# Measure Daily Output of a Solar Panel

Discussion in 'Renewable Energy' started by Moneer81, Sep 28, 2008.

1. ### Moneer81New Member

Joined:
Apr 5, 2004
Messages:
46
Likes:
0
Location:
Minnesota
Hello,

I am trying to measure and compare the total daily output of two identical solar panels: one is stationary and the other tracks the sun. I know how to measure the voltage acorss a load and determine the power but what is the best way to measure the total power produced, from sunrise till sunset, on these panels?

Thanks.

2. ### WillbeNew Member

Joined:
Aug 16, 2008
Messages:
1,236
Likes:
7
Location:
MD
Total energy. . .?

Assuming a resistive load:
.....you could make an amp-hour meter from an IC Voltage-Frequency Convertor
http://www.national.com/mpf/LM/LM331.html
and a shunt resistor. The Freq. increments a counter. Total count + total time = total amp-hours.
.....You also need a voltmeter that displays average voltage. Use another V-F convertor and counter and divide the voltage count by the hours.
.....Then you have average V and total amp-hours, and so total watt-hours.

Last edited: Sep 28, 2008
3. ### ericgibbsWell-Known MemberMost Helpful Member

Joined:
Jan 4, 2007
Messages:
21,178
Likes:
644
Location:
Ex Yorks' Hants UK
ONLINE
hi,
Are you planning to record every day or just a day or so.?

Joined:
Jan 12, 1997
Messages:
-
Likes:
0

5. ### Moneer81New Member

Joined:
Apr 5, 2004
Messages:
46
Likes:
0
Location:
Minnesota

Sorry for the late reply. I am just planning to record a day or so. Any suggestions?

Willbe - thanks for your suggestion. Is that the only method to do so?

Thanks.

6. ### mnearyNew Member

Joined:
Feb 27, 2007
Messages:
4,502
Likes:
67
Location:
California USA
Do you have an 'identical' charger and storage battery for each solar panel? Just charge both batteries and then discharge each one under identical conditions (Same inverter powering the same load. Inverter should have a low-battery cutoff. Measure time to cutoff for each battery. Swap batteries and repeat the next day.)

7. ### LeftyretroNew Member

Joined:
Mar 11, 2007
Messages:
1,420
Likes:
13
Location:
Hercules, California
Well if you are only going to compare a couple of days differences you might consider the following lower cost but more math solution.

If you wire both panels to identical load resistors and place each resistor in identical size oil baths you could trend the temperature rise of both baths over time. After you acquire the raw data you can use excel to calculate the power generated by each panel over time. I'm sure a little searching could find the proper temp/BTU/Calories/watts relationship. Again this would require more math but the hardware cost should be lower.

Lefty

Last edited: Oct 7, 2008
8. ### WillbeNew Member

Joined:
Aug 16, 2008
Messages:
1,236
Likes:
7
Location:
MD
I guess you could add charge in chunks to two capacitors using sampling and the final "volt-hours" and amp-hours would correspond to the capacitor voltages. Then you multiply once.
Exactly how to do this with rising and falling voltages doesn't exactly jump out at me. One of these, maybe. . .?
http://en.wikipedia.org/wiki/Charge_pump
but I've never designed one.

Or
a voltage controlled current source (an FET or vacuum tube) feeding the cap, with ΔV across the cap = IΔT/C. 1v cell voltage means 1 mA into the cap, 10v means 10 mA, and for a 1000 µF cap you apply the 1 mA current for 10 mS and you get 10 mV.
Adjust the current and its duration to avoid voltages higher than the cap's rating by the end of the day.
You'll need low-leakage caps.

Last edited: Oct 7, 2008
9. ### leedudeNew Member

Joined:
May 4, 2007
Messages:
22
Likes:
0
for watts:
put some resistance on the output of each panel, and measure the voltage across the resistor.

P = V²/R
power = voltage²/resistance

Last edited: Oct 7, 2008
10. ### ericgibbsWell-Known MemberMost Helpful Member

Joined:
Jan 4, 2007
Messages:
21,178
Likes:
644
Location:
Ex Yorks' Hants UK
ONLINE
hi,
As its only a few days I would suggest the following.
Interface a MCP3204 , 4 channel 12bit ADC to the parallel port of a PC.
Use two of the channels to monitor panel voltage and the other two channels for current.

This would give W1 = V1*I1 and W2 = V2*I2 which could be plotted on the PC.

If you decide on this method I could help out with the PC interface details and a simple log program.

11. ### Moneer81New Member

Joined:
Apr 5, 2004
Messages:
46
Likes:
0
Location:
Minnesota
I am aware of that. The question is how do you keep track of the power through out the day, and how do you record it and add it up?

12. ### ericgibbsWell-Known MemberMost Helpful Member

Joined:
Jan 4, 2007
Messages:
21,178
Likes:
644
Location:
Ex Yorks' Hants UK
ONLINE
Was my suggestion of any use.?

13. ### jfedisonNew Member

Joined:
Dec 1, 2008
Messages:
11
Likes:
0
Location:
Rochester NY
Logging solar irradiance data

Hello,

Another approach would be to log the solar irradiance from sensors placed next to each panel. Assuming the panels are identical, this would tell you how much energy you are getting for each configuration. The comparison will vary depending on the cloud conditions and angle of the sun and panels. A clear day would be a good starting point for comparison.

You can check out the following link for a "solar data logger" kit that logs solar irradiance and cummulative incident energy.

Micro Circuit Labs SDL-1

The kit manual describes how to compute cumulative solar energy in kWh from logged irradiance data as well as some other background on solar power.

Hope that helps.
Jeff Fedison
Micro Circuit Labs

14. ### Mike_2545Super Moderator

Joined:
Jan 4, 2009
Messages:
1,057
Likes:
29
Location:
Toledo, Ohio
Use a sunny boy inverter. Hook it to a sunny web box and presto they track the output for you!

15. ### XanNew Member

Joined:
Jan 26, 2009
Messages:
2
Likes:
0
Hi!

An idea I thought ov a while ago to do something similar to this was to get a cheap electric clock & replace the one transistor ocillator with a circuit where the freq is current dependant.

Not sure if the releationship between the time measured would be that linear using this method but if you made up two they could be used for comparison purposes.

With suitable calibration you could make the hands ov the clock move say 1 hour if there's 1 amp flowing for that time, but as I said under or over this current it might not be linear.

Hailz!

Xan

16. ### KMoffettWell-Known Member

Joined:
Oct 5, 2007
Messages:
2,496
Likes:
58
Location:
Minnesota, USA
Xan,

"Cheap" electric clocks (i.e. 1.5V desk and wall clocks) have crystal oscillators and are very accurate even as the batteries run down. They are not current dependent.

Ken

17. ### WillbeNew Member

Joined:
Aug 16, 2008
Messages:
1,236
Likes:
7
Location:
MD
If you get more than 0.15 hp per sq. yd. of solar panel, you're doing pretty well.

18. ### Mike_2545Super Moderator

Joined:
Jan 4, 2009
Messages:
1,057
Likes:
29
Location:
Toledo, Ohio
This is how I do it...

The 577 Foundation

Go to the page above and click on the charts it will show you daily hour averages and you can get monthly charts showing daily KWH production. The link also talks about BTU production from solar water heating panels. Lots of other solar charting and great stuff. I really had fun with that project.

Last edited: Feb 5, 2009
19. ### duffyNew Member

Joined:
Dec 22, 2008
Messages:
2,618
Likes:
73
Location:
nortern 'sconsin
Wow, that's a really nice setup. I like the online monitoring.

20. ### XanNew Member

Joined:
Jan 26, 2009
Messages:
2
Likes:
0
Using a Clock to measure Solar power output.

I realise that they have xtal ocillators in them. What I'm saying is disconnect that, get rid ov it & connect instead an ocillator that can have it's freq changed by a voltage or current input to the electromagnet that makes the pendulum swing.

You would have to convert the voltage to current because a solar panel's voltage stays fairly constant throughout the day if it's lightly loaded, then falls off quickly as the sun is setting. But I guess you would be loading it heavily as you want to figure out the total power it produces.

Hailz!

Xan

21. ### Mike_2545Super Moderator

Joined:
Jan 4, 2009
Messages:
1,057
Likes:
29
Location:
Toledo, Ohio
Xan,
getting a clock to run off of solar panels will not tell you output in KWH, efficiency, or in any way be a good indication of power produced. At best it would be able to tell you for how long the sun shone that day. Solar panels are not simply on or off they fluctuate with available sunlight. That is dependent on many factors such as; time of year, cloud cover, distance from equator, whether it's fixed or tracking. If it is tracking ,is it single or dual axis? All of which contribute to the performance of a given panel.