The Solar Panel Negative Connection

[Steve_switch]

I have read that using the Solar Panel Voltage to plot in a graph is a good indicator of the suns brightness and the general brightness of the daylight hitting the panels. I monitor and record all battery properties, and have recently decided that I'd like to also record the Solar Panel Voltage as a Light Guage or a brightness gauge. I don't know how I would convert that measured voltage to a unit of Lumen, but until I figure that out, I can still just go ahead and read the Solar Panel Voltage as an analog input to my Arduino Mega. I have designed an additional circuit that I have added to my daughterboard, to step the maximum solar panel voltage down to be an input to my Mega, and not destroy it.


My inquiry in this forum is to ask if I should tie the Solar Panel Negative to the Arduino Ground. Normally I would think that 'yes' I should, but I've been having some misgivings about that connection because currently the Arduino Ground connection is connected to my Negative Battery Terminal, which is connected to my Invertor Negative input, which is connected to the Grounded 120 V ac output of the inverter, which is connected to Chassis Ground on all of my electronics (Charge Controller, Battery Monitor and Solar Panel Frames and Mounting Hardware and to and outside Earth Grounding Rod. And everything works great. But I've noticed a subtle Voltage difference between the Earth ground and the Solar Panel Negative Wire. So if I measure my Solar Panel Voltage (with a Voltmeter) using Earth Ground as reference, the voltage is a little different than using the Negative Wire Lead from the Solar Panel. The difference is subtle. So until someone can point me to a good discussion on this topic, I think that I would be the safest to take my measurement of the solar panel voltage between the Positive wire of the Solar Panel and the Ground of the Arduino Mega. Can anyone shed some light on this?

 

[MisterSandals]

I don't know how I would convert that measured voltage to a unit of Lumen,

The voltage and current (aka watts) are a better indicator. Without a load, a panel can reach Voc without much light.

 

[12VoltInstalls]

Can anyone shed some light on this

That’s actually pretty funny
“don't know how I would convert that measured voltage to a unit of Lumen”
That is all dependent on a lot of things. So many that it is irrelevant.
As stated just note your wattage output.

 

[BentleyJ]

Yes, Watts is a much better indicator of solar irradiance especially considering the MPPT charge controller is ramping the current up and down, which changes the voltage, trying to find the highest Watts available for given conditions. Also, the readings would have to be temperature corrected per the PV datasheet coefficient.

 

[rogerheflin]

My panels hit VOC/VMP a few minutes before sunrise, and stay at VOC/VMP until a few minutes after sunset. There is little variation of voltage in the middle, and the panels facing straight south vs southeast only differ by 2-3 minutes. Current does significantly vary.

 

[Steve_switch]

Thank you, 12VoltInstalls, for responding so quickly to my inquiry. You quoted my statements and replied, no!!! But I didn't say that I connected the Solar Panel Negative to Earth Ground. Currently I don't have the solar panel negative connected to anything except the Solar Panel Negative input to the Charge Controller. So you're agreeing with me that the Solar Panel voltage measurement, as an input to the Arduino should be as I said, between the positive of the solar panel and the Ground of the Arduino. Again, I'm not connecting the Negative of the solar panel to anything except the negative solar panel input to the charge controller. Do you know of any documentation or literature that discusses this aspect of off grid solar?

Thank you, MisterSandals for your quick response as well. I think you both are correct, I can use my Watts reading. I'd still like to lo the panel voltage though.

I also have read a little more about the conversion to Lumens. It turns out I can convert the irradiance (W/m²) to lux (lm/m²) using an approximate conversion factor. I do collect and save the current watts going into my battery, and I know the area of my solar panels, so I think I can get some semblance of a light meter.

 

[12VoltInstalls]

You quoted my statements and replied, no!!! But I didn't say that I connected the Solar Panel Negative to Earth Ground

Yes. Sorry about that.
When I saw @MisterSandals response and he did not mention that I re-read and realized I’d misread- so I deleted that portion. Apparently not fast enough

Do you know of any documentation or literature that discusses this aspect of off grid solar?

No. Irradiance to lumens could probably be done with some calculations built into your coding; well not probably- it can be.
I do not think that to be relevant in practical use but if that suits a desire for discovering the science continually there’s no reason not to do that.

watts) are a better indicator. Without a load, a panel can reach Voc without much light.

… and as another mentioned, the ambient temperature, the state of charge (decreasing ability of the battery bank to accept any charge nearing fully charged state), the fact that voltage isn’t current, the panel’s ‘efficiency’ specification, and the variability of passing clouds are all factors.
It’s sufficient to monitor watts of input in getting an indicator of “what’s happening?”

I'd still like to lo the panel voltage though.

recently decided that I'd like to also record the Solar Panel Voltage as a Light Guage or a brightness gauge

The issue with ‘voltage only’ as a metric is that in operation your mppt SCC is going to load the array(s) to achieve the most stable and optimized output amperage at the charging voltage parameter you have selected in the SCC options. So a VOC of say 100VDC will under load be something way less, like maybe 70VDC. VOC as you probably know is open circuit- no load. When the SCC loads the panels it can only do so with ‘resistance.’ The amperage will vary with loading as influenced by irradiance and resistance and resistance (ohms) ‘sucks the voltage down’ once the circuit is no longer open.

Just worry about the watts. We aren’t going to be trying survival under a shade-dome on mars while drilling random holes looking for water. (yet)

I think I can get some semblance of a light meter.

again, if that satisfies your interest go ahead

The way I see it is that if I’m getting watts- whatever they are- then whatever else is happening is subject to the math and science that is going on already without my participation. When I “bought the SCC” I chose to abdicate any responsibilities for the engineering and design to the product manufacturer’s R&D team and I’m ok with that, now. 40 years ago I tried to conquer too many things LOL but as I aged I realized I wasn’t going to live long enough to meet the challenges of being Smokey Yunick, Robert Goddard, Ben Franklin, Nikolai Tesla, and automotive icons Bill Mitchel, Henry Miller, and Zora Arkus-Duntov.

If however you need and want the mete challenge of doing the coding you can but I don’t think you will get either useful information or accurate results:

using an approximate conversion factor

Again, my apologies for initially missing the grammar indicating you didn’t actually have the panel neg(-) connected to your earth/dirt/ground rod.
I was incredulous that a post discussing an arduino could include such a glaring oversight yet I did successfully reveal once again that ‘speed reading’ has downsides.

 

[TM48]

I would use a separate photo diode or photo transistor to measure irradiance. As pointed out the voltage curve of a typical silicon solar panel is not ideal. A quick search results in available components and sample circuits which are not complex and can be fed safety from the Arduino bus.

 

[Steve_switch]

Thank you, 12VoltInstalls for replying once again. I love the detail that you responded in, and very much appreciate your input to help me think this project through. Your comment, "40 years ago I tried to conquer too many things LOL but as I aged I realized I wasn’t going to live long enough to meet the challenges of being Smokey Yunick, Robert Goddard, Ben Franklin, Nikolai Tesla, and automotive icons Bill Mitchel, Henry Miller, and Zora Arkus-Duntov", especially made me think, because I'm starting to think that I am beginning to over think this thing a little, and your little nudging in the quote above helped me to start to realize that maybe my project may be nearing completion. As I said earlier, my off grid solar automatic transfer switch works. I continue to rethink my original design concepts and keep thinking, maybe I could add one more feature.

 

[12VoltInstalls]

starting to think that I am beginning to over think this thing a little

I’m somewhat familiar with that part of town.
Glad I could help you find your way.

 

[Steve_switch]

So, should the Negative Lead coming off the Solar Panel be tied to anything other than the Negative Solar Panel input to the Charge Controller?

 

[SolarUKWM]

But I've noticed a subtle Voltage difference between the Earth ground and the Solar Panel Negative Wire.

You're very lucky to have noticed. And that it's a small difference.

Other possibilities -- you didn't notice, it was large, you connected them together and there was a bang/smoke from something.

So many grounds, negatives to choose from, so many ways to get it wrong!

Depending on the inverter topology, you can have all sorts of AC and DC offsets between the PV positive/negative and mains earth. Same story between battery negative and mains earth.

 

[SolarUKWM]

I would use a separate photo diode or photo transistor to measure irradiance.

This would also be far safer (both to @Steve_switch and to any poor little defenceless Arduino) -- eliminating problems with grounds etc.

 

[SolarUKWM]

So, should the Negative Lead coming off the Solar Panel be tied to anything other than the Negative Solar Panel input to the Charge Controller?

I would say no -- leave the panels connected to the charge controller (+ and -) and stay off those wires!

 

[12VoltInstalls]

So, should the Negative Lead coming off the Solar Panel be tied to anything other than the Negative Solar Panel input to the Charge Controller?

No

 

[WrenchLight]

The PV negative input should never be connected to ground. I've seen people blow up inverters doing that.

 

[Steve_switch]

Does anybody know why this is so. A link to a discussion on this would be a wonderful thing. I'd love to understand the physics of this.

 

[WrenchLight]

Does anybody know why this is so. A link to a discussion on this would be a wonderful thing. I'd love to understand the physics of this.

When PV neg is grounded, it can forward bias an IGBT and dump all the capacitor energy into the IGBT, blowing it up.

 

[CaySol]

I don’t have my panels up yet. Waiting on some approvals. So all I have is my weather station that I export the data to home assistant. Fun to look at all the sun I’m not making use of.

As others have said. Your watt output should correspond to irradiance directly. 1000w/m^2 should give you full output.

 

[12VoltInstalls]

Does anybody know why this is so. A link to a discussion on this would be a wonderful thing. I'd love to understand the physics of this.

Another thing; electricity seeks any path to complete a circuit.

So if you include ‘one side’ of a variable usually higher voltage DC circuit then whatever circuitry components are in line in the way of the DC voltage potential difference completing its circuit in some way unintended.

 

[Steve_switch]

Thank you, WrenchLight, I have read about IGBT, or Insulated-Gate Bipolar Transistors and their ability to control high current with a change in voltage input. I am unfamiliar with their application inside of Charge Controllers, if that is the device that you are referencing. Please elaborate.

I know that my Windy Nation TrakMax 30L is a MPPT Solar Charge Controller. I'm really not up on or familiar with the circuitry of it. I don't have a schematic for it, but I'm starting to develop an idea of why the Negative Solar Panel input cannot be tied to anything else. The idea is that in order to electronically modify the resistance presented to the solar panel in order to present a higher load to it, the Charge Controller circuit alters its Ground Plain Leval, reference in order to adjust the load that it presents to the solar panel and find its Maximum Power Point. So instead of the Positive side of the Solar Panel Voltage fluctuating to find the Maximum Power Point, the Ground Reference is raised or lowered. I'm not sure what electronic device would do this though or is this just a matter of perspective.

​

 

[SolarUKWM]

I have read about IGBT, or Insulated-Gate Bipolar Transistors and their ability to control high current with a change in voltage input. I am unfamiliar with their application inside of Charge Controllers, if that is the device that you are referencing. Please elaborate.

The exact technicalities of how an IGBT works, and all the clever switching and control happens, is probably not helpful, but ...

Would you connect a wire between Live Mains and Ground? I hope the answer is "no" ...

Now look at this picture. If you ground the negative side of these panels (bottom left corner) there is a path that leads straight back to live utility. Via an IGBT. You would be placing that IGBT directly between live, and your newly added ground.

This is because in all-in-one devices like this picture, there is often no galvanic isolation between things. Note: The _battery_ here is shown as isolated, by a transformer. But nothing else.