Solar panels configuration help required.

[Anachem20]

I have axpert MKS 5K. I am planning to buy 9 or 10 modules of 335w. Voc 46.5

Isc 9 .31. Vmp 38.5 Imp 8.72. What will be optimum configuration. 9 panels 3 strings of 3 modules in each strings. Or 5 stings of 2 modules in each strings. I live in karachi pakistan. Weather is hot in summer temperatures in between 30 to 40 C. And in winter 10 to 20 C.

 

Sent from my SM-G925F using Tapatalk

 

 

[Don]

Hi @Anachem20,

The optimum configuration would be 3 strings with 3 modules connected in series. You want to keep your array amps as low as possible. Thinner cables and less losses. 

[Mark]

8 hours ago, Anachem20 said:

Isc 9 .31. Vmp 38.5 Imp 8.72. What will be optimum configuration. 9 panels 3 strings of 3 modules in each strings. Or 5 stings of 2 modules in each strings. I live in karachi pakistan. Weather is hot in summer temperatures in between 30 to 40 C. And in winter 10 to 20 C.

Check the max Voltage of the MPPT of the Axpert.  I think these panels may just push it over the edge of the max and then 2 panels per string will be better.  Especially if you have cold winter temperatures when the panels will really push this limit.

[ibiza]

@Don, don't you think the voltage will be little bit too high?

[Don]

VOC = 46.5 x 3 = 139.5 V. The Axpert maximum open circuit voltage =145 V, So in theory it should be fine. He also stated winter temperatures is 10-20 C. That is not cold.

[pilotfish]

I think the max OC voltage of the Axpert is 145v, the 3 series panels would be 140v - so the configuration suggested by Don is correct.

[pilotfish]

Oops, answered at the same time

[Don]

Lol, happens a lot with me too.

[Don]

3 Strings in parallel = 140 Voc and say 27 Amps.

5 Strings in panels = 93 Voc and 45 Amps.

With the Axpert the optimum efficiency is achieved with PV voltage between 110-115 Volt, if I remember correctly. (Can't find it now). The 93 Voc to start off is way too low, taking into account that once the panels start to take load the voltage would drop to around 75 Volts. The 45 Amps is just too high for me. 

With the Axpert, the optimum configuration = 3 panels in series. If the Voc is to close for comfort, I would then rather get slightly lower rated panel of say 315-320 Watt to reduce the Voc slightly. 

 

[___]

45 minutes ago, Don said:

voltage would drop to around 75 Volts.

Vmp = 38.5, so it will drop to 77V to be exact, assuming no losses in the cable. That is ample if a little on the low side. On the other side, the 140V open circuit feels a bit high, even on my 150V blue stuff that would make me feel a little uneasy. 140/145 leaves a mere 4% space, and 140/150 a mere 7%. So this seems to me a task of picking the lesser of two evils.

MPPTs -- being buck converters -- have inherent constraints. They usually cannot vary the duty cycle with infinite precision (the MCU will have a discrete number of possible step sizes on the PWM outputs), they cannot regulate to below their internal reference voltage, and usually the FET requires a bit of off-time so a 100% duty cycle is often not possible. Also, most losses occur at low input voltage and high output current.

So I tend to agree with you that the 3x3 configuration is best... but man... that Voc gives me the heebie-jeebies.

I think this just again confirms that with the Axpert, you really want to use strings of 3, but 60-cell modules rather than 72.

[Don]

40 minutes ago, plonkster said:

I think this just again confirms that with the Axpert, you really want to use strings of 3, but 60-cell modules rather than 72.

72 cell is not a problem. I have 315 Watt panels with a stated of STC = 45.79 Voc and NOCT = 42.2 Voc. I have 2 arrays. My actual Voc is around 124 - 126 V with 3 panels in series and 2 strings in parallel per array. Not even close to 145 Volts.

[pilotfish]

The full Voc would be obtained at an insolation level of 1kw/m2, which to my understanding is obtainable in a laboratory or by a perfectly aligned panel on the surface of earths atmosphere.

[Don]

Yes, the STC is laboratory figures with Irradiance 1000 W/m2, cell temp @ 25°C. That should be the absolute max the way I see it. I might be wrong.

[___]

19 minutes ago, Don said:

1000 W/m2,

Oh, that's fairly easy to obtain. I believe that's a perfect sunshiny day with good panel alignment in the Northern hemisphere, at 1.5 atmospheres. It's not some difficult-to-obtain only-in-the-lab number. But the Voc, I believe that is a difficult-to-obtain figure, it requires bright sunshine and very low temperatures. Only reason why I'm quoting the 4% margin is simply just in case the Chinese embellished the number a little, as they sometimes do, that is to say the 145V might be a theoretical maximum that included no tolerances. I don't know... it's just a rule of thumb for me to leave 10% if I can. ;-)

[Mark]

On my small system 600W (2x300W panels), draadkar as @Chris Hobson called it. I could routinely got 110% and my best was 127% of rated maximum.

[Chris Hobson]

Karachi has a temperate climate with no low temperatures. So high voltage due to low temperatures should not be a problem. What is confusing is the specs of the Axpert. PV V_oc is 145 VDC while MPPT range is given as 60 to 115VDC. Don has similar panels running in series multiples of 3. Don what is your experience with MPPT voltages above 115 VDC?

[pilotfish]

@Chris HobsonThe 115v spec relates to operating voltages (ie knee point voltage), the 145v spec relates to never exceed Voc.

[pilotfish]

If for example your panels where sitting in bright sunshine with your panel isolator off then your panel voltage would rise to V(open cct), at the moment of switching on your panel isolator the system would be exposed to Voc until current is drawn and voltage drops to a level set by the MPPT. 

[___]

9 minutes ago, Chris Hobson said:

MPPT range is given as 60 to 115VDC.

Buck converter is optimised for a certain range. A buck converter basically pumps charge into an inductor and then squeezes that charge out into a smoothing capacitor on the other side. The higher the input voltage (more specifically, the higher the Vin/Vout ratio), the quicker you fill up the inductor, the shorter the duty cycle has to be (taking into account that switching electronics have slew rates of their own), so for larger ratios you want a larger inductor and/or a faster switch. That means the size of the inductor and the switching electronics operate best in a certain range. It still works outside that range, just not as efficiently. It could even be that it cannot switch the FET at the required duty cycle to hit the MPP correctly, so it might even operate a little bit below the MPP (meaning you lose power). This is all conjecture... theory.

[Don]

5 hours ago, Chris Hobson said:

Don what is your experience with MPPT voltages above 115 VDC?

@Chris Hobson, the voltage normally only goes above 115 as @pilotfish said, when there is no load on the panels = Open circuit voltage. Mine goes up to 124-126 Volt. As soon as I put load on the panels the voltage would drop right down. My panels are doing 3450 Watts right now and both inverters sitting at 95 Volts.

With low loads, both inverters would cut back, with the one inverter operating between 100 and 110 Volts and the other inverter would happily sit at 120 Volts doing the controlling going up and down, depending on the load. See an example below. At low loads, Inverter 2 is idle doing just about nothing, cutting right back to about 120 Volts.

With only 1 inverter, I doubt the voltage would go above 115 Volts and if it does, I don't see a problem with that. That would just be because of very low load on the system. As I said before, I read somewhere that the Axperts perform optimally at 110 or 115 volts. I just cannot remember where I read it. You therefore want to operate as close as possible to that voltage. That is why I don't think to only have 2 panels in a string is such a good idea, as they would operate probably between 75 and 80 volts with high loads, which is very far from 110-115 Volts. 

The ideal panel setup would be where the maximum open circuit voltage is as high as possible, but not exceeding 145 Volts and under full load operate at close as possible to 115 Volts. That is the way I see it. Therefore the 335 Watt panels @Anachem20 are looking at, might just be the optimum panel size for the Axperts, with 3 panels in series in a string, with Voc = 139.5 and Vmp = 115.5 (to be exact).

[Don]

So yes, you can take a Formula 1 car around a track at 120 km/h stuttering and backfiring as it was not designed to do that, but at 300 km/h they purr around the track like a cat.

[Anachem20]

I installed 9 panels in 3 strings. My peak yield is 2350 watts. Distance between inverter and panels is 85 feets. I used 25mm2 single core cable. Panels facings are south and tilt angles are optimum. Is this yield is ok ?

Sent from my SM-G925F using Tapatalk

[Chris Hobson]

HI Anachem 20 I am happy with your cabling. For your specs you are keeping losses about 1%. I would expect a little more power as you go into summer but I don't think there is a problem with you array.Momentary peaks is not what I would be looking at rather what steady output are you getting between 10am and 2 pm. Peaks close to 3000W tend to occur on partly cloudy days rather than days with clear skies.

[Don]

Hi @Anachem20, that seems to be a little on the low side.

1.) Were the panels fully loaded at that stage?

2.) What is the angle of your panels?

3.) What is the temperature there? 

4.) Do you have any shading from nearby buildings that fall onto your panels?

It would be a great help if you registered your system on PVOutput.org and posted your outputs there. That way we can see what is happening during the day.

[Don]

4 minutes ago, Chris Hobson said:

Peaks close to 3000W tend to occur on partly cloudy days rather than days with clear skies.

I agree Chris. On clear days I would have expected the panels to run closer to a peak of 2700W.