Relative Low Performance of Domestic PV System

[binky]

so the 2 arrays have been wired as a single string (there's 2 cables per string), no wonder performance is down, the 2 arrays will be outputting peaks at different times which makes it hard for the MPPT to extract the maximum from either array becuase it is trying to track 2 different power curves. This video may help

 

[Bob Smith]

If this is confirmed all the panels have been put on one string and knowing that half of them are facing a different way (opposite direction), I think it would be fair to say that they have not installed the system for optimum performance.

@Jena Please check carefully your system (number of DC switch, check online what your inverter is doing) before shouting at your installer
Just to be sure that what we are saying is correct.

Why is it important in your situation to have your panels connected in two strings?
Each string of your inverter needs a minimum of power to start converting DC to AC.
In a string the panels are installed in series and to make a long story short, if one panel is not performing this affect the entire string.
Therefore, since your panels are facing West and East they are not exposed to the light the same way. Even if one side west or east produces enough to kick off the invert the other side receiving less light will affect the entire string.

I do not know the type of inverter you have. Some still have only one string. This could be the reason for such set up but then the question becomes why having installed an inverter with one string.

Looking forward to receiving your confirmation (or not) so we can say we have found the root cause of the poor performance of your system.

 

[Joules]

This reminds me I did have a problem when I got my system in 2016. I bought my system with a friend living about 500 yards. We place the ordered together to get a discount. We got the exact same equipment, The only difference my 16 panels are SE while his are 8 E and 8 West- His roof has a slightly higher pitch than mine. Our systems were commissioned mid of jan 2016.
It was expected that he would generate earlier and later in the day and in winter, because of the pitch, also more. This is what happened.
However, came April and his system could peak at about 3kW while mine did not exceed 2kW. I started to really wonder.
I contacted the installer and they found out only one string was producing. They came and found out one string had not been connected correctly!!

I would have found out eventually but not knowing when would my panels generate the most power, it would have taken longer.
Comparing is good
So maybe as Joules says, maybe another cause is that some panels or string are connected correctly.

Thank you for your really useful comments Joules. I interrogated the inverter as you suggested and have attached images of the first two items on the scrolling information list. The first shows a voltage of 167.0V and 0.2A. The second line on the list displayed a voltage of 1.6V and 0.1A. Lines three and four were in a similar format, but both showed voltages of 0.0V and a amperages of 0.1A. There are only two cables entering and then exiting the isolation box located before the inverter (please see image). Forgive my ignorance, but could one cable be from one string and the other from the string located on the adjacent roof? Alternatively, could one cable be from a single string made up of all 10 panels, whilst the second cable just completes the circuit back to the panels? As I wrote earlier, I am no electrician.

Probably not the best day to be looking at the system performance figures given that it is exceptionally gloomy today, with current power output only being a miserly 20W and I guess readings being subject to inaccuracies from baseline noise, perhaps. I’m not sure whether this helps you to comment further on our system, but any additional thoughts would be appreciated. I do possess a clamp voltmeter and wondered whether trying to measure DC current and voltage in the cables entering the inverter might be useful. Yesterday was also overcast and we managed only 0.4 kWh and our nearby friends produced 1.6 kWh. Many thanks.

Thank you and well done! Can you please provide us with a pic of the exact model of inverter, there is a label on the side. The reason why is your 3.6 model comes in two flavours: one with a single MPPT for use with two pairs of panel strings and the other with two MPPT for two pairs of strings. What the installer may not be aware of is something that happens with East/West installs when one string pair is used to link East/West. When panels have differing light levels or insolation the ones with the weaker light act as a resistance preventing the other higher level light ones from generating properly. Similarly, one MPPT served by two strings may create the problem you are experiencing. If you can need a pic looking up at the bottom of the inverter. You have options but more on this later.

 

[Jena]

Jena,

You should have one DC switch per string. The DC Switch is the black one we see on your photo.

I assume you have a online logging to see what your panels/inverter are doing.

Mine looks like this

View attachment 14599

This is the two string (PV1 and PV2), right now as you can see they are producing nothing.

From my auntie it looks like this
View attachment 14600

you can see that the two strings have not performed the same. Her panels are similar to yours. Some facing west (one one string) and the rest facing Est (on the other string)



Check on your online account and see if you can see something similar.

I only have the one DC switch Bob, with the two cables feeding in and out of it. There is an app I use for monitoring the system/inverter performance called Soliscloud. There is nothing in the app indicating that the two sets of panels can be monitored independently.

 

[Joules]

Thank you for your really useful comments Joules. I interrogated the inverter as you suggested and have attached images of the first two items on the scrolling information list. The first shows a voltage of 167.0V and 0.2A. The second line on the list displayed a voltage of 1.6V and 0.1A. Lines three and four were in a similar format, but both showed voltages of 0.0V and a amperages of 0.1A. There are only two cables entering and then exiting the isolation box located before the inverter (please see image). Forgive my ignorance, but could one cable be from one string and the other from the string located on the adjacent roof? Alternatively, could one cable be from a single string made up of all 10 panels, whilst the second cable just completes the circuit back to the panels? As I wrote earlier, I am no electrician.

Probably not the best day to be looking at the system performance figures given that it is exceptionally gloomy today, with current power output only being a miserly 20W and I guess readings being subject to inaccuracies from baseline noise, perhaps. I’m not sure whether this helps you to comment further on our system, but any additional thoughts would be appreciated. I do possess a clamp voltmeter and wondered whether trying to measure DC current and voltage in the cables entering the inverter might be useful. Yesterday was also overcast and we managed only 0.4 kWh and our nearby friends produced 1.6 kWh. Many thanks.

Before you upload a photo of the inverter product label please redact or obscure its serial number. To move forward the model number is required. Thx.

 

[Joules]

I only have the one DC switch Bob, with the two cables feeding in and out of it. There is an app I use for monitoring the system/inverter performance called Soliscloud. There is nothing in the app indicating that the two sets of panels can be monitored independently.

Solis Cloud is why the serial number should remain private to you only.

 

[Bob Smith]

I only have the one DC switch Bob, with the two cables feeding in and out of it. There is an app I use for monitoring the system/inverter performance called Soliscloud. There is nothing in the app indicating that the two sets of panels can be monitored independently.

This screenshot is from SolisCloud but from a web-browser. Not sure what info you have from the app but you could simply log to it from a web browser

 

[MHarv]

Look at the inverter spec sheet or look underneath the inverter to see how many PV inputs the inverter has, might just have one set of PV inputs; one for + positive and one for - negative therefore your east/ west are on one string which means your inverter is single mppt

 

[Jena]

so the 2 arrays have been wired as a single string (there's 2 cables per string), no wonder performance is down, the 2 arrays will be outputting peaks at different times which makes it hard for the MPPT to extract the maximum from either array becuase it is trying to track 2 different power curves. This video may help

Thank you both Binky and Bob for your continued advice. The video you attached, Binky, is excellent and explains the principle behind MPPT function really well. There’s one point that’s become slightly confused as this thread has progressed, in that, our panels face south and south south east ( please see my initial post for how the panels are arranged). Presumably, the performance impact of having the two arrays on one string is much less when the angle between each array is 22.5 degrees (as ours is) rather than 180 degrees, had they been positioned east and west facing. However, you have established that our panels comprise only one string. This takes me back to a thought I had initially, that not all of the ten panels have actually been connected. The poor performance difference we see between our system and our nearby friends is about 40-50%. Without any solid evidence at this stage, a possible cause could be that the four panel array has not been successfully connected. This would equate to the percentage performance difference when compared with our friends’ system. I was hoping that there was a way to easily check this without having to access the panels externally, via the roof. Is there?

I know that my installer will stick to his guns and say it is early days and he needs to monitor performance over 90 days. It’s difficult to argue with this without solid evidence. He may be right, but I cannot comprehend such a significant difference in power generation. He has cited a few possible reasons to account for the difference including inverter efficiency (our Solis 3.6 reviews well when googled) and inverter start-up voltage difference. Our inverter has a lower start-up voltage than does our friends’. Our roof pitch is also slightly steeper, which in winter I believe, is an advantage. Nothing adds up to me. Many thanks.

so the 2 arrays have been wired as a single string (there's 2 cables per string), no wonder performance is down, the 2 arrays will be outputting peaks at different times which makes it hard for the MPPT to extract the maximum from either array becuase it is trying to track 2 different power curves. This video may help

 

[Bob Smith]

Hi Jena

sorry if we brought with confusion with the orientation of the panels. Effectively with your set up (South and South East with a 22.5deg difference) the situation is not as extreme as for a 180°.

Nonetheless, unless you purchased knowingly a one string inverter. It is always best practice when going for a string inverter to have at least 2 string.
Even when all the panels are facing the same way and do not experience any shade, you are not prevented for any type of fault and if a panel fails the whole string has a reduced performance (unless you have optimizer (kind of bypasser ...)

You mention that your inverter has a lower minimum voltage to start. What is it?

It is clear that in winter and having the panels just now, it is difficult to correctly judge the impact of your set over a whole year vs the same configuration but on two strings....
At least you now have a reasonable root cause to challenge your installer (in a nice and polite way, you still need him lol )
In any case he must be able to justify his decision and bring you pros and cons vs a two-string installation (he may not have done a written analysis during his design phase but he must have installed many arrays on 2 strings, and maybe just mentally he decided that for yours one string would be enough). Then you can challenge his analysis.

The benefits must be for you and not for him (eg easier installation).

 

[Jena]

Bob, Binky and all of the others who have helped, thank you. You might be interested to know that I undertook a thorough search of our loft space this morning, hoping, but expecting, to trace the panel wiring further. The attached images shows what I found. It appears that the string of four panels on our south south east facing roof have not been connected! I couldn’t believe my eyes, but was relieved at the same time. This explains and corresponds to our mathematical estimate of a 40% deficit in power production. There’s always (well nearly always), a silver lining to come out of situations like this. Through the advice you have kindly proffered, I now know how the second string should be connected to the inverter to maximise power output through the second MPPT channel and increase both system resilience and safety. I’ll insist that the installation is completed to achieve this. Thank you to everyone for your thoughts and assistance and sharing my conundrum. Wishing everyone a happy and peaceful Christmas.

 

[Bob Smith]

Hi Jena

I am glad you now better understand the technical aspect but all this would have not been possible with a dose of common sense that led you to think something was wrong.

There could be several reasons for those cables.
Just throwing hypothesis

1. They expected to connect the panels to a second string of your inverter but they found out your inverter had only one string. So left the cables (forgot to remove them?) and connected all the panels on one string
2. If they expected to have 2 strings, well a previous photo shows only one DC switch.
3. Keep an open mind because the truth could be elsewhere

In any case, they should have not left those cables like this.

It seems that you are in a very good position and have leverage for the conversation ahead.

Keep us posted

PS: I have said that for my installation the cable to the inverter for the second string was not attached correctly. They realized their mistakes and compensated me for the loss of the generation.

 

[Jena]

Hi Jena

I am glad you now better understand the technical aspect but all this would have not been possible with a dose of common sense that led you to think something was wrong.

There could be several reasons for those cables.
Just throwing hypothesis

1. They expected to connect the panels to a second string of your inverter but they found out your inverter had only one string. So left the cables (forgot to remove them?) and connected all the panels on one string
2. If they expected to have 2 strings, well a previous photo shows only one DC switch.
3. Keep an open mind because the truth could be elsewhere

In any case, they should have not left those cables like this.

It seems that you are in a very good position and have leverage for the conversation ahead.

Keep us posted

PS: I have said that for my installation the cable to the inverter for the second string was not attached correctly. They realized their mistakes and compensated me for the loss of the generation.

Hello Bob. Keeping an open mind is always good advice. Our inverter’s spec is dual MPPT, so hopefully option one does not apply (see attached image). There is no evidence of any connection from the other string to the cables leading from the ‘working’ string. I can trace this string’s cables from where they pass into the roof space and then run to the DC switch. So no real evidence of a last minute splice into these cables and change in circuit plan, if indeed, there ever was such a plan. We’ll see on Thursday morning when the electrician arrives, but I remain hopeful. Yes, of course, I’ll update the thread after Thursday’s visit.

 

[Jena]

Look at the inverter spec sheet or look underneath the inverter to see how many PV inputs the inverter has, might just have one set of PV inputs; one for + positive and one for - negative therefore your east/ west are on one string which means your inverter is single mppt

Thanks MHarv, there are two pairs (see image) labelled DC1 and DC2, so looks like it has dual MPPT as per the spec sheet.

 

[Bob Smith]

When (hopefully not If) they acknowledge that they forgot to connect some panels or that they install them all on the same string while you have a 2 string inverter, it will be time to discuss money because during all that time you lost in generation and therefore money.
You have already evaluated 40%, this is without mentioning all the time you wasted and this situation could have lasted a lot longer if you had not felt there was something wrong (thanks your neighbour )

You can also ask them to make a donation to electricianforum.co.uk without this forum maybe you will still be scratching your head

 

[binky]

Bob, Binky and all of the others who have helped, thank you. You might be interested to know that I undertook a thorough search of our loft space this morning, hoping, but expecting, to trace the panel wiring further. The attached images shows what I found. It appears that the string of four panels on our south south east facing roof have not been connected! I couldn’t believe my eyes, but was relieved at the same time. This explains and corresponds to our mathematical estimate of a 40% deficit in power production. There’s always (well nearly always), a silver lining to come out of situations like this. Through the advice you have kindly proffered, I now know how the second string should be connected to the inverter to maximise power output through the second MPPT channel and increase both system resilience and safety. I’ll insist that the installation is completed to achieve this. Thank you to everyone for your thoughts and assistance and sharing my conundrum. Wishing everyone a happy and peaceful Christmas.

Do not handle the ends of those cables! DC is nasty. Can you see if they exit through the roof somwhere close to the 4 panels? If so, your installer hasn't got a leg to stand on, get them back to complete works.

 

[Jena]

When (hopefully not If) they acknowledge that they forgot to connect some panels or that they install them all on the same string while you have a 2 string inverter, it will be time to discuss money because during all that time you lost in generation and therefore money.
You have already evaluated 40%, this is without mentioning all the time you wasted and this situation could have lasted a lot longer if you had not felt there was something wrong (thanks your neighbour )

You can also ask them to make a donation to electricianforum.co.uk without this forum maybe you will still be scratching your head

Hmmm. Don’t know about the installer, but I’m more than happy to.

 

[Jena]

Do not handle the ends of those cables! DC is nasty. Can you see if they exit through the roof somwhere close to the 4 panels? If so, your installer hasn't got a leg to stand on, get them back to complete works.

Yes, they correspond exactly with where the four panel string is positioned. I’ve left the cables well alone, even though the chaps that fitted the panels taped-up the cable ends. The installer is sending an electrician this Thursday. Thanks again Binky.

 

[binky]

Yes, they correspond exactly with where the four panel string is positioned. I’ve left the cables well alone, even though the chaps that fitted the panels taped-up the cable ends. The installer is sending an electrician this Thursday. Thanks again Binky.

Your welcome

 

[Andy1733]

Not sure if this is helpful. You mention you use Solis cloud. I've Inc some images from my Solis app and in the first image along the top it says device. Click on this and it opens the 2nd image which shows my 2 strings listed as pv1 and pv2.