Selling home, solar inspection issue with VOC measurement.

[Sunking]

Not if the home was already sold. OP says the house was sold. The question is if the buyer who already agreed to buy is justified in using this as a reason to exit the deal.

My bad, If they have closed, deal is final unless the agent or seller failed to disclose vital information.

[AzRoute66]

OK, my first post so I might be missing something. Not that the previous arguments weren't good, but I think the first thing that should be taken into consideration is that the inspecting electrician made his measurements at 0800 hours. By my calculations, assuming the date was June 21, and Shasta Lake, CA was in the PDT (GMT-7) time zone then solar noon was at 1311 (over five hours away), and at 0800 hours the altitude of the sun would have been 24.35 degrees and the azimuth would have been 79.22 degrees (not even south enough for East yet). This is a LONG way from Standard Test Conditions, and perhaps this intuitive argument would be easier to digest than the temperature coefficient of Voc. Use the suncalc.org tool to put in the correct date, exact location, and play around with the time of day.

The fact that he got the same voltage on both strings is also a good intuitive argument for 'all is well'.

Frankly I am almost surprised he got as high of a Voc as he did.

[sensij]

OK, my first post...

Frankly I am almost surprised he got as high of a Voc as he did.

Stick around, there is lots to learn! Voc doesn't change with irradiance, and Vmp does only a little.

[AzRoute66]

sensij, Thanks for the encouragement. I know I have lots to learn. But I do think that Voc does change with irradiance, especially when the panels are getting no direct sunlight, only diffuse. Remember, his panels face south at 28 degrees and the sun was at only 24 degrees BEHIND them. This is why I encouraged him to play with the suncalc.org tool. I am also ciphering on the powerline shadows he saw in the photo, but thought I would let him double check the situation before I asked about that as it isn't really relevant to reassuring his buyers.

[sensij]

sensij, Thanks for the encouragement. I know I have lots to learn. But I do think that Voc does change with irradiance, especially when the panels are getting no direct sunlight, only diffuse.

No, irradiance only affects the amount of current that can be generated, not the voltage. The same Voc will be observed in diffuse or direct light, if the temperature is the same.

[max2k]

No, irradiance only affects the amount of current that can be generated, not the voltage. The same Voc will be observed in diffuse or direct light, if the temperature is the same.

details, details- senij is correct in 97% of the cases but if the diffused light is so diffused that it changes Voc then panels are out of their operating range anyway which makes him 100% correct .

[Sunking]

sensij, But I do think that Voc does change with irradiance, especially when the panels are getting no direct sunlight,

Wrong. You can take any solar panel, put it in a dark room, turn on a flashlight and the panel will have full Voc. Current is what changes with irradiance. Solar panels are current sources.

[J.P.M.]

sensij, Thanks for the encouragement. I know I have lots to learn. But I do think that Voc does change with irradiance, especially when the panels are getting no direct sunlight, only diffuse. Remember, his panels face south at 28 degrees and the sun was at only 24 degrees BEHIND them. This is why I encouraged him to play with the suncalc.org tool. I am also ciphering on the powerline shadows he saw in the photo, but thought I would let him double check the situation before I asked about that as it isn't really relevant to reassuring his buyers.

Voltage changes very little as a direct function of irradiance. Voltage changes almost entirely in response to and inversely as panel or cell temp. Cell temp. is f(POA irradiance, wind and cell output) as determined by an energy balance on the device. Current changes pretty much as a direct and linear function of POA irradiance to the point that arrays in unshaded areas arrays actually make pretty good pyranometers once calibrated to a standard. See it every day.

POA irradiance has 3 components: Direct (beam), diffuse (received from the sky dome as scattered by the atmosphere), and reflected, which can be either beam or diffuse (or possibly) concentrated from specular reflecting or diffuse reflecting objectstive objects onto the array. Some calcs I SWAGged for Redding, CA on a clear morning, 06/21 at 0700 STANDARD time put the POA irradiance at ~ 350 - 375 W/m^2.

For a good treatment of irradiance calcs and estimates, see Duffie & Beckman.

I'm not too concerned about the discrepancies between measured voltages and the temps. that might lead to those voltages as the somewhat non specific nature of the dates, times temps. and irradiance conditions don't warrant more accuracy. The panels being 5 yrs. old and probably fouled doesn't increase the confidence level, but none of that not in any way to imply the array is not operating nominally.

[J.P.M.]

details, details- senij is correct in 97% of the cases but if the diffused light is so diffused that it changes Voc then panels are out of their operating range anyway which makes him 100% correct .

AzRoute66: Get ahold of a copy of any decent, undergraduate solar text. Your education is incomplete at this time.

Max: For any solar PV device you can buy, Sensij is correct in 100 % of the cases.

The intensity of the irradiance, be it direct, diffuse, or reflected usually and commonly as expressed as "Plane of Array" Irradiance is what will affect the temp. of the array, and the array's current.

An energy balance on the array will show that, as the irradiance increases, the array temp. will increase, or vice versa. That temp. increase is what will decrease the array voltage, or conversely, an array temp. decrease will increase the voltage, BUT, to be clear, irradiance, in and of itself will not change the voltage in any measureable way. Irradiance increases will increase the device temp. That temp. increase is what will decrease the voltage.

[max2k]

AzRoute66: Get ahold of a copy of any decent, undergraduate solar text. Your education is incomplete at this time.

Max: For any solar PV device you can buy, Sensij is correct in 100 % of the cases.

The intensity of the irradiance, be it direct, diffuse, or reflected usually and commonly as expressed as "Plane of Array" Irradiance is what will affect the temp. of the array, and the array's current.

An energy balance on the array will show that, as the irradiance increases, the array temp. will increase, or vice versa. That temp. increase is what will decrease the array voltage, or conversely, an array temp. decrease will increase the voltage, BUT, to be clear, irradiance, in and of itself will not change the voltage in any measureable way. Irradiance increases will increase the device temp. That temp. increase is what will decrease the voltage.

what I was referring to is the following pic showing different IV curves at different levels of irradiance. If Voc was not dependent on that all of them would hit the same point on V axis. On the graph it shows Voc increases with irradiance despite its negative temp coefficient. Bad source?

My own LG305 panel spec shows similar pic- Voc increases by 3.5V when irradiance goes from 200 to 1000 W/m2. If I'm reading this correctly it might explain OP troubles as this effect is stronger than temperature dependency and they work in opposite way: if one measures Voc on cloudy but hot day he might get
35.5 V / panel instead of 40.5V / panel.
pro_bulb1.jpg

[bcroe]

I don't see the concern about the exact value of Voc. The higher irrad curves are pretty close together.
We already know the effect isn't linear over orders of magnitude (moonlight measurements demonstrated
that), secondary effects come into play. I would probably just lump it into a "leakage" effect. Bruce Roe

[J.P.M.]

what I was referring to is the following pic showing different IV curves at different levels of irradiance. If Voc was not dependent on that all of them would hit the same point on V axis. On the graph it shows Voc increases with irradiance despite its negative temp coefficient. Bad source?

My own LG305 panel spec shows similar pic- Voc increases by 3.5V when irradiance goes from 200 to 1000 W/m2. If I'm reading this correctly it might explain OP troubles as this effect is stronger than temperature dependency and they work in opposite way: if one measures Voc on cloudy but hot day he might get
35.5 V / panel instead of 40.5V / panel.
pro_bulb1.jpg

Most all panels have similar graphs. The Vmp as shown at the Pmp on such graphs is due to the circuitry rooting around looking for the Mpp at an irradiance level. It's pretty much a function of the electronics and circuitry rooting around looking for an Mpp. One physical manifestation of all the math of the 5 or 6 parameter cell model is that the change in the slope of the locus line of Mpp's as irradiance changes is mostly caused by the circuitry adjusting the voltage to get to the Mpp as the current increases in response to an increase in irradiance. The goal is to get max. power out of a panel.That means, if everything is working according to plan, neither the Voltage or current will likely be at a max. at that point. Strictly speaking, depending on the model (and some on the method of looking for the Mpp), the MPP current at any point, including an MPP will change logarithmically with the coincident voltage (see Duffie & Beckman, 2d ed., eq. 12.2.1, p. 771). But again, whatever that Voltage is at any point, it's value is due to the search for the Mpp. See Duffie & Beckman or the SAM documentation for more information and a more complete description of how all this works.

FWIW, my array's Voltage is always greater when irradiance is low, typically in the A.M. But, at the same POA irradiance levels later in the afternoon (and I've measured it several times) after things heat up, the array string Voltages are lower. I'm pretty sure that decrease in Voltage is pretty much and entirely caused by the array being at a higher temp. Unless I've got some way to instantaneously increase irradiance, and measure any changes VERY quickly, measuring any Voltage change as f(irradiance) is a very tricky and imprecise business.

[bcroe]

Unless I've got some way to instantaneously increase irradiance, and measure any changes
VERY quickly, measuring any Voltage change as f(irradiance) is a very tricky and imprecise business.

I'd throw a saw tooth current drain (dummy load) on it at perhaps 5 HZ, low enough frequency
to ignore system inductance but high enough to have minimal thermal change. Watch the
waveform of the voltage relative to the current; it will probably be pretty flat. Better measure
voltage at the panel terminals so resistance of the copper doesn't affect it. Bruce Roe

[max2k]

I'd throw a saw tooth current drain (dummy load) on it at perhaps 5 HZ, low enough frequency
to ignore system inductance but high enough to have minimal thermal change. Watch the
waveform of the voltage relative to the current; it will probably be pretty flat. Better measure
voltage at the panel terminals so resistance of the copper doesn't affect it. Bruce Roe

it would correspond to one of those IV curves above for the given irradiation level. That's probably how they obtained those curves in the first place except may be combined with calibrated irradiation source they could control in steps so the whole thing would get all those curves in one run of multiple 'tooths'- one per irradiation level. Similar approach was used to measure transistor's IV curves- the device would supply base current in steps and then swipe set range of collector voltages measuring collector current. Transistor wouldn't noticeably heat up during this scan. The scan voltage or current in your case can be any smooth waveform as long as its value can be measured and its range controlled. The form just affects the 'speed' of the swipe not the values. Those transistor IV testers simply used unfiltered rectified AC.

[bcroe]

it would correspond to one of those IV curves above for the given irradiation level. That's probably how they obtained those curves in the first place except may be combined with calibrated irradiation source they could control in steps so the whole thing would get all those curves in one run of multiple 'tooths'- one per irradiation level. Similar approach was used to measure transistor's IV curves- the device would supply base current in steps and then swipe set range of collector voltages measuring collector current. Transistor wouldn't noticeably heat up during this scan. The scan voltage or current in your case can be any smooth waveform as long as its value can be measured and its range controlled. The form just affects the 'speed' of the swipe not the values. Those transistor IV testers simply used unfiltered rectified AC.

Actually I think my idea FAILS, because its changing current instead of irradiance. Oh well. Bruce Roe