/P solar panel in PA.
This page was last updated : 07-Sep-26 10:11:24
This page I document my finding and measurement regarding a few low voltage solar panels I have. To get the motivation behind it, I suggest you start with "How it all started".
Enjoy.
- How it all started
- 23th of December 2025 - grey overcast winter day with no sun at all
- 26th of December 2025 - sun is shining and measurement automation is on.
- 27th of December 2025 - nice typical Dutch overcast day
- 28th of December 2025 - fog
- 7th of February 2026 - sun and a cnc340x220-9 solar panel
- 8th of February 2026 - overcast and a cnc340x220-9 solar panel
How it all started
In the last months of the year 2025, the Dutch government started a campaign "Denk vooruit" (think ahead) to get the Dutch be prepare for an emergency situation. The idea behind it, is that you could manage and help yourself in the first 72 hours at home, should an emergency situation arise.
The basic thoughts behind this campaign, is that in the Netherlands, we almost always have electricity, water and internet. But in an emergency, such as extreme weather, war or a major outage, those supplies could suddenly fail. It might take a little while for help to arrive. So it’s smart to be properly prepared. That requires action from everyone: from the government, but also from you and those around you. By preparing properly, you could help yourself and others if the need arises. The website and the information booklet tells you how you could to do that.
One of the things they say we should to is putting together an emergency kit for 72 hours and they have created a PDF for that to download.
Aside from water, food, medications etc, they also advice to put in a battery-operated radio and mobile with power bank to stay informed (interesting though that the mobile network would last for more then one day, but ok) and a flashlight with extra
batteries, candles and matches.
In a Dutch Facebook group "Voorbereiden Noodpakket" a lot of people talk about small "emergency radio's" and flashlight with little solar panels and crank generator in it. And other people advice to use off-grid or portable solar panels as a power supply. Knowing something about solar panels in various small sizes (used to build a few solar powered IOT/LoRaWAN nodes) and have owned a few powerbanks with small solar panels on it, I was interested what these solar panels could do in typical Dutch December weather (grey and clouded).
I own a "small" 60W /P solar panel (amazon link / manufacture link) which I bought for field radio experiments and camping, so I decided to put that to the test on a typical Dutch overcast day in December.
The 60W /P solar panel is Ecosonique 60W Portable / Foldable Solar Panel with a maximum output of 60W based on the optimal voltage of 18V and a operating current of 3.33A. It has 3 output connectors (DC, USB-A and USB-C) with various output voltage/current combination (I assume some PowerDelivery protocol is used for configuration).
It comes with a DC to DC cable, some adapters for various DC busses and a adapter cable with a XT60 and Anderson PowerPole connector on it (big plus).
To put this panel to the test, I'm using a semi-pro Maynuo M9812 electronic load to measure the :
- open voltage
- current for a constant voltage of 13.8V
- current for a constant voltage of 12V
- current for a constant voltage of 5V
All measurements are on the (unregulated) DC port but maybe in the future I'm going to test the USB ports also.
23th of December 2025 - grey overcast winter day with no sun at all
With the setup shown below I started to measure.
Preforming the measurements, I got these values.
| Open voltage | CV 13.8V | CV 12V | CV 5V | |
|---|---|---|---|---|
| Voltage | 18.325 V | 13.8V | 12.0V | 5V |
| Current | -- | 0.0224A | 0.021A | 0.0391A |
| Power | -- | 0.31W | 0.25W | 0.2W |
Sorry for the crappy table, but my CMS has some problems with those :-(
Conclusion : using this solar panel in these condition, even a 60W panel as used in this test, it will take 284 hours* to charge a 10Ah powerbank.
* for this result we use a Realistic Efficiency of 90% in the following formula.
Charge time(hours) = ( Capacity(Ah) / Current(A) ) / efficiency =Time(hours) (10Ah / 0.0391A) / 0.9 = 284 hours
Side note : I posted my findings in the Dutch Facebook group "Voorbereiden Noodpakket" and a interesting discussion was forming between people who do have some technical knowledge and could apricate the post, and people who didn't take the time to really read to post. Some told me that almost all emergency radio's and light do have dynamo cranks to power them and the solar panels did work. Why I know they didn't read the post, because after a short introduction, I wrote the conclusion indicating that it with the 60W solar panel on a clouded grey day, it would take 284 hours to charge a 10Ah power bank. Only after this conclusion, I summed up all the measurement data etc. because I know not everybody is interested in the technical mambojambo. Then later that day, one of the moderators send me a FB message requesting to remove the amazon link from the post (it wasn't allowed to have product links in a post but a screenshot is allowed). But I didn't responded fast enough (a FB message from somebody not a friend will be handled as a "message request" and not shown into my notifications), the post was removed by the moderator. So that was my last time I will post technical stuff on that group.
26th of December 2025 - sun is shining and measurement automation is on.
On the 26th of December the sun was kind of shining and I setup everything to do some measurements. Because it's kind of freezing outside (-5 dgr C), this time I used a think 2.5mm2 twin cable between the M9812 to the solar panel and fed it through the cantilever window, allowing me to control the M9812 within arm's reach and from the comfort of my warm living room. (I like cold outside, but behind the desk I don't) :-)
Based on the Maynuo M9[78]xx Library (source: https://github.com/djacobow/maynuo_load), I wrote a small python script to control my Maynuo M9812 and make a constant voltage measurement sweeping from 21 to 3 volt in 0.5 volt steps. The script is still work in progress, but for now it works for me.
Setup 60W panel direct sunlight
For the first test of this day I setup the 60W solar panel on a garden chair and pointed to the sun for maximum exposure and then run the python script.
$ python3 solar_test.py -f '2025-12-26 - 60w - sun.csv'
The scripts generated a CSV file for post processing and also shows a chart having the current and power set against the constant voltage in a range of 21 to 3v volts with 0.5 volt steps. The chart is also automatically saved in PNG and PDF file format.
Click on image for PDF or download the data in CSV format.
Conclusion : Please keep in mind that it's December and the solar power isn't that strong (according the KNMI, it was just just 1 on a scale of 1 to 10) and even in this dull sun, this is where these things are made for.
With 6 hours of this sun you could fully charge your 10Ah power bank and under these (not even great) conditions, you may even have enough power available to make a few QSO's with your Yeasu FT817 and/or Elecraft K2 at a low power setting.
A QRP-Labs QMX with a TX current of around 0.7A for 5W with 12V you even have spare to charge your mobile phone.
Setup 60W panel NO direct sunlight
After just 13 minutes the sun has set so much that with the same setup I could a "no direct sun" measurement.
My shack cat Moppie was also aware of the sun setting and was trying to get the last of it on her windowsill place.
$ python3 solar_test.py -f '2025-12-26 - 60w - no direct sun.csv'
And as expected the current / power output in this case has dropped dramatically.
Click on image for PDF or download the data in CSV format.
Conclusion : although better then the 23th of December 2025, it's still not much. You could try to charge your 10 Ah powerbank (aprx 0.13A at 5V) in just 85 hours. But at 12V the current of 0.119 A isn't hardly enough to power a common /P /QRP hamradio radio like the Yeasu FT817 in RX mode or even a Elecraft K2 RX and battery save mode on (150 mA battery save mode). A QRP-Labs QMX with a receive current 80mA could be you goto rig in these conditions.
12V 2W solar panel
After this measurement I wasn't finished yet for today. In my junk box I have a few small 12V 2W solar panels gathering dust and I had plans to use them to charge a battery powered Meshcore repeater on the roof. So I pulled them out for some measurements to see if they are OK for that.
Note that there a no labels on the panel and it's clearly mend for indoor usage, so that will need some work.
Setup 12V mini panel direct sunlight
I place the panel in direct sunlight and executed :
$ python3 solar_test.py -f '2025-12-26 - small - sun.csv'
Click on image for PDF or download the data in CSV format.
Conclusion : Looking at the result seems that this solar panel is limited to 100mA output. At 5V it should be able to charge a 2Ah 16550 battery over multiple days (expected to need 22 hours for a full charge).
At 0.1A, it will still take 111 hours to fully charge a 10Ah powerbank.
Setup 12V mini panel NO direct sunlight
I pulled out the sun and executed :
$ python3 solar_test.py -f '2025-12-26 - small - no direct sun.csv'
Click on image for PDF or download the data in CSV format.
Conclusion : As expected the current / power output in this case has dropped dramatically and it couldn't even reach the 12V requirement. But using it for 5V I still generated a current of 9 mA.
Which could be enough for trickle charging a 16550 battery. And don't get me started on a 10Ah powerbank.
Overall conclusion for today : putting a little bit effort into measurement automation make live much easier. I was able to fast collect source data (less : change in light conditions) and use that data for generating charts which make it easier to interpret the results.
although better then total darkness, y
27th of December 2025 - nice typical Dutch overcast day
On the 27th of December it was a nice typical Dutch overcast day, so I wanted to repeat my measurements from the 23th of December.
To get a feeling of the light intensity, I used a LUX meter on my mobile phone to measure the light intensity showing a value of 5400 Lx (should have done that on the other days also).
For the 60W solar panel with a m2 size of 0.337m2, in lumens(lm) that would be : 1819.8 lm
For the small 12V solar panel with a m2 size of 0.012m2, in lumens(lm) that would be : 64.8 lm
For the conversion from Lx to lm I use this online calculator : https://www.rapidtables.com/calc/light/lux-to-lumen-calculator.html
Setup 60W panel overcast day
For the first test of this day I just lay the 60W solar panel on a garden table and then run the python script.
$ python3 solar_test.py -f '2025-12-27 - 60w - overcast.csv'
Click on image for PDF or download the data in CSV format.
Conclusion : although better then the 23th of December 2025 (it's probably less overcast), it's still worse then the no-direct sunlight measurement of the 26th of December. You could try to charge your 10 Ah powerbank (aprx 0.08A at 5V) but it will take 138 hours. But at 12V the current of 0.07A isn't hardly enough to power a QRP-Labs QMX with a receive current 80mA.
Setup 12V mini panel overcast day
I pulled out the sun and executed :
$ python3 solar_test.py -f '2025-12-27 - small - overcast.csv'
Click on image for PDF or download the data in CSV format.
Conclusion : As expected the current / power output in this case has dropped dramatically and it couldn't even reach the 12V requirement. But like the previous measurment (no-direct sunlight) at 5V I still generated a current of 9 mA. I would have expected it to be less.
Overall conclusion for today : overcast day suck regarding solar power efficiency.
28th of December 2025 - fog
On the 28th of December started out very nicely, and I event went into the field to make some QSO's (see this link), but within 30 minutes the temperature dropped from 7 dgrC in the sun to 0.1 dgrC.
So when I was back home, I used a LUX meter on my mobile phone to measure the light intensity showing a value of 1370 Lx (should have done that on the other days also).
For the 60W solar panel with a m2 size of 0.337m2, in lumens(lm) that would be : 461.69 lm
For the small 12V solar panel with a m2 size of 0.012m2, in lumens(lm) that would be : 16.44 lm
Doesn't look that promissing.
Setup 60W panel fog afternoon
For the first test of this day I just lay the 60W solar panel on a garden chair in the sun direction and then run the python script.
$ python3 solar_test.py -f '2025-12-27 - 60w - fog.csv'
Click on image for PDF or download the data in CSV format.
Conclusion : you could trickle charge you powerbank with it. For the rest, not really usefull.
Setup 12V mini panel fog afternoon
I pulled out the sun and executed :
$ python3 solar_test.py -f '2025-12-27 - small - fog.csv'
Click on image for PDF or download the data in CSV format.
Conclusion : the small solar panel is totally useless on in this fog, no current availible.
Overall conclusion for today : fog day suck big time regarding solar power efficiency.
7th of February 2026 - sun and a cnc340x220-9 solar panel
On the 7th of February started out very nicely, and because I still had a cnc340x220-9 solar panel in stock which I wanted to use to power my MeshCore repeater.
After some digging on google I found out that it should be a 10W solar panel which is also available in 18, 12 and 6 V versions.
Because I can't find the specifications for the 9V version, I used the specifications for the 18 volt version to do some math (which may be totally bullshit) :
- Color : Blue / Black
- Material: Polycrystalline solar cell
- Output Voltage: DC 18V
- Open Voltage: 20V
- Power 10W
- Working Current: 550mA
- Size: 34mm x 22mm x 0.3mm
- Certification: CE ROSH
- Standard Test Condition: 1000W/m2、25℃、AM1.5
According to this page, 1000W/m2 from the sun is aprx 98000 lumens(lm) for 1m2 which is equal to 98000 lux.
Bullshit starts here . . .
When we take the size of this panel into account, we can calculate that the test conditions, according this calculator, would result in 7350 lumens(lm).
Because we have a winter sun and defiantly not lab conditions, I don't expect it to deliver 100%.
To get a idea what to expect I used a LUX meter on my mobile phone to measure the light intensity showing a value of 56505 Lx.
When I converted this value for solar panel with a m2 size of 0.075, to lumens(lm) that would be : 4237.875 lm .
I'm not sure if there is any linearity, but I could expect a bit above 50% current / power, so aprx 5W of there is any linearity in these solar panels.
On stack-overflow I found this answer :
The photovoltaic effect itself is quite linear (but not 100% due to serial resistance and other effects), You have other factors that comes in on a system.
The major one, is the temperature dependency of the solar array, which is negative for the Pmax and quite important. As the irradiation increases, the temperature on the array does as well, which lower the efficiency of the panels.
Another factor, is that as the irradiation increases, the panel produces more current, and thus produces more losses in the cables, inverter and so forth as Ploss=R*I^2.
The spectral irradiation will also have an impact as c-Si panels are more sensitive in infra-red, lower irradiance in the morning/evening may actually have slightly more efficiency as IR traverse the atmosphere with more ease.
In the end of the day, many factors affects the power output.
So overal, there are to many factor to expect a output of aprx 5W. And also I don't use a calibrated LUX meter.
Bullshit ends here . . .
Important note : I found this paper which I want to study to get more knowledge about how to measure solar power efficiency and conversions.
Note : because I didn't have any weather protection, I designated it as a indoor panel.
Setup cnc340x220-9 solar panel sun
For the first test of this day I just lay the cnc340x220-9 solar panel on a garden chair in the sun direction and then run the python script.
$ python3 solar_test.py -f '2026-02-07 - indoor - sun.csv'
Click on image for PDF or download the data in CSV format.
Conclusion : So it confirms that this is a 9V solar panel (hench the -9 in the identification) with a peak power of 3.4W at 9.6V under these winter sun condition.
This may be even a good start to power my MeshCore repeater.
Setup cnc340x220-9 solar panel indirect sun
For the second test of this day I just lay the cnc340x220-9 solar panel on a garden chair facing off the the sun so it only collects indirect sun light and then run the python script.
$ python3 solar_test.py -f '2026-02-07 - indoor - indirect sun.csv'
Click on image for PDF or download the data in CSV format.
Conclusion : in a indirect sunlight condition, this may become a problem but still could supply 50% of the peak current used by the repeater node.
Setup cnc340x220-9 solar panel behind window sun
For the third test of this day I put the cnc340x220-9 solar panel behind a HR++ glass window directly facing the sun as shown below.
This will be the way it will be used. I'm going to 3D print a holder for it and use suction cups to place it on the attic window (also HR++ glass)
and then run the python script.
$ python3 solar_test.py -f '2026-02-07 - indoor - window sun.csv'
Click on image for PDF or download the data in CSV format.
Conclusion : Still usable, although we would dropped a bit of current. I expected to look simulair to direct in sun graph, but maybe a thin cloud got in front of the sun. That the problem when measuring stuff and you are depending on nature.
Overall conclusion for today : This cnc340x220-9 solar panel is a good starter for my MeshCore repeater. Although it doesn't supply much power, using the Wavershare Solar Power Managment Module i have confidence that it can power the repeater and even charge the batteries on a sunny day.
Update : the panel is installed and its up and running 24/7 from the moment it was installed.
Live data can be seen on this page : https://ssl.ph2lb.nl:8943/meshcore/rpt_tel_chart.php (the SSL certificate is self signed so you need to ignore the warning to continue).
8th of February 2026 - overcast and a cnc340x220-9 solar panel
The 8th of February was a typical dutch overcast day and a excellent moment to test the cnc340x220-9 solar panel in these conditions.
Again I used a LUX meter on my mobile phone to measure the light intensity showing a value of 10021 Lx.
When I converted this value for solar panel with a m2 size of 0.075, to lumens(lm) that would be : 751.575 lm .
Setup cnc340x220-9 solar panel overcast
For the first test of this day I just lay the cnc340x220-9 solar panel on a garden chair in the sun direction and then run the python script.
$ python3 solar_test.py -f '2026-02-08 - indoor - overcast.csv'
Click on image for PDF or download the data in CSV format.
Conclusion : with a maximum of 50mA, it's just enough current to power the Heltec V3 based MeshCore repeater. On heavy trafic it will start to drain the batteries.
Setup cnc340x220-9 solar panel overcast behind window
For the second test of this day I just lay the cnc340x220-9 solar panel panel behind a HR++ glass window directly facing the sun as done yesterday and then run the python script.
$ python3 solar_test.py -f '2026-02-08 - indoor - overcast - window.csv'
Click on image for PDF or download the data in CSV format.
Conclusion : with a maximum of 25mA, this may become a problem but it still can supply a part of the required current. This graph also shows that the behind the glass measurment from yesterday potentialy was effected by a thin cloud got in front of the sun.
Overall conclusion for today : Lets see if the panel can keep the repeater up during winter. If not, I need to look for another solar panel.
Next steps . . .
- Repeat the measurement on the 60W and 12V panel again on a overcast day in December.
- Repeat the measurement on the 60W, the 12V and the cnc340x220-9 in spring and summer to see what we can pull out of them one.
- Buy some emergency radio/flashlight/etc gadget with build in solar panels and repeat measurements on solar panels to see if the addition of the solar panel is worth it (bought 2 already, check out this page).
- Order a few Power Delivery adapters to do some measurements on the USB ports on the 60W panel
- To make things easier, the Maynuo M9812 has interface allowing me to write commands for configuration and measurements, I'm thinking about writing a program which automates the measurement for me.
For example I would be interested to sweep between 15V and 5V with steps of 0.1V in constant voltage mode to create a voltage/current chart.
To be continued . . . . .
Continue reading on this page.
Previous page: Twisting enamel wire
Next page: Gadgets with solar panels