PWM VS MPPT TROPICS

[BriRich]

Hi Gals and Guys

It's Joie and Brian the OP's. Our decision is the BlueSky SB2000E 25a MPPT.

Our panels are:

we have two panels 130W X 2 = 260W, these are panels I bought off ebay that have Bosch mono elements.


Cells size (mm). 156X130.57
Module Size(mm) 670X1245X35
Approximately Power Tolerance +/- 3%
Nominal Pick Power (WP) 130 watts
Nominal Voltage (V) 18 volts Nominal
Current (A) 7.22
NOTC 45+/- -2%/C
Voltage Temperature Coefficient –0.33%/C
Current Temperature Coefficient +0.05%/C
Power Temperature Coefficient -0.23%/C
Open Circuit Voltage (Voc) 21.6
Short Circuit Current (lsc) 7.8
Conversion Efficiency 17.56%
Operating Temperature -40 to +85C
Max System voltage 715 DC

We plan on +300ah of flooded cells batts. We decided on the MPPT vs PWM as we do not want to take up any more space with panels and want to get as much output as possible from the panels we have. Also while the tropics are hot there are plenty of rain days and seems the MPPT is more versatile in charging in different conditions. Palau gets 3 meters of rain a year.


We're stoked with all the debate going on we're gonna try parallel hook-up and see how it goes.


Cheers Gals and Guys


Brian & Joie

[sailorchic34]

Originally Posted by s/v Jedi

Hmm yes but now you are shading 1 panel out of 3 panels, which is NOT partial shading.


Do the example for 1 panel full sun and two that are partially shaded, let's say one of their two strings are shaded. In parallel, you loose 2/3 of the output because both partially shaded panels loose half their output voltage so can charge at 13-14V anymore.

But the series setup is happily charging away with only a 1/3rd reduction in output. Pretty cool huh

Hum.. Really it depends on if there is one bypass diode per panel or one diode per string. Looking at kyocera panels it looks like they have one bypass diode per panel on some and on others there is a bypass diode per string. Where the bypass diode is one per panel you loose the panel weather serial or parallel.

Ok OK, if and its a big IF you have panels with blocking diodes on seperate strings, then series might be better. However, I'm having a hard time coming up with manufacturers data that shows that. Kyocera on their installation instructions indicates that there is a bypass diode per panel. So weather wired parallel or serial its the same drop in the system. Well with a 1/2 volt drop more in serial, due to the bypass diode.

The Bypass diodes function is to protect a series string of panels from being pulled down. They do nothing in a parallel wired panel installation.

Here is the link to the kyocera installation instructions: http://www.kyocerasolar.com/assets/001/5155.pdf

For land base systems with a clear site, that is no shading, series wiring is superior as the wire size is smaller for each series run. On a sailboat at anchor there is always some shading for part of the day..

As I stated before, most shading issues with panels on boats, cause a shade line across the width of the panel, there by pulling down both strings where there is a bypass diode on each string.

In most cases with shading on a boats PV array, you loose the panel irregardless of weather the panels are wired serial or parallel. In serial you get an added .5V per panel drop (or 1V depending on construction) in serial that you don't get in parallel..

So both methods offer pluses and minuses. From my standpoint Parallel has fewer failure points. This as a loss of a bypass diode does not pull the whole array down in parallel. But that's just me.

BTW Jedi, thanks for a delightful discussion.. Its good to get my poor brain working once and a while.

Serial does offer the chance of higher MPPT conversions so that's a plus in the serial column. I still feel that the cost of the typical Mppt controller would be better spent adding one more panel to the array. It would seem you would get more amps per dollar invested that way.

But to each there own....

[s/v Jedi]

I have the old Shell/Siemens panels that are not sold anymore but have 3 bypass diodes iirc. We had a thread like this one a year or so ago and all the current panels were made up of multiple strings. I don't think they sell panels with just one bypass diode.

here's more: http://www.solaredge.com/files/pdfs/...in_shading.pdf

cheers,
Nick.

[sailorchic34]

Originally Posted by s/v Jedi

I have the old Shell/Siemens panels that are not sold anymore but have 3 bypass diodes iirc. We had a thread like this one a year or so ago and all the current panels were made up of multiple strings. I don't think they sell panels with just one bypass diode.

here's more: http://www.solaredge.com/files/pdfs/...in_shading.pdf

cheers,
Nick.

Oh power optimizers on panels are cool. The solar edge folks are selling MPPT power optimizers for each panel. What the optimizer does is optimize the output of each panel taking into account shading effects of that panel to maximize the output of each panel. These outputs then are taken to another MPPT controller for the whole array. Way cool. For those with a lot of money, using a power optimizer on each panel gives you the most output per panel.

Well I was just quoting from the kyocera instructions, which indicate one diode per panel. Really its hard to find out just how many diodes are in each manufactures panels.

Still be it 1 or 3 bypass diodes, on a sail boat, 90+ percent when there is shading, the shading will be full width across a panel, so all bypasses on a panel will be in play. Plus with three diodes in series your getting a 1.5V drop across the shaded panel in series. That's 1.5V drop to the rest of the array. I'm thinking sometimes a single bypass diode per panel would work better for sailboats.

Still you are right that with a partial shade over one string of cells in a panel, you will get more output in series. But only for when there is shading on just one cell string. Once the shade crosses all strings of cells, parallel would be better as there is no additional voltage drop in a parallel array.

So best I can come up with is we're both right, part of the time OK My brain hurts so I need to go play for a while...

Still love the idea of the panel optimizers...

[noelex 77]

The 36cell Kyocera panels in the most used boat sizes have 8 bypass diodes.
Most panels have less. 2-3 is common. The individual cell will be damaged by overheating if there are no bypass diodes so less than 2-3 is not common.
If wiring in series a bypass diode around the whole panel is sometimes installed.

[s/v Jedi]

Originally Posted by noelex 77

The 36cell Kyocera panels in the most used boat sizes have 8 bypass diodes.
Most panels have less. 2-3 is common. The individual cell will be damaged by overheating if there are no bypass diodes so less than 2-3 is not common.
If wiring in series a bypass diode around the whole panel is sometimes installed.

There you go, 8 diodes in a Kyocera; I guess 6 strings and an extra diode bypassing 3 strings in one go.

Partial shading is why panels must be mounted in a fore-aft direction.

[Nicholson58]

Originally Posted by BriRich

I am refitting my new to me Cape North 43. I have a civil engineering degree but lord help me if I can figure the electrical stuff...all the acronyms .

There's some exhaustive info here on CF on MPPT vs PWM and a lot of acronimonimiis (acronyms) that us dirt engineers need more schooling on.

Thanks & Peace

Brian

HandyBob's Blog « Making off grid RV electrical systems work I consider the complete reading of every long-winded discussion on Handy Bob's site the pre-requisite to any solar job on a boat. Bob lives off the grid in the SW and primarily deals with motor homes, a pretty fair comparison. We boaters will tend to make decisions based more on keeping the equipment footprint small and perhaps pay a bit more to do so.

He discusses MPPT and PWM. THe short of it is that MPPT is used when the voltage out from your panels is well above your battery charge voltage. I my case, my Sanyo panels are 54 volts open circuit. The MPPT can repacakge that energy into lower voltage and higher amps. As the sun goes down and panel voltage drops, the MPPT continues to charge as long as the pnael voltage is above battery charge voltage. I'm sure that the technology marches on and better stuff will be available every year. Mine is two years old and probably obsolete.

[btrayfors]

Great discussion, folks. Thanks for all the insights.

Don't forget, there's shading and then there's shading. A panel with a high PV to start with....say 65VDC open circuit like the 350 watt panel on one of my current client's boat, can put out usable volts (and amps) even when the ambient light is very low, like dawn, sunset, storm conditions, etc.

Don't forget, in sunny conditions there's also light reflection, e.g., from the decks up to the bottom of the panel and down again even in the "shaded" spots. Some panels now can use light from the bottom side as well as the top side.

It would be interesting to do some simple tests to measure the output of the new panels when part -- or all -- of the panel is "shaded".

Bill

[sailinglegend]

Originally Posted by btrayfors

.....It would be interesting to do some simple tests to measure the output of the new panels when part -- or all -- of the panel is "shaded".

Bill

Interesting post as I have just upgraded to a Morningstar MPPT controller and switched to series connection with two 68 watt fairly new Semi-flexible Solaras. The MPPT is so much better than the Morningstar PWM.

So I've done some tests and it very clearly shows that parallel is better than series - with any kind of shading - certainly with my panels. I have very thick wires to the regulator so there was no difference in the charge current in series.

To put some figures on the test - in Turkey at 1130 hours constant sun, all loads disconnected, battery 91% charged, still in absorption mode, all the shade applied to only one panel. The same figures were obtained by shading the other panel only:

SERIES:.............No Shade-3/3 shade-2/3 shade-1/3 shade-Hand/arm
Charge current....4.85........0.11.........0.16.........0 .16.........1.66
ARRAY VOLTS.....32.88.......32.88.......32.88.......32.8 8........32.88

PARALLEL:..........No Shade-3/3 shade-2/3 shade-1/3 shade-Hand/arm
Charge current....4.95........2.45.........2.53.........2 .58.........3.4
ARRAY VOLTS.....15.33.......15.33.......15.33.......15.3 3........15.33

You notice a slight increase in the parallel connection currents with no shade as the panels are mounted on a curved deck. The more the sun comes around and sinks the less output the port panel gives.

All the theory we have had is all very misleading to most people. I have a degree in Electrical and Electronic Engineering, and I was giving up, so I would urge everyone to try their own tests. I will do the tests again in cloudy conditions and see if there is a difference, but the above result are dramatic. It confirms what appears to be the thoughts of most people that if there is any inbalance, even 5%, between two or more panels, like their output wattage or panel voltage, then they should be wired in parallel.

[s/v Jedi]

Originally Posted by sailinglegend

Interesting post as I have just upgraded to a Morningstar MPPT controller and switched to series connection with two 68 watt fairly new Semi-flexible Solaras. The MPPT is so much better than the Morningstar PWM.

So I've done some tests and it very clearly shows that parallel is better than series - with any kind of shading - certainly with my panels. I have very thick wires to the regulator so there was no difference in the charge current in series.

To put some figures on the test - in Turkey at 1130 hours constant sun, all loads disconnected, battery 91% charged, still in absorption mode, all the shade applied to only one panel. The same figures were obtained by shading the other panel only:

SERIES:.............No Shade-3/3 shade-2/3 shade-1/3 shade-Hand/arm
Charge current....4.85........0.11.........0.16.........0 .16.........1.66
ARRAY VOLTS.....32.88.......32.88.......32.88.......32.8 8........32.88

PARALLEL:..........No Shade-3/3 shade-2/3 shade-1/3 shade-Hand/arm
Charge current....4.95........2.45.........2.53.........2 .58.........3.4
ARRAY VOLTS.....15.33.......15.33.......15.33.......15.3 3........15.33

You notice a slight increase in the parallel connection currents with no shade as the panels are mounted on a curved deck. The more the sun comes around and sinks the less output the port panel gives.

All the theory we have had is all very misleading to most people. I have a degree in Electrical and Electronic Engineering, and I was giving up, so I would urge everyone to try their own tests. I will do the tests again in cloudy conditions and see if there is a difference, but the above result are dramatic. It confirms what appears to be the thoughts of most people that if there is any inbalance, even 5%, between two or more panels, like their output wattage or panel voltage, then they should be wired in parallel.

It looks like your test is flawed somehow. At 1/3 shading the series connection wins. At no shading, the series connection has more than double the output from parallel connection. The 2/3 and 3/3 tests would indicate no bypass diodes in your panels, but the 1/3 test invalidates that.

Your test does show that you loose 50% with partial shading in parallel... Kind of.

[sailinglegend]

Originally Posted by s/v Jedi

It looks like your test is flawed somehow. At 1/3 shading the series connection wins. At no shading, the series connection has more than double the output from parallel connection....

You have totally misread the results table.

At 1/3 shading the charge current is 0.16 amps in series, and 2.58 amps in parallel!!!!!!

With partial shading with one arm/hand - in series the charge current is 1.66 amps and parallel is 3.4 amps. Sorry I didn't include the word "amps" in my table, but charge current seemed to be enough.

The double output you refer to is actually the double input array volts when the panels are in series.

You just seem to have shot yourself in the foot - that's a British saying for......

How can we believe anything else you have written in this thread?

[daddle]

Ahhh ... Just get some reasonable 18v panels, wire them in parallel with a controller, maybe MPPT, and go cruising. It's what everybody is doing. Try to keep the panels in the sun and the beer in the fridge.

[s/v Jedi]

Originally Posted by sailinglegend

You have totally misread the results table.

At 1/3 shading the charge current is 0.16 amps in series, and 2.58 amps in parallel!!!!!!

With partial shading with one arm/hand - in series the charge current is 1.66 amps and parallel is 3.4 amps. Sorry I didn't include the word "amps" in my table, but charge current seemed to be enough.

The double output you refer to is actually the double input array volts when the panels are in series.

You just seem to have shot yourself in the foot - that's a British saying for......

How can we believe anything else you have written in this thread?

Okay then, I see that you added an extra column then. What is that extra column? I see that you partially shaded each string... you should only shade 1 string so that the others can work. I am starting to think you mounted the panels sideways on your boat... or a curved deck?


On electrics you need some basic understanding: the power output is voltage times current.


This means that your parallel output unshaded is 4.95 x 15.35 = 76 Watt
This means that your series output unshaded is 4.85 x 32.88 = 160 Watt

The series output is more than double your parallel output.

You have two 68W panels which makes that 136 Watt. Which means both numbers above are wrong.

Now, tell me who they will believe in this thread? How can you have a degree in Electrical and Electronic Engineering but are unable to calculate power output?! I call your post a hoax and BS

Really, I'm not the only engineer on CF... why do you even try?

[neelie]

Originally Posted by sailinglegend

All the theory we have had is all very misleading to most people. l.

And certainly beyond my meagre intellect.

So, SailingLegend, would your results not depend on how many blocking diodes and their arrangement in each of your panels? I am trying to reconcile why your results are so clear cut - in other words, are your test results correct only for your brand/model of panel? .

In series, it seems with "only" 1/3 shade, the total output is reduced to almost zero. On a sailboat, shade happens, and a series connection will be catastrophic.

I plan on installing 2 x 240W and am trying to figure out which is the better option. Parallel or Series?.

Your tests show this decision to be critical, yet the previous debate indicated that the expected differences should have been somewhat less dramatic.

[noelex 77]

Firstly thanks sailinglegend for doing those measurements.

I assume you are quoting the amps into the batteries after the voltage conversion from the MPPT.
I think Jedi assumes you are quoting the amps from the panel, upstream of the MPPT controller. Perhaps if you could clarify it would help.

I must admit I find the numbers surprising. I expected a much smaller difference.
I wonder if there is a fault with the bypass diodes in your panels (they do sometimes blow), or some other fault of experimental design.

Finally could everyone on this thread please remember we are taking about solar panels. There is no need for personal comments.