Best solar panels for the buck

[Sumner]

Thanks. I've attached the IV curves for my 210 watt Photowatt PW2050-210 panels.....................I probably need to divide voltage in half since the two panels are in series....

You are running them in series? I looked at the specs on those panels and the voltage could be between 26.8 and 32.8. If so you could be running 60 volts or so through the lines to the controller?

What voltage range does the controller operate over and do you feel comfortable with 50-60 volts on the boat?

Sum

[FONT=Arial, sans-serif]Our Endeavour 37[/FONT]

[FONT=Arial, sans-serif]Our Trips to Utah, Idaho, Canada, Florida[/FONT]

[FONT=Arial, sans-serif]Our MacGregor S Pages[/FONT]

[FONT=Arial, sans-serif]Mac-Venture Links[/FONT]

 

[prroots]

You are running them in series? I looked at the specs on those panels and the voltage could be between 26.8 and 32.8. If so you could be running 60 volts or so through the lines to the controller?

What voltage range does the controller operate over and do you feel comfortable with 50-60 volts on the boat?

Thanks. Our Morningstar TriStar MPPT 45 is rated for 150 Volts input. I could actualy put 4 of these panels in series without problems. Morningstar advised me to run them in series for better performance. I am not concerned about the Voltage especially since we already have 120 VAC running around the boat.

As an aside, the TriStar MPPT 45 is (also) a data acquisition system allowing you to monitor or log 31 separate variables. As one example, I have attached a graph showing array voltage as a function of time of day. As you can see it averages around 55 Volts during most of the day.
Pete

 

[prroots]

Using this source

http://eosweb.larc.nasa.gov/cgi-bin/sse/grid.cgi?email=
​

I get an average of 4.12 kWh/m2/day on a horizontal surface in February at my location. My question is how to use this number in conjunction with my IV curves as given in a prior post. I think I can divide it by 1000 watts/m2 to get 4.12 hours per day. The value of 1000 watts/m2 represents standard conditions upon which basis the panels are rated. This time represents the equivalent period over which I'm receiving 1000 watts/m2. If I then apply a 10% temperature de-rating, I should then get about 2 X 200 watts for 4.12 hours on average each day. That would convert to 2 panels X 200 watts/panel X 4.12 hours / 12.25 Volts = 134 Ah/day which is about double what I've been seeing. Of course, I haven't accounted for any shading and I do get shadowing from my wind generator during the morning hours. It does account for overcast conditions since the 4.12 is an average over the last 22 years. Does this sound right? As an aside, the Ah collected represents 31% of the rated power of panels (ie, 134/420) which is very close to the 1/3 rule of thumb for the tropics.
Pete

 

[Maine Sail]

Using this source

http://eosweb.larc.nasa.gov/cgi-bin/sse/grid.cgi?email=
​

I get an average of 4.12 kWh/m2/day on a horizontal surface in February at my location. My question is how to use this number in conjunction with my IV curves as given in a prior post. I think I can divide it by 1000 watts/m2 to get 4.12 hours per day. The value of 1000 watts/m2 represents standard conditions upon which basis the panels are rated. This time represents the equivalent period over which I'm receiving 1000 watts/m2. If I then apply a 10% temperature de-rating, I should then get about 2 X 200 watts for 4.12 hours on average each day. That would convert to 2 panels X 200 watts/panel X 4.12 hours / 12.25 Volts = 134 Ah/day which is about double what I've been seeing. Of course, I haven't accounted for any shading and I do get shadowing from my wind generator during the morning hours. It does account for overcast conditions since the 4.12 is an average over the last 22 years. Does this sound right? As an aside, the Ah collected represents 31% of the rated power of panels (ie, 134/420) which is very close to the 1/3 rule of thumb for the tropics.
Pete

Pete,

With your panels in series any shading across all "strings" can really hurt performance. Even just a little bit can take a BIG bite out of performance.. People often underestimate how much just a little shading can hurt. Series can be better, or worse, but it will depend upon your situation and your type of shade. There are pro's and cons to each..

Also be sure your bank can actually "accept" the current being thrown at the bank and keep battery "acceptance" in mind when figuring these numbers. Ideally this testing should be done in bulk mode for the most accuracy.

 

[prroots]

Pete,

With your panels in series any shading across all "strings" can really hurt performance. Even just a little bit can take a BIG bite out of performance.. People often underestimate how much just a little shading can hurt. Series can be better, or worse, but it will depend upon your situation and your type of shade. There are pro's and cons to each..

Also be sure your bank can actually "accept" the current being thrown at the bank and keep battery "acceptance" in mind when figuring these numbers. Ideally this testing should be done in bulk mode for the most accuracy.

Thanks. I am making sure that batteries stay in bulk mode. As far as parallel vs series goes, I need to find a treatise on this subject. I will say that I am getting shading from the wind generator during the morning, but I am always getting about 55 Volts output with the current varying. At the moment, it is pouring rain and I am getting 51.4 Volts and 0.63 amps out of the panels.

I need to rig up a jury-rig that allows me to switch back and forth between parallel and series. I did size my cables to go either way
Pete

 

[Maine Sail]

Thanks. I am making sure that batteries stay in bulk mode. As far as parallel vs series goes, I need to find a treatise on this subject. I will say that I am getting shading from the wind generator during the morning, but I am always getting about 55 Volts output with the current varying. At the moment, it is pouring rain and I am getting 51.4 Volts and 0.63 amps out of the panels.

I need to rig up a jury-rig that allows me to switch back and forth between parallel and series. I did size my cables to go either way
Pete

Pete,

With series poly or mono panels you are creating a big loop. If you shade across the panel it can reduce the output of both panels depending upon where the shade is and where the by-pass diodes are for the "strings". Series can get you to a higher voltage earlier in the day and keep it up later in the day but you really need to test it on your boat for your application. In parallel shading of one panel does not affect the output of the unshaded panel but in series both panels can be affected, depending upon the lay of the shade... Pro's & con's to each...

 

[Bill Roosa]

MS is right on the money (as usuall). If you consider each panel as being composed of 0.5 volt batteries (the individual cells) what happens if one of them goes dead (is shaded)and they are all in series? the whole panel has to push all the current through a dead cell.
What happens if they are wired in parallel? You loose the output of that one cell and amps go down a bit.
This is why bypass diodes are so important. they allow the dead cell to be bypassed with little effect. To be effective though you need a diode for each cell (not the blocking diode for each panel BTW)
There is a move afoot to wire all the cells in parallel for an output of 0.5ish volts then boost convert to the required voltage. Seems the cost of all those bypass diodes is greater than 1 inductor, 1 transistor, 1 diode and a PIC Micro processor to control the transistor.

 

[prroots]

MS is right on the money (as usuall). If you consider each panel as being composed of 0.5 volt batteries (the individual cells) what happens if one of them goes dead (is shaded)and they are all in series? the whole panel has to push all the current through a dead cell.
What happens if they are wired in parallel? You loose the output of that one cell and amps go down a bit.
This is why bypass diodes are so important. they allow the dead cell to be bypassed with little effect. To be effective though you need a diode for each cell (not the blocking diode for each panel BTW)
There is a move afoot to wire all the cells in parallel for an output of 0.5ish volts then boost convert to the required voltage. Seems the cost of all those bypass diodes is greater than 1 inductor, 1 transistor, 1 diode and a PIC Micro processor to control the transistor.

Morningstar has advised me to wire the panels in series. According to them the best option depends on the type of shadowing. In the case where one panel is shaded and the other is not, then parallel is preferred. On the other hand if both panels have a little shading then series is preferred. In my case, it's not a case of one panel being shaded and the other not, so I'm better off in series. Series has the following advantages overall:

• Better overcast performance
• Earlier collection in day
• Later collection in day
• Better performance under light shading

As mentioned earlier, I would love to find a treatise on parallel vs series. It's certainly not the case where one is preferred in all cases.
Pete

 

[Bill Roosa]

Well it depends on how you panels are constructed and the actual shape of the shadow vs. the configuration of the bypass diodes. More diodes results in less effect of shading of any kind. if the shadow covers more than one "string" of cells (more than one bypass diode comes into play) then output suffers more. If only one "string" is shadowed then you only loose that string's output.
I did see a treatise on this subject when researching Mpp and boost-buck converters and as I recall they too indicated that it depended on how you where using / shading the panels and how the panels where constructed.

 

[prroots]

Here's a simple explanation of how blocking and bypass diodes function in conjunction with solar panels:

http://www.solar-facts.com/panels/panel-diodes.php

​

According to the following spec for my two model PW2050-210 panels, each panel has 4 bypass diodes:

http://www.photowatt.com/documents/produits/PW2050_EN.pdf
​

I assume each bypass diode covers 1/4 of the cells. Does that sound right? If I understand things correctly, one string could be shaded and I would still get 75% of the power out of that panel.

I am still trying to figure out whether to wire them in series or parallel. I've pretty much decided to test them both ways under similar levels of irradiance.
Pete

 

[joker460]

I would put them in parallel. That's what I did. 2 x 100w = 200 watts theoretical /12 = 16.6 amps Theoretical. Go for more power as opposed to more voltage.

Just my 2 cents worth

 

[prroots]

I would put them in parallel. That's what I did. 2 x 100w = 200 watts theoretical /12 = 16.6 amps Theoretical. Go for more power as opposed to more voltage.

Just my 2 cents worth

Thanks. With an MPPT controller things aren't so simple. The controller adjusts the voltage and current at output of panels to achieve maximum power regardless of whether in series or parallel. Series has a number of advantages such as:

• earlier startup
• later shutdown
• better overcast output
• lower cable voltage drop

Note: to realize the rated output of my panels they must be operated close to 26.4 Volts not 12 Volts. The issue that I'm faced with has to do with shadowing. I believe that will ultimately dictate the decision although if shadowing wasn't an issue, the obvious choice would be series.

As an update, we redid our wiring so that it will be easy to switch back and forth between parallel and series. If we can't figure it out on theoretically, we'll surely figure it out empirically
Pete

 

[joker460]

Mine is the Bluesky MPPT controller and I chose to go parallel because it was the best choice for my setup. Charges the battery to my needs while sailing.

Good Luck.

 

[Bill Roosa]

It all has to do with the actual shadows and the configuration of the by-pass diodes.
A long thin shadow over one string of cells all controlled by a single by-pass diode will show the same drop as a small shadow over one cell. If there is a small shadow that cover any part of two cells that are on two separate strings (two diode strings) then the drop will be double of the former example.

 

[prroots]

It all has to do with the actual shadows and the configuration of the by-pass diodes.
A long thin shadow over one string of cells all controlled by a single by-pass diode will show the same drop as a small shadow over one cell. If there is a small shadow that cover any part of two cells that are on two separate strings (two diode strings) then the drop will be double of the former example.

Interesting, but I suspect things are more complicated. I would theorize that the size of the shadow as a percent of cell size matters as does the distance between the object creating the shadow and the cell. Surely the shadow created by a piece of twine is not the same as a shadow created by a 1-1/2 inch wind generator mast even though both cast a shadow across the same cell(s). Also the same mast at a distance of 3 inches, 3 feet, and 30 feet, probably create a much different effect.

As an aside, I found post #27 by noelex 77 to be particularly insightful:

http://www.cruisersforum.com/forums...rallel-series-or-series-parallel-59513-2.html
​

He quotes in part:

...shade on one panel will reduce the power produced by that panel and the reduction will be a lot more than you would expect from the surface area of the panel. Other panels connected in parallel will be unaffected by the shaded panel and will still deliver there full power. Panels can only be connected in series with a MPPT regulator. Panels in full sun connected in series to a shaded or partially panel can be effected. Depending on the degree and amount of shade the power from the good panel can be drastically reduced, in some situations it is not effected much and in other situation the MPPT controller can extract slightly more power from the shaded panel when it is connected in series. In other words panels in series with a MPPT can produce much less power or a bit more. Once the voltage of the solar panels is well above the battery voltage there are no further possible gains from series connection and the risks of losses increases with every panel added in series. So there is no advantage in connecting more than 2 (nominally 12v) solar panels in series (There is a very slight possible gain from a reduction in wire losses, but this is small on the short runs found on a boat)

There are several significant points I took from this:

• Shading is not black and white; the depth of shading is important which really complicates the analysis
• Since my panels are 18 Volts nominal I may not see a significant improvement with a series connection since they are already well above battery voltage
• An MPPT controller can actually improve the performance of a partially shaded panel in series or drastically reduce output.

It seems to me that in the very complex shading situation you find on a typical sailboat, it's likely that the parallel situation will work best especially since my panels are already at a much higher voltage than batteries.
Pete

 

[prroots]

I found the following video to be interesting. It's a training video for the Morningstar Tristar MPPT 45 and 60 controllers. There are a number of interesting points raised if one has the stamina to sit through the entire seminar I especially liked the section on how the controller dealt with bumpy IV curves caused by shadowing:

http://www.morningstarcorp.com/en/support/item.cfm?ItemId=444
​

Pete

 

[prroots]

After redoing my wiring to make it easy to switch back and forth between parallel and series we finally have a clear sky for awhile. Here is a graph showing the power in both conditions. The lowest curve is array voltage which shows whether we're in series or parallel mode. One anomaly is that input power is lower than output power while in parallel mode; I'm guessing that this must be due to errors in the shunt resistors.We see that the spread between input and output power increases while in series mode which verifies that controller efficiency decreases with higher Voltage. Of course, we also have greater line loss while in parallel mode. Since the output power is what matters, we see hardly any difference between the two configurations. Note: there was no shading at the time on either panel.

Since the output power seems to be unchanged in either mode, the conclusion I draw is that the loss of controller efficiency at higher voltages associated with series mode is almost exactly offset by the increased line loss in parallel mode. Our controller is providing all the results you see

The boat is oriented so there is no shading on either panel. Once I complete these tests I will be investigating the effects on shading. That could get interesting; stay tuned
Pete

 

[Witzend]

Now would it be nice if someone would come up with a circuit that would change between series and parallel by testing the power output periodically to optimize. I guess that would be a bit anal, but possible. With a moving platform it is pretty difficult to determine what your shadowing will be.

 

[prroots]

Now would it be nice if someone would come up with a circuit that would change between series and parallel by testing the power output periodically to optimize. I guess that would be a bit anal, but possible. With a moving platform it is pretty difficult to determine what your shadowing will be.

Thanks. I was already thinking that a switch would be nice As it is I have to climb up and move the spade lugs around. It only takes me a couple of moments, but I'd rather not make it a habit. Our Morningstar TriStar MPPT 45 includes a comprehensive data logging capability which makes these graphs possible!
Pete

 

[Sumner]

If I'm reading that correctly it looks like you are getting up to around 280 watts. I'm think that isn't bad if the panels are not perfectly orientated towards the sun.

That would equate to somewhere in the neighborhood of 23 amps at 12 volts. We were usually getting about 9-12 amps during the mid-day out of 180 watts, so I feel you are doing well. We were still in bulk charge, but the battery was voltage was still up there. I don't think we ever had our batteries much under 60%. As the batteries started to top off the amps went down as they should. I think that the state of battery charge at any moment is going to throw a curve ball into the whole deal.

Interesting what you are doing. I don't have the patience for it . I just put as many panels up there as I can find space for and live with the results :cry:,

Sum

[FONT=Arial, sans-serif]Our Endeavour 37[/FONT]

[FONT=Arial, sans-serif]Our Trips to Utah, Idaho, Canada, Florida[/FONT]

[FONT=Arial, sans-serif]Our MacGregor S Pages[/FONT]

[FONT=Arial, sans-serif]Mac-Venture Links[/FONT]