irradiance/potential in your geographical area, weather and smog conditions, time of day, and angle of the solar panel toward the sun, will all affect the maximum output you can get from it, and in most cases that is going to be somewhat less than the maximum rating. So you always want to size your solar panels a bit larger than your energy requirement.

In this case, I would say the panel is very likely overrated. It's physically very small and has an even smaller effective surface area as the individual PV cells occupy only a small amount of the total area! I don't see how you could ever get 1W out of that.

Rough math:
Panel dimensions 132mm x 63mm, total surface area 0.008316 m2

Based on the pictures, I'm estimating only about 40% of that is effective area (actual PV cells): 0.0033264 m2

1 watt / 0.0033264 m2 = 300.625301 watts per square meter

The sun energy reaching the earth: 1360 watts per square meter, only about 70% of this or about 952 W/m2 actually reaches the surface, at most.

That would make your panel theoretically 31.5% efficient in order to produce 1W of power. Even the best of the best solar panels money can buy nowadays, are about 21-23% efficient.

So yes, I'd say that eBay panel is greatly overrated.

[Manul]

I like the "2000mAh" part of its description 

If you tested it under clear sunlight, not angled and not through window glass, I guess it is quite overrated.

[Peabody]

Aside from the price, would this be a reasonable choice?

https://www.digikey.com/en/products/detail/anysolar-ltd/SM531K08L/9990469

I was hoping for battery charging current of about 120mA, and about 40mA for the Arduino and other stuff.

[Renate]

I think that you have to really check your load current, your battery capacity and the worst conditions.
Will you want this to run 24/7 in the wintertime where you only have 8 hours of sunshine and a couple of overcast days?

Are you limited for space? Does this have to be installed on the device/package?

Spending $18 for a Watt doesn't seem like good value.

Here's 15 Watt for $30: https://www.digikey.com/en/products/detail/tycon-systems-inc/TPS-12-15W/14296570
Note also that this gets you a real outside unit with glass and frame, not just a slice of silicon.

With solar there is always some time when your capacity is overkill and some time when it's insufficient.

(I run 2 x 300W and in the worst winter weather usage is restricted, in summer the electrons fly out my ears.)
300 Watt < $1/Watt

[Peabody]

What I'm trying to do is come up with a better alternative to Andreas Spiess' cheap and simple solar power circuit, with power path, for small hobby projects - Arduino, ESP8266 or ESP32:



His circuit doesn't work very well because the power path circuit normally used for fixed 5V supplies doesn't work with variable solar power.  So I want to test an alternative which uses an opamp to drive the mosfet gate, and which works perfectly in simulation.

But in addition to the input voltage limit of the TP4056, the opamp I want to use has a maximum 6V rail, and most very low dropout 3.3V linear regulators also max out at 6V input.  So I really need to stick with 5V solar panels at this point.  Andreas links to an AliExpress listing that has various 5V and 5.5V panels, including one 5V 2.5W panel for under $6.

https://www.aliexpress.com/item/1005001880431126.html

It is described as:

Code: [Select]

5V 500mA 2.5W Solar Panel

Size :130*150MM
Rated Voltage：5V
Rated Current:500mA
Max Power: 2.5W

No indiction of Voc however, which needs to be under 6V.  Of course I don't know if their ratings are any good, but I would certainly try that 2.5W panel except that it would take a month to get here.  I can't find anything similar from a US domestic Ebay source.  There are several multi-watt 5V panels which have female type A USB connectors, but they appear to have some kind of voltage regulation built into the panel, and I don't want that, at least at this point.  I don't have Prime, but probably should look on Amazon too.

So I was thinking the Digikey panel would at least let me test the circuit, although I agree it's pretty expensive  for what you get.  With tax and shipping it would be $25 per watt.

One other question.  If you use two panels in parallel to get higher current, is it necessary to use reverse current blocking diodes on each one, or can you just wire them directly in parallel? 

[MIS42N]

you might have a dud panel. It is hard to estimate the active area, but the quoted area is about .008 square meters. A good sunny day should deliver around 1000W/sq m. If the panel is 12.5% efficient then it should deliver 1W. Any shading anywhere will stuff it up, like the shadow of your arm on one cell. Any haze drops it to about a half. I bought some single cells and soldered them in series to get 12V, they are about 12% efficient. It was interesting but I'd not do it again, just buy a proper panel. Too much effort to get anywhere near a professional result, and only marginally cheaper.

[Peabody]

I found an Amazon listing for a pair of 2.5W 5V panels rated at 500mA each.  But reviews say they're only 250mA.  Still, with two of them, there should be plenty of current.  Reviews also say they cloud over after a year or so, and won't work after that.  Is this common with solar panels?

https://www.amazon.com/dp/B074TYH68Z

So at $13, I guess I'll order these.

[Renate]

So I really need to stick with 5V solar panels at this point.

This seems like an arbitrary constraint.
This is the 21th century, everything has a zillion DC converters.
Why would you expect a solar panel to deliver a constant voltage?

The specs for solar panels are based on high noon in the Sahara desert with ice cubes under the panel to keep the junction temperature 20°C.
Anything short of these conditions will generate less power.

Reviews also say they cloud over after a year or so, and won't work after that.  Is this common with solar panels?

No, because solar panels are not commonly covered in epoxy.
Black solar panels in the sun get pretty hot.
Ever take a heat gun to epoxy?
Most solar panels use tempered glass.

My suggestions were based on my thinking you wanted an unattended data station running 24/7.
If you want something to drag out on sunny days to demonstrate something, that's different.

I thought that the panels that I suggested were pretty neat.
They are the kind of things I see on unattended hydrology measurement points.

[Peabody]

I don't expect the panel to have a constant voltage.  I just want its maximum voltage (Voc) to be less than 6V so downstream parts won't be damaged - without having to add a converter to achieve that.  This is a hobbyist thing where parts count matters.  I'm obviously new to solar panels, but from what I've seen, the nominal panel voltage is more like the MPP voltage, with the open circuit voltage being somewhat higher.  The Amazon "5V" panel has a Voc of about 5.7V per the reviews.

Thanks for pointing out the tempered glass thing.  I would have thought that glass would block some of the high-energy components of sunlight, such as UV, but apparently not.

[Renate]

I would have thought that glass would block some of the high-energy components of sunlight, such as UV...

It does!

A photon has to have a certain amount of energy (band gap energy) to knock out an electron.
For silicon that's about 1.12 electron volt = 1100 nm, infrared.
Ultraviolet has tons (~5X) more energy but you don't get a payoff for that, just the same single electron.
(This is all Planck's constant and energy is proportional to frequency.)
At shorter wavelengths you run into limitations of the silicon and reflection.
Besides, there's not as much UV coming down as visible and infrared light.

https://www.pveducation.org/pvcdrom/solar-cell-operation/quantum-efficiency
https://www.pveducation.org/pvcdrom/solar-cell-operation/spectral-response

Multi-junction solar cells have multiple band gaps and can collect the payoff of the higher energy, shorter wavelengths.
https://en.wikipedia.org/wiki/Multi-junction_solar_cell

[Peabody]

The Amazon panels turned out to be exactly as represented.  The "2.5W" has open circuit voltage of about 5.85V, and the short circuit current is 550mA.  So they will work fine.