Strange strange solar panels.

[deepakpandey2403]

Solar panels of different make behave in weird ways. I hope you'll all agree with me by the end of this thread.

Here's my observation (and I would appreciate if somebody could correct my errors)
1) I have an inverter (off grid, running MPPT) which has been tested around with different make of cells. Its a 2.2kW inverter and i have hooked up a 3kW configuration to it. Now when there is a sudden change in weather (cloud passing by , rains etc) , different responses have been observed with different makes.
a) sudden change in intensity means sudden drop in cell voltage, this will cause the inverter to trip in some case, while in others it gives sufficient time for the algo to work and gradually reduce the voltage(and current) intake (2-3 seconds etc) from the inverter.

Anyone who can put a light on this property of different makes and also a possible solution to make my algo independent of the same.

I would appreciate if somebody could share some good out-of-box solar PV character document/link/ref etc.

Thanks

[inetdog]

Solar panels of different make behave in weird ways. I hope you'll all agree with me by the end of this thread.

Here's my observation (and I would appreciate if somebody could correct my errors)
1) I have an inverter (off grid, running MPPT) which has been tested around with different make of cells.
Anyone who can put a light on this property of different makes and also a possible solution to make my algo independent of the same.

I would appreciate if somebody could share some good out-of-box solar PV character document/link/ref etc.

All PV cells to a greater or lesser degree will deliver a constant open circuit voltage over a wide range of incident light. But the current available from them and the position of the Voltage versus Current graph will change as the light changes. This may result in the voltage dropping if the panels cannot supply all of the power the load needs, and this in turn reduces the amount of power available from the panels even more than just the drop in current would.

Well, this is one reason that all (except one line from SMA) GT inverters will not run off grid and off grid inverters will not run without batteries.

If you used an MPPT charge controller to feed batteries which fed the inverter, the fluctuation in solar PV output would have little immediate effect on the off-grid inverter.

Basically you cannot easily use a grid tie inverter off grid because there will not be a load capable of absorbing the full but varying output of the panels. The main exception is when you use the panels to directly drive a pump motor which has been designed for this kind of use.

And you cannot use an off grid inverter which tries to supply whatever the loads try to draw without a battery or equivalent energy storage of some kind.

More info please....

[deepakpandey2403]

All PV cells to a greater or lesser degree will deliver a constant open circuit voltage over a wide range of incident light. But the current available from them and the position of the Voltage versus Current graph will change as the light changes. This may result in the voltage dropping if the panels cannot supply all of the power the load needs, and this in turn reduces the amount of power available from the panels even more than just the drop in current would.

Well, this is one reason that all (except one line from SMA) GT inverters will not run off grid and off grid inverters will not run without batteries.

If you used an MPPT charge controller to feed batteries which fed the inverter, the fluctuation in solar PV output would have little immediate effect on the off-grid inverter.

Basically you cannot easily use a grid tie inverter off grid because there will not be a load capable of absorbing the full but varying output of the panels. The main exception is when you use the panels to directly drive a pump motor which has been designed for this kind of use.

And you cannot use an off grid inverter which tries to supply whatever the loads try to draw without a battery or equivalent energy storage of some kind.

More info please....

Hey inetdog

Thanks for the reply. Two things here
Number one : I am aware of PV cells I-V-P characteristics, Temp-P characteristics.
Number Two: I myself has worked on off-grid inverters w/o any battery system and has almost achieve the ideal MPPT characteristics with some of the cells. So, there exists an inverter in the market throughout the world, on which I have worked on, which runs w/o any battery system.
I have seen other inverters as well w/o battery system.
P.S : My discussion is on off-grid inverter which normally do not extend above 22kW rating. Sorry if i had not mentioned earlier.

Please counter me , it will be appreciated. No offence meant in any way.

Regards
Deepak

[inetdog]

Hey inetdog

Thanks for the reply. Two things here
Number one : I am aware of PV cells I-V-P characteristics, Temp-P characteristics.
Number Two: I myself has worked on off-grid inverters w/o any battery system and has almost achieve the ideal MPPT characteristics with some of the cells. So, there exists an inverter in the market throughout the world, on which I have worked on, which runs w/o any battery system.
I have seen other inverters as well w/o battery system.
P.S : My discussion is on off-grid inverter which normally do not extend above 22kW rating. Sorry if i had not mentioned earlier.

Please counter me , it will be appreciated. No offence meant in any way.

Regards
Deepak

The main thing which I can think of that could cause an unexpected voltage drop even though the inverter automatically reduces its current demands would be partial shading on the panels in the series string. If the panels are not built with bypass diodes, the current output of the entire string will be reduced to that of the most shaded panel. This can have the side effect of severely dropping the output voltage at Imp, under your algorithm.
If the panels are equipped with bypass diodes, then you will see little reduction in the string current, but the string voltage will sharply drop by the total Vmp contribution of the shaded sections, plus an additional loss of up to 1 volt for each bypass diode that is activated.

in this respect you will see a great difference between the behavior of a mono or polycrystalline silicon panel versus a thin film amorphous panel.

[deepakpandey2403]

The main thing which I can think of that could cause an unexpected voltage drop even though the inverter automatically reduces its current demands would be partial shading on the panels in the series string. If the panels are not built with bypass diodes, the current output of the entire string will be reduced to that of the most shaded panel. This can have the side effect of severely dropping the output voltage at Imp, under your algorithm.
If the panels are equipped with bypass diodes, then you will see little reduction in the string current, but the string voltage will sharply drop by the total Vmp contribution of the shaded sections, plus an additional loss of up to 1 volt for each bypass diode that is activated.

in this respect you will see a great difference between the behavior of a mono or polycrystalline silicon panel versus a thin film amorphous panel.

Hey inetdog

Thanks for the answer. Helps ! I did not get your bypass diode concept though. The drop(of 1V, or 0.7V ) is understood, but could you throw some more light on bypass diodes in PV panel.

Thanks

[inetdog]

Hey inetdog

Thanks for the answer. Helps ! I did not get your bypass diode concept though. The drop(of 1V, or 0.7V ) is understood, but could you throw some more light on bypass diodes in PV panel.

Thanks

A set of cells that is shaded will produce a lower short circuit current than ones that are in full sun. If you try to push more current through those shaded cells they will act as a high resistance and may be damaged by the heating.
For this reason a diode is connected internally (or in the j-box) across every 10 or 20 volts worth of cells. This diode is back-biased (non-conducting) when the cells are producing power, but get forward biased and conducts current around the cells when the external circuit is trying to push more current through than those cells can handle at the moment. Hence the name bypass diode.
For DIY panels, there are usually two or three diodes (or pairs of diodes) mounted in the junction box, and connected to bus wires from the middle of the cell stack (two diodes) or the 1/3, 2/3 points (three diodes) and to the overall + and - leads. If there are no bypass diodes, the panels will behave very badly when partially shaded and may actually be damaged.

[deepakpandey2403]

I do no tbelieve a word you are saying.

That's what they said to Galileo.

Anyway, I am not here to check your understanding or to prove myself. But try doing it yourself.
"MPPT inverters without battery runs fine."
If you have trouble understanding, ask!

No offence meant to anyone.

[Mike90250]

...."MPPT inverters without battery runs fine."
If you have trouble understanding, ask! ...

So what we have here is a failure to communicate.
MPPT is a power harvest maximizing function
Inverters convert DC to AC.

If you take the DC provided by a MPPT device, and feed it into an inverter, you need two more things:

1) a specially designed inverter to limit it's input consumption to less than what the MPPT can provide,
preventing collapse of the MPPT and solar array. (most inverters rely on a battery to provide surge response) This must anticipate cloud events.

2) inverter smart enough to not over voltage it's output. (GT inverters rely on the Grid to do this)

Both of these exist, but not in the same package, AFIK. SunnyBoy has a model that can provide
a limited AC output of about 10% of the PV array, but a brief cloud event will shut that down. If a fridge is running, the compressor will likely stall at restart.

How about if you give us the Mfg & Model of this magical device that only you have heard of. Or else you are running a system with only a few % usage of PV array . (50 w load, 500w PV)

[deepakpandey2403]

So what we have here is a failure to communicate.
MPPT is a power harvest maximizing function
Inverters convert DC to AC.

If you take the DC provided by a MPPT device, and feed it into an inverter, you need two more things:

1) a specially designed inverter to limit it's input consumption to less than what the MPPT can provide,
preventing collapse of the MPPT and solar array. (most inverters rely on a battery to provide surge response) This must anticipate cloud events.

2) inverter smart enough to not over voltage it's output. (GT inverters rely on the Grid to do this)

Both of these exist, but not in the same package, AFIK. SunnyBoy has a model that can provide
a limited AC output of about 10% of the PV array, but a brief cloud event will shut that down. If a fridge is running, the compressor will likely stall at restart.

How about if you give us the Mfg & Model of this magical device that only you have heard of. Or else you are running a system with only a few % usage of PV array . (50 w load, 500w PV)

Well certainly a communication gap here. Ofcourse MPPT is Max. Power Point Tracking (if we see the PV cell I-V-P char, we will be talking more sense here) and an inverter , is an inverter, the name says it all. By the way thanks for pointing the obvious.
Well, the magical system which the US or Turkey might not have seen by now is developed by ABB India (I am not marketing here, since you asked I am mentioning here). I work in ABB and was in a team which dev the MPPT. It is for the European, the African and the Mexican markets where there is a demand.When you said Inverters are not smart, what do you mean. You have DSP running the most complex algorithms to govern the switching of IGBT's in a way to get an almost pure sine wave. You have thousands of protection inbuilt to protect your pump. You have Fieldbus written on it. What else do you expect. I would never expect my inverter to walk away from its installation in order to show smartness.

Note1: If you know what a perturb and observe in MPPT is , you wouldn't have talked about battery in first place.
Note2: I have not seen Japan, but I still believe that Japan exists, if you know what i mean.

Cheers!

[inetdog]

Well certainly a communication gap here. Ofcourse MPPT is Max. Power Point Tracking (if we see the PV cell I-V-P char, we will be talking more sense here) and an inverter , is an inverter, the name says it all. By the way thanks for pointing the obvious.
Well, the magical system which the US or Turkey might not have seen by now is developed by ABB India (I am not marketing here, since you asked I am mentioning here). I work in ABB and was in a team which dev the MPPT. I would never expect my inverter to walk away from its installation in order to show smartness.

I guess that the product is somewhat hard to find. When I search through the AAB India products on the web, I see only grid tie inverters.

– these transformerless, single-phase inverters are suitable for small and medium-size photovoltaic systems connected to the public electricity network.

Can you be more specific without providing an actual link?
Thanks.

[russ]

Well, the magical system which the US or Turkey might not have seen by now is developed by ABB India

ABB India ? Develops nothing - thy have to pick up the phone and call Europe to find out what day it is.

[deepakpandey2403]

I guess that the product is somewhat hard to find. When I search through the AAB India products on the web, I see only grid tie inverters.

Can you be more specific without providing an actual link?
Thanks.

Hello
The grid tie inverters (PVS300 and PVS800) are generally deal with higher ranges (coupled to generate MW range). For domestic purpose (off grid solar pump), we are in testing (beta) phase and hence has not been put on the site. However, within a month or so , it will be available as a product for commercial use and will be available on site.

[deepakpandey2403]

ABB India ? Develops nothing - thy have to pick up the phone and call Europe to find out what day it is.

FYI, not to be harsh , ABB India has the largest R&D facility of ABB Group outside Europe contributing to $4.5bn revenues to the group in an year. A significant amount of group assigned revenues for R&D are allotted here and a good amount of business generated.
For instance, ABB Traction devices were developed from scratch in ABB India - Bangalore facility and then the manufacturing outsourced to ABB Germany.
All major software developments and testing are done here.

As for the matter of asking the day and time is concerned, ABB Europe (and for you as matter) are still behind time, so when we write codes here, you guys are still in dreaming mode.

I have always wondered why people have been so ignorant about the silicon valley of south Asia.

[SunEagle]

25 years is a pretty long time . I really wonder how slow a learner you are. If I was in Turkey/ 'dealing' with Turkey for 25 long years, I would have been knowing in and out of its history. BTW, patience is the virtue. Wait till you actually see the product.

My thread here was on PV panels but strangely, no body is able to explain me why the problem put here actually happens. People dealing for 25 years do not make sense now.

Let's not discuss what happens to India or Turkey or the US. Let's make the planet greener by answering my question to the point. Had you been knowing the answer, the answer would have come in the first reply.

OK. Getting back to your original question which I believe was, why do different pv manufactured panels give different results for the same sun conditions?

Forgetting that you are using a "special" inverter to measure each type panel I would say the different outcomes are based on the materials of each of the panels or how they are made. Some manufacturers do not use the same quality materials which could be a reason their product does not perform as well as another manufacturer. Another reason could be the quality control of the finished panel is not the same for each manufacturer.

Just because a panels nameplate information is similar to another does not mean they will really perform the same under real life conditions.

Since you really didn't provide specific differences between each product you tested I can only presume that some panels performed better than others but may still fall within the manufactures specification.

[FloridaSun]

25 years is a pretty long time . I really wonder how slow a learner you are. If I was in Turkey/ 'dealing' with Turkey for 25 long years, I would have been knowing in and out of its history. BTW, patience is the virtue. Wait till you actually see the product.

My thread here was on PV panels but strangely, no body is able to explain me why the problem put here actually happens. People dealing for 25 years do not make sense now.

Let's not discuss what happens to India or Turkey or the US. Let's make the planet greener by answering my question to the point. Had you been knowing the answer, the answer would have come in the first reply.

hahahaha! you're funny, coming on, claiming you're a 'programmer' who works for ABB, has a new 'magic box' inverter from that company but... has no technical back up from the company and looking for answers.. and how did you acquire this new design, I wonder ? Or is the tech/RD dept at ABB so sad they have to look online for development answers? I'm still waiting to see the product. Also wondering why you're connecting 3Kw of panels to a 2.2Kw inverter. You've given only obtuse evidence about this grand new invention soon to come... (or most likely never be). Only certain thing I understand is that it doesn't quite work as you expected.
"...good out-of-box solar PV character document/link/ref etc." is available online or from the manufacturers and can be googled easy enough. Why do you come here for that answer?