PWM VS MPPT TROPICS

[Adelie]

Originally Posted by goboatingnow

Well pv are current sources rather then voltage sources so the panel voltage isn't that important anyway.

Dave

Photovoltaic panels are a source for electrical energy so both current and voltage are important, though the specifics of how they are handled can effect final efficiency and make one somewhat more important than the other.

Based on reading of latter posts I think what you are trying to say is that PVPanels are sources of constant voltage and the energy produced varies mostly by varying the current.

[btrayfors]

Originally Posted by Adelie

Photovoltaic panels are a source for electrical energy so both current and voltage are important, though the specifics of how they are handled can effect final efficiency and make one somewhat more important than the other.

Yes, absolutely.

Originally Posted by Adelie

Based on reading of latter posts I think what you are trying to say is that PVPanels are sources of constant voltage and the energy produced varies mostly by varying the current.

That, too, would be wrong. They are NOT sources of constant voltage. Their voltage output varies as a function of the light falling upon them, among other things (load, heat, etc.).

Bill

[Adelie]

Originally Posted by btrayfors

That, too, would be wrong. They are NOT sources of constant voltage. Their voltage output varies as a function of the light falling upon them, among other things (load, heat, etc.).

Let me rephrase, they are a source of approximately constant voltage. As light intensity, temp., etc vary, changes in the energy produced are reflected in slight changes in voltage and relatively much larger changes in current.

The thing that is not included in the "etc." is partial shading of the panel(s) which results in more significant changes in voltage.

[btrayfors]

Originally Posted by Adelie

Let me rephrase, they are a source of approximately constant voltage. As light intensity, temp., etc vary, changes in the energy produced are reflected in slight changes in voltage and relatively much larger changes in current.

The thing that is not included in the "etc." is partial shading of the panel(s) which results in more significant changes in voltage.

Nice try, but still missed the mark.

Take the 350-watt panel on one of my client's boats. It puts out 65VDC open circuit. But, in low light conditions (overcast, early morning, late day, etc.), the OC voltage drops precipitously. I've seen it as low as 20VDC.

IMHO it would be more accurate, based on real-world performance, to characterize solar panels as "a source of DC electrical energy transformed from light energy (usually the sun), which varies in both voltage and current output depending upon the amount of light falling on all its component cells".

Bill

[goboatingnow]

Originally Posted by Adelie

Let me rephrase, they are a source of approximately constant voltage. As light intensity, temp., etc vary, changes in the energy produced are reflected in slight changes in voltage and relatively much larger changes in current.

The thing that is not included in the "etc." is partial shading of the panel(s) which results in more significant changes in voltage.

No they are non ideal current sources. ( which I know many non EES struggle with the concept ) a battery on the other hand is a non ideal voltage source.

What I was trying to say in that is that the shading figures presented could be approximately evaluated using array current as panel voltage in itself is not a defining factor. So the figures indicated that the panels the OP had had very few bypass diodes or such diodes had failed ( which can very easily happen in series configuration. ) this would account for the near zero current that flowed , ie one panel was blocking the current in the other.

You don necessarily need panel watts to make comparisons ( within limits )
Dave

[s/v Jedi]

We have solar panels to power loads and to charge batteries. All the science on how to connect and which controller is about what a guy named Jacobi came up with in the 1840's so really this isn't new. Impedance matching and maximum power transfer are the key words. Start here: http://en.wikipedia.org/wiki/Maximum_power_theorem and don't forget to follow the links to impedance matching. I have been doing this sh/t for 40 years on HF which is a bit more complex than DC, but would not know how to start to explain it here.

In short: the solar array outputs maximum power into an unknown impedance value. An MPPT controller will measure PV power output (volt x amps) for a whole range of impedances that it presents to the array. From this info it establishes it's initial input impedance to achieve maximum power transfer. The thing is that this optimum impedance changes as the sun moves across the sky or when a cloud comes etc. The MPPT controller will keep varying the impedance a bit to see if either direction results in more power transfer, i.e. it tracks that point of maximum power transfer.

But, what is important is the power transfer during partial shading. I do concur that this might follow current drop somewhat. However, the tables provided have a constant voltage which is a fake measurement and also current to battery which is totally weird.

We just need to know power transfer for each shading level, stated in Watts. We also need to define partial shading. It does not mean partially shade every string of cells. It means to shade one or two strings while giving the other string(s) full exposure to the sun. Discussing how panels become shaded on boats and how to mount them is another thing.

If you want to test all this you need a dark room with an constant artificial light source. Some carton to cover strings etc. But this has been done to death by the manufacturers which is what they based their cell & string configurations with by-pass diodes on. A bit silly to not believe them and test it again.

[neelie]

Originally Posted by s/v Jedi

We just need to know power transfer for each shading level, stated in Watts. .

Jedi... sorry to pick out just one line of your post, but below is a question that relates to it.

Would I be wrong in assuming that the voltage out of his MPPT controller is approximately the same for both the series and parallel cases? (i.e. 14.4- ish?)

If this is so, then surely the charge amps as listed by SL are directly proportional to the power transferred.

And if this is so, I don't understand the 15 fold difference between the power transferred in the series and parallel cases in the 1/3 shade test, its simply too good (or bad) to be true.

[goboatingnow]

Originally Posted by neelie

Jedi... sorry to pick out just one line of your post, but below is a question that relates to it.

Would I be wrong in assuming that the voltage out of his MPPT controller is approximately the same for both the series and parallel cases? (i.e. 14.4- ish?)

If this is so, then surely the charge amps as listed by SL are directly proportional to the power transferred.

And if this is so, I don't understand the 15 fold difference between the power transferred in the series and parallel cases in the 1/3 shade test, its simply too good (or bad) to be true.

Yes that's what I was saying , panel current can be used as an approx guide to what's happening , what having watts would do is simply to account for the non linearity of PV cells at the extremes and to compute power transfer efficiency. But in reality that's not what's being done here all that is being done is to evaluate shading , I mean you could just replace all the mppt and batteries with a resistor and that would work for a comparison.

While the results are quite dramatic and ignoring voltages which may be suspect. , the situation is consistent with a panel with few or no bypass diodes in series and in shade.

Dave

[Adelie]

Originally Posted by btrayfors

Nice try, but still missed the mark.

Take the 350-watt panel on one of my client's boats. It puts out 65VDC open circuit. But, in low light conditions (overcast, early morning, late day, etc.), the OC voltage drops precipitously. I've seen it as low as 20VDC.

IMHO it would be more accurate, based on real-world performance, to characterize solar panels as "a source of DC electrical energy transformed from light energy (usually the sun), which varies in both voltage and current output depending upon the amount of light falling on all its component cells".

Bill

Through the bulk of it's performance envelope, the voltage should remain approximately constant. At the very low end slight differences in temp and manufacturing tolerances will cause individual cells or strings of cells to drop out as the diodes kick in at different times and that's where you see the big voltage drop off.

[sailinglegend]

This has been a very interesting if somewhat emotional topic. As yet I am the only one who has done any series/parallel shading tests. Please can someone else try their panels so that this topic can try and reach a conclusion. Because of all the “opinions” expressed here I have repeated my tests this morning at 0900 hrs MED time.

With my two Solara M-series 68 W semi-flexible panels into a Morningstar MPPT Sunsaver controller there is no difference to my original conclusion that Parallel connection is a dramatic winner over Series connection. Depending on the amount and angle of shading on one panel the output of the two in series is reduced to at worst to near zero, when in parallel the output was reduced to a half. I do however have some comments to add from today’s testing. My figures have all come from the Morningstar software, via a series interface to my computer. I have tested the output diodes on both panels are they work correctly.

This time I have tested both panels individually to see their shading performance – both were exactly the same. With a lower early morning sun there was a drop in the Array voltage with different shadings as the MPPT did its job of tracking the maximum power point tracking voltage. Yesterdays midday sun it was pretty constant. There was a slight drop in the battery charging voltage but this is to be expected, it is the current that's important.

Today I have adjusted my shading from the simple 1/3 to 3/3 to have more 1/3 sections and hand/arm shades on different parts of the panels. It is very clear that the output from one panel can be dramatically reduced from 2.8 amps to zero with only a small shadow in the wrong place. In contrast a 1/3 shading across another part of the panel may produce only half the battery charging current. I have also noted that the array power in all cases is proportional to the battery charging current. As others have said this is all you need to measure, amps into the battery for so many hours = Ah. When both panels were connected in Series the output charging current fell to zero with the wrong kind of shadow on just one panel. In Parallel the same shadow dropped the charging voltage by half. These tests have taken several hours this morning.

I have always pushed for Series connection when several panels are connected to one MPPT controller. This way the current in series is much lower than in parallel, so smaller cables can be used, or the same cables can take double the wattage of solar panels. But if your panels occasionally reduce the output to zero then you are better off in parallel - with maybe more power loss on the cable than you would have like.

You have to do the tests yourself with your panels and your cabling on your boat and decide if parallel is better than series for your situation.

[neelie]

Originally Posted by goboatingnow

Yes that's what I was saying , panel current can be used as an approx guide to what's happening , what having watts would do is simply to account for the non linearity of PV cells at the extremes and to compute power transfer efficiency. But in reality that's not what's being done here all that is being done is to evaluate shading , I mean you could just replace all the mppt and batteries with a resistor and that would work for a comparison.

While the results are quite dramatic and ignoring voltages which may be suspect. , the situation is consistent with a panel with few or no bypass diodes in series and in shade.

Dave

Mate, you've really lost me here.. just explain in simple English for a dummy like me and especially avoid terms like "non linearity"....

(I'm not trying to score any points here - I'm just trying to understand whats going on..and I'm still swimming in glue here, my engineer friends)

Were my assumptions approximately right in the real world or not?
Especially regarding the output voltage of SL's MPPT controller?
And that the validity of stating that the charge amps are directly proportional to the power produced?

As I stated, I plan on having 2 x 240W panels with an MPPT controller - so this is crucial that I get a bit of grip on this topic.

[deckofficer]

What I got out of this thread is what I thought going in, correct me if I'm wrong.

For the best bang for the buck, go with 60 cell, 30 volt residential panels since they are down to $1 per watt, run parallel if more than the 280 watts typical of the bigger panels is needed, and use a MPPT controller that allows user set points and as Jedi explained, matches impedance for all the variables of shading, heat, and exposure angle.

Do I have it right?

[s/v Jedi]

Originally Posted by deckofficer

What I got out of this thread is what I thought going in, correct me if I'm wrong.

For the best bang for the buck, go with 60 cell, 30 volt residential panels since they are down to $1 per watt, run parallel if more than the 280 watts typical of the bigger panels is needed, and use a MPPT controller that allows user set points and as Jedi explained, matches impedance for all the variables of shading, heat, and exposure angle.

Do I have it right?

Pretty much. Depending on how/where you mount the panels, it might be better to put two in series for 60V. This only makes sense for 4 or more panels like 2x2, 3x2 etc. setup. With two panels, one will probably be on one side of the boat and the other on the other side and I would not put them in series in that case. Mark @ Reach had that explained earlier on.

p.s. I use the Outback Flex-something controller, which is the best... imho of-course.

ciao!
Nick.

[deckofficer]

Originally Posted by s/v Jedi

Pretty much. Depending on how/where you mount the panels, it might be better to put two in series for 60V. This only makes sense for 4 or more panels like 2x2, 3x2 etc. setup. With two panels, one will probably be on one side of the boat and the other on the other side and I would not put them in series in that case. Mark @ Reach had that explained earlier on.

p.s. I use the Outback Flex-something controller, which is the best... imho of-course.

ciao!
Nick.

Nick,

You know my plans are for electric propulsion at 48 VDC and of course a 48 VDC inverter for AC loads, so I'd probably go with a 3 series string and parallel the strings for desired wattage.

How is the Outback on the RFI front?

[s/v Jedi]

Originally Posted by deckofficer

How is the Outback on the RFI front?

I must still try that... little sun today, only got 2 kWh out of the array...