#1 Jan. 5, 2016 14:29:59

prodigymad
Registered: 2015-02-06
Posts: 17
Reputation: +  0  -
Profile   Send e-mail  

Dual MPPT:- Single string or two?

Hi,

I recently had a 2.4kWp PV array (URN: 8389) added to my garage roof (in addition to a ground mount system URN: 7115) and I was wondering what peoples opinions were about Dual MPPTs.

The new 2.4kwp array is using a Solis 2.5k inverter which has a dual MPPT function built in:

http://www.ginlong.com/index.php?s=/en/pro_sheet/mid/4/tid/27/id/10.html
Input Side(DC)
Max. input power 3kW
Max. input voltage 600V
Start-up input voltage 120V
MPPT voltage range 100-500V
Max. input current 10A+10A
MPPT number/Max. input strings number 2/2

The installer has just wired the whole array up as a single string as they said that splitting into 2 strings would mean that the array would struggle with low voltage and fail to start up unless it was very bright sunlight.

They reckoned that a single string would hit the 100v start-up voltage easier and therefore start generating sooner.

However, surely if this was the case then why would the manufacturer even offer this function. Is it 100v per MPPT string?

Cheers!


Oh, P.S. the array is 8x 300w LG Neon Black panels:
http://www.lg.com/us/commercial/solar-panels/lg-LG300N1K-G4

Edited prodigymad (Jan. 5, 2016 14:33:15)

Offline

#2 Jan. 5, 2016 16:28:14

SteveRogers
Registered: 2011-12-08
Posts: 201
Reputation: +  0  -
Profile   Send e-mail  

Dual MPPT:- Single string or two?

Your installer is right about the startup/operating voltage under low light conditions. My installation was set up as two strings because the 18 panels are divided either side of a dormer. Unfortunately, the first inverter was not matched to the string output (inverter startup voltage was 315V!) and the system struggled and even stopped working in full sun as the panels warmed up. Changing the inverter fixed the problem (after 3 months of wrangling).

Your 8 panels have an MMP voltage of 8*32.5 = 260V so that's well below the maximum voltage for the inverter. The maximum inverter current is 10A per string which is more than the maximum panel output so that should be just OK too. The MMP voltage for a string of 4 panels would be 130V which is only just above the start up voltage of the inverter so the system would not work well as two strings.

The advantage in having a 2 string system comes if there is a shadow for part of the array (e.g in the case of my dormer at certain times of the day) which will drag the output of a whole string down. In such a case it's better to have two strings as at least one would be working without shadow. However, in your case I don't think 2 strings would work well because of the startup voltage as indicated above. I'd stick with a single string for a couple of months and see how it goes.

Given the present weather I guess no-one's system is working too well!

Steve
URN 215

Edited SteveRogers (Jan. 5, 2016 16:32:41)

Offline

#3 Jan. 6, 2016 10:39:22

prodigymad
Registered: 2015-02-06
Posts: 17
Reputation: +  0  -
Profile   Send e-mail  

Dual MPPT:- Single string or two?

Well, the main reason for asking is because I just submitted my first quarter readings and First Utility rejected them as they were below the threshold they expected.

According to my MCS certificate it is estimated that the system will produce 2180kWh over the year. However, I've put the system through PVGIS and I can't get anywhere near this figure. I get a figure closer to 1,760kWh (array is pointing to about 220deg rather then due South).

For reference the array was commissioned on the 29th Sept and the meter read done on 31st Dec was 125kwhs.


Here is the response from First Utility:

“We use the estimated annual generation figure (EAG) on your MCS statement and input this figure into a solar radiation model. This takes into account a number of factors and calculates how many Kws your system should produce on a daily basis. We are then able to predict what your meter reads should be from read to read. We have a 10% tolerance and if the read falls outside this tolerance Ofgem require us to obtain evidence of the read before we are able to use it for payment.

We were expecting a meter read of 261kws for an EAG of 2182. I have also calculated that if the EAG was changed to 1760 we would have expected a meter read of around 205 kws. Most customers who fail the tolerance levels are only between 5 to 9 % higher or lower than the tolerance. As the read you provided was 51% lower then expected it would indicate that their may be an issue with your system.”


I'm intrigued to see what happens with the “compare against my neighbours” feature to see if this is really under performing.

My older 3.2kwp system generated 210kwhs over that same quarter so I can't see how the 2.4kwp system should have generated 205. We've had some dire weather up here in the North West so I'm not expecting massive figures this quarter.

Offline

#4 Jan. 6, 2016 11:43:20

prodigymad
Registered: 2015-02-06
Posts: 17
Reputation: +  0  -
Profile   Send e-mail  

Dual MPPT:- Single string or two?

In fact, that's got me thinking about the other 3.2kwp array as that is installed on dual MPPT channels.

It is a single ground mounted array facing 130deg - 140deg from North, comprising of:

10x Seraphim 320w panels (SRP-320-6MA): http://www.seraphim-energy.com/en/file/pdf/SRP-Datasheet-6MA.pdf
Vmp = 37volts.

They are wired in to a Samil SolarRiver 3400TL-D as two banks of 5, like this:
1 1 1 1 2
1 2 2 2 2

http://www.enfsolar.com/pv/inverter-datasheet/552

So, presumably it is 185 volts per MPPT channel, with a startup voltage of 100v.

Again, given the number of “grey” days we get, surely a more optimal configuration would be to stick all 10 panels through a single MPPT channel, at 370volts.

Apparently the “nominal voltage” of the Samil is 380v.

Hmm, I'm wondering if its worth getting that one re-wired to a single channel

Offline

#5 Jan. 7, 2016 12:25:48

prodigymad
Registered: 2015-02-06
Posts: 17
Reputation: +  0  -
Profile   Send e-mail  

Dual MPPT:- Single string or two?

I've got a day off today so I can check out the live inverter readings for the smaller 2.4kwp array.

It's a grey overcast day and it seems that the array didn't start-up until about midday!

Before midday:

DC1 Input Voltage - 250v
DC1 Input Current - 0.0A

DC2 Input Voltage - 0v
DC2 Input Current - 0.0A

AC Output = 0.0W

——————————-

After midday, once started up:

DC1 Input Voltage - 274v
DC1 Input Current - 0.6A

DC2 Input Voltage - 19v
DC2 Input Current - 0.0A

AC Output = 142W

Strangely, there's not supposed to be anything wired into DC2 so I'm not sure why it has any volts going through it!

Offline

#6 Jan. 8, 2016 13:14:55

SteveRogers
Registered: 2011-12-08
Posts: 201
Reputation: +  0  -
Profile   Send e-mail  

Dual MPPT:- Single string or two?

Have a look at the met office anomoly maps for sunshine (map type = 1981-2010 anomoly) for the months in question and your location.
http://www.metoffice.gov.uk/climate/uk/summaries/anomacts

Where I am November and December were rough - about 80% of the expected figures - whereas September and October were roughly on target. The sunshine levels from the anomoly maps agree well with the production figures expressed as a % of the expected target.

My 4kW system (directly S facing and near ideal roof angle) produced 415 kWh through Oct-Dec. My reading for FIT was 488 kWh for mid-Sep to mid-Dec. January 2016 is looking disastrous - less than 25% of the expected target so far. Every day is well below target (3.9 kWh) and I've only had 2 days producing more than 1 kWh so far.

You can model the predicted monthly production yourself using PVGIS (search on these forums for details). The model will take into account your location, panel orientation, roof angle, shading and other losses (guesses) as well as the local climate/sunshine levels that might be expected.

My panels are starting up around 9 or 9:30 am and cutting out around 2:30-3:30 so I think something may be wrong with your system (unless your weather has been even worse or you have a big shading problem). On the other hand that DC1 voltage and current after startup look OK.

Finally, I think the startup voltage is supplied by the panels operating open circuit (Voc) when they are actually producing a higher voltage (39V per panel?). As soon as the inverter starts to draw current then the voltage should drop to Vmmp (32.5V).

Just noticed you are in the NW so your installer is likely to palm you off given the appalling weather, I'm afraid :-(

Hope that helps,
Steve
URN215, 3.96 kWp, S facing, 39 degree roof, minor shading, North Yorkshire

Edited SteveRogers (Jan. 8, 2016 13:20:28)

Offline

Board footer

Moderator control

Powered by DjangoBB

Lo-Fi Version