diy solar

diy solar

Optimization of solar in winter/grey days

My 4 panels are CanadianSolar CS3U-380MS
Pmax: 380 not 385W, Vmp: 40.0V, Imp: 9.5A, Voc: 47.8, Isc:10.1A
OK, Here are the key specs for the panels
1631986089109.png
1631986132967.png
Also I go down to -40C in winter averaging -22C so how does that affect the 174Voc @ -10C?

With two in series the string voltage @ STC is 2*47.8=95.6V. That leaves 54.4V or 54.4/95.6 = 57% headroom. Anything above 25% headroom is plenty for even the most extreme cold conditions. 3 in series would come to 143.4V and that is too high for cold conditions.

If you want to learn more about adjusting Voc for temperature, this resource may help.


Current on PV line:
Even the short circuit current of these panels is 'only' 10.01A so you should never see much above 30A with 3 strings in parallel. However, for safety reasons the NEC requires a multiplier of 1.55. Even with that multiplier the max would only be 46.6A. I am still baffled on how that would ever trip a 100A breaker (even in very hot conditions).

The from the Pv to the combiner box are #6 AWG
6AWG is more than enough from a safety point of view with 3 strings in parallel and since the normal operating current is going to be closer to 28A, it is good from a voltage drop point of view for runs as long as 100 ft from the controller to the panels. (Longer will still work but the voltage drop will start inching up above 3%)

For 4 strings in parallel, the max current with the NEC safety margin would be 62A. The 6AWG can easily handle that from a safety point of view.
Operating current will be around 38A so runs as long as 80 ft would still be less than a 3% voltage drop. (Since the controller will probably cap the current at a lower value, you can probably still go 100ft without any significant worries about voltage drop)

The big remaining question is can the Classic handle over-paneling with a 4th string. I am pretty sure it can.... but would like to get confirmation from Midnite. If over-paneling is allowed, you should be able to just add the 4th string and not worry about taking them in and out of the circuit.
 
It think I’m understanding better. (Abt time! Lol) The Hyper Voc is the max amt of PV I can drive to.
The Hyper Voc (174V) is the voltage you can drive to without damaging the classic. However, the classic will shut down at 150V to protect itself.
Right now you have two in series on each string. That means you are at 95.6V@STC. If you added the 3rd in series, it will be so close to 150V that as soon as it starts getting cold the Classic will shut itself off... not good.

To add panels, you will want to add more in parallel so you don't raise the voltage of the array.
 
The other 2 panels are 385W, Isc: 10.1A, Voc: 48.21, Imp: 9.61, Vmp : 40.05
Hmmmm. I don't generally like mixing different panels, but the specs are close enough that it should still be OK.. I would feel better about it if the 2 new panels were the same model as the old panels.

I might even be tempted to mix the old and new panels in the strings. Two strings with both old and Two strings with one old and one new. This would make the voltages of the strings a bit closer together....but I don't have enough experience to say this is a good or bad idea.

One mitigating factor is that once the Classic 'tops out', the panels are going to be throttled back and it won't make much difference anyway.
 
i learned that the #16 & #17 breakers are really for disconnecting the PV #16 and batteries #17 and not so much about the current from the PV although,
I agree that #16 is just a disconnect and serves no over-current protection function. With the properties of the solar panel and the size wires you have, even a short circuit anywhere along the PV wiring is not a safety issue.

However, #17 is a bit different. If there is a short between the breaker and the Classic, the battery could supply enough current to start a fire. The breaker will prevent that.
 
I agree that #16 is just a disconnect and serves no over-current protection function. With the properties of the solar panel and the size wires you have, even a short circuit anywhere along the PV wiring is not a safety issue.

However, #17 is a bit different. If there is a short between the breaker and the Classic, the battery could supply enough current to start a fire. The breaker will prevent that.
One more thing: Because the breaker is there to protect against too much current from the battery, If it is a polarized breaker the positive should be toward the battery.
 
You mentioned that you are having a shading problem.

The best way to deal with shading is to avoid it all together by either removing whatever is causing the shade or placing the panels where there is no shade...... but I assume you would have already done this if it was an option.

If you can't avoid the shade, the best but most expensive way to deal with shading is to have smaller arrays, each with it's own MPPT controller. The extreme of this would be to have a controller per panel. Each controller can be lower voltage and lower power (Lower cost per controller) but even then it can be very expensive.... particularly when you consider the wiring costs.

Another way to deal with shading be to have more parallel strings. That way, shade on one string would not impact the other strings. Ideally the individule strings would be positioned in a way that the shade is less likely to hit all of the strings at the same time. NOTE: in this case it would be good to put a blocking diode on each string so a shaded string can't pull down the other strings.

The next way to combat shading is to add bypass diodes. A bypass diode effectively takes shaded cells out of the string so they don't impact the rest of the string. Most modern panels come with at least one bypass diode. Some come with up to 3 bypass diodes. In these panels, each diode can bypass a strip of panels that is 1/3 the width and the full length of the panel. The really good news for you is that your panels have the 3 diodes:


1631993107328.png
@Will Prowse did a good video about bypass and blocking diodes:
 
If you can't avoid the shade, the best but most expensive way to deal with shading is to have smaller arrays, each with it's own MPPT controller. The extreme of this would be to have a controller per panel. Each controller can be lower voltage and lower power (Lower cost per controller) but even then it can be very expensive.... particularly when you consider the wiring costs.
Optimizers can be used to help simulate more controllers. I have never designed with them so I really can't say much about them but my understanding is that they will optimize the panels they are on and then pass the power down to the main controller. I don't know how effective they are.
 
One more thing: None of the shade mitigation methods I mentioned are mutually exclusive. You could have a system with multiple controllers, each with multiple parallel strings with blocking diodes and bypass diodes.
 
Hmmmm. I don't generally like mixing different panels, but the specs are close enough that it should still be OK.. I would feel better about it if the 2 new panels were the same model as the old panels.

I might even be tempted to mix the old and new panels in the strings. Two strings with both old and Two strings with one old and one new. This would make the voltages of the strings a bit closer together....but I don't have enough experience to say this is a good or bad idea.

One mitigating factor is that once the Classic 'tops out', the panels are going to be throttled back and it won't make much difference anyway.
That is what I’ve done 2 x 580 panel series with 2 in parallel and 2 x 585 panel series and in parallel. Couldn’t get two more 580 and so I got the 585s . This is best as I understand it. 580s are paired and 585s are paired to get 24V
The 6 AWG is only 32 feet from the Midnite.
The MidNite shutting down at 150 V is good to know. So that’s more the limiting factor.
Sorry but where does STC is 2*47.8=95.6V. come from,what is “STC”? I get the same number when I calculate my number to be in spec. But I don’t understand it the 47.8 the way you explained it. Got it 47.8, is the Voc! Are you saying each panel is 24v When loaded?
I understood then to be 12v and why I put 2 in series to get 24v. Light bulb moment, maybe. Vmp is 40V.
should they all be in series (6s ie 6 x 40v)? Maybe I don’t understand how it’s connected at the shutoff?
 
It think I’m understanding better. (Abt time! Lol) The Hyper Voc is the max amt of PV I can drive to.
ie if I add two more panels at 385W, I’m adding 770W combined with (4@380+2@385) 2290W of PV must not exceed 174Voc
ie Add 2 panels, I’ll go to 3060W. If divides by 24V, that’s a 127.5Voc which is < 174Voc max
so it looks like I can do it?
I’d recommend reading the following to do your string calculations. You want to optimize your voltage to be compatible with your DC wire runs. You may see large losses running on 6AWG if you get these wrong.
 
One more thing: None of the shade mitigation methods I mentioned are mutually exclusive. You could have a system with multiple controllers, each with multiple parallel strings with blocking diodes and bypass diodes.
I don’t have any blocking diodes that I am aware of unless they are in the shutoff box.haven’t looked in there yet.
 
Here’s a DC wire size calculator. Remember this is all about making sure the wire can carry the power from the panels to the Mppt charger efficiently. Note that watts are not part of the calculation, it’s all about the flow, distance and resistance. A good analog to this is water flow through a firefighter‘s hose. You never see them using a residential hose. https://www.solar-wind.co.uk/info/dc-cable-wire-sizing-tool-low-voltage-drop-calculator
 
That is what I’ve done 2 x 580 panel series with 2 in parallel and 2 x 585 panel series and in parallel. Couldn’t get two more 580 and so I got the 585s . This is best as I understand it. 580s are paired and 585s are paired to get 24V
The 6 AWG is only 32 feet from the Midnite.
The MidNite shutting down at 150 V is good to know. So that’s more the limiting factor.
Sorry but where does STC is 2*47.8=95.6V. come from,what is “STC”? I get the same number when I calculate my number to be in spec. But I don’t understand it the 47.8 the way you explained it. Got it 47.8, is the Voc! Are you saying each panel is 24v When loaded?
I understood then to be 12v and why I put 2 in series to get 24v. Light bulb moment, maybe. Vmp is 40V.
should they all be in series (6s ie 6 x 40v)? Maybe I don’t understand how it’s connected at the shutoff?
OK... back to basics.

When putting panels in series the voltage adds and the current stays the same
When putting panels (or strings of panels) in parallel the voltage stays the same and the current adds.

So, in your current (and Future) configuration you will have 2 panels in series and either 3 or four strings in parallel.
To calculate the voltage you have to add the Voc of each panel in series in a string. Consequently, in both your present and future set up the voltage will be 2xVoc @STC (More about STC later)

Also, in your current configuration you have 3 strings in parallel. Each strings highest current is Isc so your total current is 3times Isc (Before adding the NEC safety factor) When you add another string of 2 panels, it will be 4 times Isc (Before you add the NEC safety factor)

1632000744872.png


STC is Standard Testing Conditions. If you look at the data sheet for your panels, you will see this:
1631999564779.png
All panels are tested and rated under these conditions so it makes it easier to compare panels. However, in the real world you rarely get the full 1000W/m^2 of irradiance and the Cell temperature is almost always hotter. Consequently, you rarely get the rated wattage out of your panels.
Unfortunately, there are many other factors that impact performance so comparing at STC is not perfect. One example is that some panels can disapate heat better than others. This means they will stay cooler and therefor perform better. Another difference is the panels ability to capture light coming in at an angle. If a panel can do a better job of creating electricity from off-angle sunlight it's real world performance is likely to be closer to the STC rating.

The 25C of the STC is particularly important when calculating the effects of cold temperature because you use the temperature difference between the 25C of the test conditions and the coldest your panels will see. (You mentioned you get as cold as -40C. That means that in the calculations to determine cold weather Voc, you will us 25 - (-40) = 65C as the difference in temperature).

Are you saying each panel is 24v When loaded?
I understood then to be 12v and why I put 2 in series to get 24v. Light bulb moment, maybe. Vmp is 40V.
should they all be in series (6s ie 6 x 40v)? Maybe I don’t understand how it’s connected at the shutoff?
When calculating the voltages on the PV circuit, you (mostly) ignore the battery voltage. The solar charge controller converts the solar panel voltage and current to the appropriate Voltage and Current for the battery. The voltage on the solar panel side will almost always be higher than on the battery side of the controller. The current on the solar panel side will almost always be lower than on the battery side.

The one time that the Battery voltage comes into play on the solar panel side is that the voltage on the solar panel side usually has to be some amount higher than on the battery side for the controller to turn on and function.
 
Last edited:
I don’t have any blocking diodes that I am aware of unless they are in the shutoff box.haven’t looked in there yet.
There is a reasonably good chance you don't have any blocking diodes. In your set-up they would most likely be in the combiner box.
 
Here’s a DC wire size calculator. Remember this is all about making sure the wire can carry the power from the panels to the Mppt charger efficiently. Note that watts are not part of the calculation, it’s all about the flow, distance and resistance. A good analog to this is water flow through a firefighter‘s hose. You never see them using a residential hose. https://www.solar-wind.co.uk/info/dc-cable-wire-sizing-tool-low-voltage-drop-calculator
That looks like a nice calculator. The key is picking the correct voltage and current to put into the calculator.

For calculating the wires for solar panels, it is a two step process,

1) I first look at the wire size needed for a really short run but I use the current required by the NEC (1.55x total Isc). This gives me the wire size needed to ensure safety.

2) I then calculate the wire size needed for normal operating conditions (i.e total Impp) and the actual distance between the panels and the controller. This gives me the wire size needed to ensure low voltage drop. (If you want to be more conservative you can use straight Isc)

I then use the larger of the two calculated wire sizes.

If you use the 1.55x total Isc to calculate wire size for voltage drop, you are probably going to end up with a wire size that is way bigger than necessary.....and with the price of copper what it is, that can get expensive.
 
When calculating the voltages on the PV circuit, you (mostly) ignore the battery voltage. The solar charge controller converts the solar panel voltage and current to the appropriate Voltage and Current for the battery. The voltage on the solar panel side will almost always be higher than on the battery side of the controller. The current on the solar panel side will almost always be lower than on the battery side.

The one time that the Battery voltage comes into play on the solar panel side is that the voltage on the solar panel size usually has to be some amount higher than on the battery side.
One HUGE caveat: My comments are all in relation to MPPT controllers like you have. When using PWM controllers, things are very different and it is important to appropriately size the Solar voltage to the Battery voltage.

The concept of 12V solar Panels and 24V solar Panels harkens back to when PWM controllers (or even no controllers) were the only game in town.
To have a good setup on a 12V battery with a PWM controller you would use a 12V panel and know the solar voltage was about right. (16-18V) For a 24V system, you would use 2 12V panels in series or one 24V panel.
 
I have spent all day on calculating the actual Voc and Isc for the total array. Again thank you so much!
I have learn a tremendous amount of information! Thank you.
I have been working on ideal principles and not realistic. I now have a much better understanding.
I read all the documents you sent and probably will again, over and over. Thank you
Sorry for sending you back to first principles. Lol
I just have not seen many of the hidden details which was confusing me because I could not get my calculations to work.
I forgot to mention the 385W panels were Q. Peak Duo L-G5.2 385 and I only have two which are in parallel.
Wow thank you so much! And the diagrams are really good!
Tremendous!
 
I got a message back from Midnite. As I suspected, it is OK to over-panel the classic as long as the voltage does not exceed the limits.


Subject: MidNite Classic Date: Sep 20, 2021 From: Todd Geers - SUPPORT
Hi Paul,

Yes you can over-wattage the Classic, just ensure the string voltage is below the CL150 input limit of 150V plus extra for cold weather headroom.

Todd


So, you should be able to add the extra string and not worry about swapping them in and out.
 
I got a message back from Midnite. As I suspected, it is OK to over-panel the classic as long as the voltage does not exceed the limits.


Subject: MidNite Classic Date: Sep 20, 2021 From: Todd Geers - SUPPORT
Hi Paul,

Yes you can over-wattage the Classic, just ensure the string voltage is below the CL150 input limit of 150V plus extra for cold weather headroom.

Todd


So, you should be able to add the extra string and not worry about swapping them in and out.
Thank you so much! I’ve been doing lost of research since we last texted. I found the MinNite calculator just last nite and when I ran the numbers it came back green for go and identified I was over wattage .
so I was going to call Monday. So I will do a 2S4P as it does appear to work.
‘Again thank you so much! It is greatly appreciated. And I learned a lot!
 
And I learned a lot!
That is the important thing and I am glad to hear you say it. My primary purpose for posting is to help people learn. The calculators are handy, but if you don't understand the underlying reasoning and math, it is easy to still get things wrong.
 
Back
Top