Correct Inverter Choice?

[Dave-geo]

I hope someone on here can help with an opinion on my install.


The system was installed in December to beat the original deadline. I was quoted for 14 Bosch or similar panels and an SMA or similar inverter. Basically they said they would install what they could obtain as supplies were short.


The panels supplied ended up being 245watt Bosch panels giving a total system of 3.43Kw and a Theia Valera 4.4 HE-t inverter.


The panels are installed on a south facing 50 degree slope roof. There is a large chimney that shades some of the panels early on in the day. First of all it covers three panels then two then one. By mid-day the panels are completely shade free for the rest of the day. They are split into two strings of 7 panels with all the shaded panels being on one string. The DC run is approximately 24M.


There is also an additional shading issue in winter caused by nearby trees. These were a problem this winter but the sun is now high enough and so do not present a problem. I have a chainsaw and these will be dealt with before the Autumn :devilish:


The system hasn’t lived up to my expectations or the prediction provided by the installer although it is early days yet. So far the prediction to date would have been for 385Kwh’s and the system has actually produced 220Kwh’s.


The question is do I have the correct inverter for the system installed. The Theia only has 1 MPP tracker unlike others which appear to have two, does this make a big difference? The inverter also appears to be a larger size than needed, does this give any problems? Should a setup like this have used two smaller inverters?


Today it has been sunny all day so I have attached a graph of todays generation and you can see that up until mid-day the production is poor compared to the rest of the day. On a fully sunny day in March I think I could expect 15 (is that realistic for South facing sunny South Wales?) or more KWh but in fact got 10, is that reasonable?


I see on another thread that I should have been given a SAP calc. I was told a rough payback time, approximately 3000Kwh per year and quoted a return on investment but was never given a Sap calc. Does that matter?


Should I have received a Part P certificate for the work?


I should say that I am happy with the inverter. I connected it up to my router and it gives me all the stats I could ask for. Is that facility available with other inverters?


Sorry for all the questions, any opinions you may have on the above would be hugely appreciated as I can then have an educated chat with the installer if required.

 

[babba]

Inverters oversized... yes you should get a part p..saps calc you shoud get but its not that important most are flawed by about 20% on a decent system from what I see...BUT you agreed to any system so I cant see you have any come back...

 

[provr6]

Your predictions would only be based on using a Sap calculation and would most likely assume no shade whatsoever. The shade will impact on your generation but if you are shade free by noon in the winter you will still get a healthy amount of sun come summertime.

I don't think you would necessarily need a dual inverter as all your panels are on the same roof.

I got 8kw today from a 2.7kw system (although it is not wired up correctly yet, so would hope to improve on that)

 

[Dave-geo]

BUT you agreed to any system so I cant see you have any come back...

Thanks for your input babba. The agreement was for a correctly specified and correctly installed system with whatever equipment was available. Just because the equipment was in short supply should not mean that I end up with a deficient system. I get on well with the installer so don't think this should be a problem.

The inverter is due to be replaced anyway as it often has a fault code "low PV isolation resistance" so I am trying to find out if there is something more suitable that I can try and specify at this stage.

Anyone know what causes the above fault?

 

[babba]

: The inverter LCD screen display “PV Isolation Low”, how can I deal with it? : PV Isolation Low is insulation fault. Every time before inverter trying to connect to grid, inverter will check the whole system’s insulation condition, if the insulation resistance is too low, inverter will show “PV Isolation Low” on its LCD screen and won’t connect to gird. Here are some suggestions to solve this problem:

1) Please first check whether the AC output wiring were connected properly or not. The PE wire in AC output must be connected to the earth. It is better connect to PE line from grid than to your resident earthing.
2) Please check casing of panels were grounded properly. Meanwhile, do not connect PV+ or PV- to ground.
3) Check whether DC switch or AC switch was moistened. There might be a little water inside switch, check and clean.
4) Use a multi-meter to measure the insulation resistance between PV+ (or PV-) and ground, it must be greater than 8 M ohm. Please measure at night.
Note: The PV isolation Low detection is very accuracy, it rarely have problems, so please check earth conditions carefully. The final suggestion is you take an additional inverter as a replacement to the site, replace the inverter if the problems still persist in. If the 2 inverters all have PV Isolation Low issue, it indicates the problem is not from the inverter but from the ground. The function of replacement is just for you to focus on attention to ground conditions.

 

[pww]

If you are getting 'low pv isolation resistance' fault this is usually due to a wiring problem not the inverter. When in this condition the inverter will not connect to grid. Sounds like a rushed and poor system. I would suggest getting the company back to re-install it properly!

 

[babba]

PS this is a copy n paste thank the internet..

 

[Eco Deal]

I think a dual MPPT would be better as when the three panels are shaded then all 14 are working to the three shaded panels !!!

if a dual MPPT was there the 7 on the unshaed string would work to full.

 

[Gavin A]

I think a dual MPPT would be better as when the three panels are shaded then all 14 are working to the three shaded panels !!!

doesn't necessarily work like that, and doesn't seem to be from the graph. It looks like it's pulling the power at the peak MPPT of the unshaded string, so that one's unaffected, it just means that the 2nd string isn't operating at it's peak point.

So not ideal, but not as bad as if it was all run in one string.

 

[babba]

Swings and roundabouts

 

[Gavin A]

So far the prediction to date would have been for 385Kwh’s and the system has actually produced 220Kwh’s.

If they only supplied you with a rough 3000kWh a year estimate where does this 385kWh to date prediction come from?

 

[moggy1968]

220kw seems about ok to me for the time of year and the lmitations you have described

 

[kyle simpson]

Hi dave, I think there is severe problem with the system.I have check the pramater of the inverter as well as the panel.
Inverter, MPPt 230~500v, input power 4600w, output power 4400w.
Panel, 245w, open voltage37.7V, operating voltage 30.1V.

Are you sure there are in two separate strings? If so, the system's open voltage is 264V, operating voltage is 211V(which is out of the inverter's MPPt voltage range), which is too low for the inverter. It means the inverter can't working properly...

Meanwhile, I'd like to say some about the 'shade' and the 'dual mppt' issue.
As some of the panels are in shade sometimes, a dual-mppt is needed. I agree with MEP ELEC LTD, who said----

"I think a dual MPPT would be better as when the three panels are shaded then all 14 are working to the three shaded panels !!!if a dual MPPT was there the 7 on the unshaed string would work to full."

The voltage of the shaded panels will drop. If the two strings are not equal in voltage, the higher will charge the lower_this will reduce the lifetime of the panels!!!

 

[Dave-geo]

: Inverter will show “PV Isolation Low” on its LCD screen and won’t connect to grid.

My inverter still connects to the grid when this fault is showing.

 

[Dave-geo]

If they only supplied you with a rough 3000kWh a year estimate where does this 385kWh to date prediction come from?

I have a friend who had a 4Kw system installed by the same company before me. He lives about 15 miles from me as the crow flies and he was given a prediction. I have kept the meter readings for the two systems and increased my readings by 16.65% to account for my system being smaller. As you can see from the attached graph he is slightly behind prediction but I am considerably behind what I would have hoped to acheive.

 

[babba]

Meanwhile, I'd like to say some about the 'shade' and the 'dual mppt' issue.
As some of the panels are in shade sometimes, a dual-mppt is needed. I agree with MEP ELEC LTD, who said----

"I think a dual MPPT would be better as when the three panels are shaded then all 14 are working to the three shaded panels !!!if a dual MPPT was there the 7 on the unshaed string would work to full."

The voltage of the shaded panels will drop. If the two strings are not equal in voltage, the higher will charge the lower_this will reduce the lifetime of the panels!!!

Got a question on the back of this Quote

..what about the bypass diodes...?????

 

[Dave-geo]

Hi dave, I think there is severe problem with the system.I have check the pramater of the inverter as well as the panel.
Inverter, MPPt 230~500v, input power 4600w, output power 4400w.
Panel, 245w, open voltage37.7V, operating voltage 30.1V.

Are you sure there are in two separate strings? If so, the system's open voltage is 264V, operating voltage is 211V(which is out of the inverter's MPPt voltage range), which is too low for the inverter. It means the inverter can't working properly...

Meanwhile, I'd like to say some about the 'shade' and the 'dual mppt' issue.
As some of the panels are in shade sometimes, a dual-mppt is needed. I agree with MEP ELEC LTD, who said----

"I think a dual MPPT would be better as when the three panels are shaded then all 14 are working to the three shaded panels !!!if a dual MPPT was there the 7 on the unshaed string would work to full."

The voltage of the shaded panels will drop. If the two strings are not equal in voltage, the higher will charge the lower_this will reduce the lifetime of the panels!!!

Thanks Kyle this is really useful information.

The system is two strings, typical DC voltage is 220-228 so as you say the MPPT cannot work. I hadn’t spotted that before.

I have just spoken with the installer and his first thought was to put it onto one string to increase the voltage. I mentioned that the shaded panels could then have an adverse effect on all the panels and he agreed. He then volunteered to change the inverter to a twin MPPT model which I guess is the best outcome given this situation.


I have asked if he could supply an inverter which gives me some decent output of stats/information as the HE-t is very good for this. He has agreed. Any suggestions for the most appropriate inverter for this system that will give me lots of information as well?


Regarding the life of the panels mentioned in a previous post what is the best way to maximise that? Do the bypass diodes help or do they only help output when in a shaded system and is this a potential warranty problem if things do go wrong in the future?


Thanks to everyone who has posted on this thread it has really helped me.

 

[babba]

I vote for a power one 3.0 outd

 

[kyle simpson]

My friend explained to me that the shaded panels are not to be re-charged and its life-time shall not be affacted by the bypass diode, sorry, I made a mistake...

Anyway, the power points of the two strings are not the same when there is a shade, so the inverter will not find the Max. power point of the PV array, it's sure that there is a little bit waste of power in this panels. Do you agree with it?

And the most important problem in this system is that the array voltage is not in the inverter's MPPT voltage range...

 

[Dave-geo]

An update on my situation.

I have spoken with the installer and he has agreed to change the inverter to an SMA with twin trackers. This would be free of charge to me and although he made a mistake in fitting the original inverter I believe this is a responsible approach by him.

The second option would be to change to a Solar Edge system. He would change the inverter free of charge and fit the power boxes free of charge. I would have to buy the power boxes and pay for the scaffold.

I believe given the harsh shading to three of the fourteen panels that the Solar Edge system would be a significant improvement. I need to budget about £650.00 for the power boxes and maybe £400.00 for the scaffold so it is not a cheap option having already paid for the system. Do you think it is worth the extra at this stage? It appears that the installer might actually be making some money if we go this way due to the reduced cost of the inverter and limited work required to fit the power boxes.

If we went with Solar Edge would we be better using just one string instead of the current two? The DC run is about 24M.

I have to supply the power boxes. Any suggestions of a supplier who would deal direct with a consumer at a sensible cost? As you can see I have budgeted £400.00 for the scaffold, is that about right for a width of 7M and height to gutter of about 5.5.M?

Thanks for all your help it is really appreciated.

 

[Glen Powell]

Power one would be a good choice as its MPPT starts at 100v however you would be right on the knuckle for maximum input power so could be classed as undersized, A steca 2010+ with slave will give you multiple MPPT and start at 80v which means even on a very shaded roof useful electricity will generated.

As already stated in this post the shaded panels can pull down the rest of the system.

In my opinion if the system is incorrectly designed and the installer has given you a different product then that product should be equivalent or better than the original specification. This should all be at the cost of the installer, if he has made a mistake he can rectify it and sell on the inverter or claim it under his insurance variant on the insurance cover that is.

I do not have PVsol on my home computer so cannot do any simulations to assist you but maybe I could look at this in work should I get 5 minutes unless one of the guys on here have it to hand. What I will say is PVsol does not like the steca as it is a combination inverter system so it will not give you the option to string it correctly on the simulation software.

Some people will say this install is an ideal candidate for micro inverters, please do not go down that route, they sound good in principle but they do need a lot of power from the panel before they start up.

 

[Glen Powell]

Which SMA some also have a high start voltage?

 

[Glen Powell]

Maybe a samil solar river 3300TL would be better sized its just a shame that the MPP range starts a 200v which is not good for shading issues.

 

[moggy1968]

to be honest, if your voltage is dropping to that amount then your output will be so low as to be irrlevant anyway. voltage isn't the main thing affected by shade, current is. picking between, say, 120v and 200v as start up current is pretty meaningless if it is just for reasons of shade or trying to extend the time the array is generating in a day (rather than total voltage limitations due to the number of panels). Most systems will still be generating good voltage right at the end of the day when current is down to 0.1a
voltage is affected more by temperature.

 

[Glen Powell]

I totally agree but if there is only 3 panels shaded he could put them onto the single string, better with 4 to ensure maximum production from the rest on the other string as both the powerone and steca are capable of unbalanced loads.

 

[moggy1968]

hmmn, tricky, might be worth sacrificing one panel onto the smaller string to get the voltage up.

 

[Dave-geo]

Another update!

The system has now been connected temporarily on one string. This brings the string voltage in line with the inverter so the MPPT is now working. The difference in output is incredible at about 45% increase.

Of course the shading now drags down the complete system not just the one string and this is shown well in the following graph from yesterday. It is a huge improvement but the shade is still a considerable problem. If you look at the graph the early morning production is poor when three panels are shaded. When the shading is reduced to just one panel there is a big jump in production followed by a further jump when the system is shade free




I now have to work out what the final solution should be and believe I have three choices.

1. Install PowerOne inverter on two equal strings of seven panels. This would be a simple operation and only involve changing the inverter. The downside is 7 panels would still be dragged down by the shading.

2. Install PowerOne and change the strings so ten panels work on one string and four on another, the four panels are the ones that receive shading from the chimney. This would minimise the shading issue and still give a decent string voltage on the four panels. The PowerOne should be able to deal with the unbalanced strings. The problem with this is the 50° roof pitch as I expect we would require scaffolding to complete this.

3. Change the system over to Solar Edge. Then the shading is only a problem on the individual panels. The downside is the cost of the power boxes and again scaffolding may be required.


Any thoughts on which of the above options would be best and if would be scaffolding be required?


Any comments would be really appreciated.

Note - Image above is correct, thumbnail below is previous Monday and I can not work out how to remove it!

 

[SolarCity]

Thanks for the post. This is really handy data.

To answer your question, I would personally go for the Solar Edge system. If the cost is an issue then I'd go for the Power One option. I would definitely check out getting 4 panels onto one of the strings and see how it comes out.

 

[Dave-geo]

If the cost is an issue then I'd go for the Power One option. I would definitely check out getting 4 panels onto one of the strings and see how it comes out.

Thanks for your reply.


Regarding cost I am quite happy to pay for the power boxes on a Solar Edge system but believe this should be subsidised by the cheaper inverter that would then be used.


Scaffolding is the real issue regarding cost. If we went for a 10/4 string split then scaffolding may be required and likewise may also be required to fit the powerboxes on a Solar Edge system. Therefore I think the cost should be split between the installer and me but we can’t agree on that at present.


He also hasn’t confirmed if scaffolding would even be required hence my question regarding this.

 

[SolarCity]

I'd say that a Solar Edge system would be an upgrade and that is reasonable that you should expect to pay the difference.

However, I think that as the inverter was badly installed in the first place I wouldn't expect to pay anything towards scaffold nor replacing of the inverter.

 

[jupestar]

interesting post and it's a tough call with out seeing the job it's hard to know. power one, or solar edge both good options. if you need a scaffold might as well do the solar edge, but without seeing it istough.

i will add this, my dad had one of the first fit eligible installs. and had a steca inverter which lasted less than a year, it completly stopped working. the installer replaced it with a fronius but changed it to single string mppt (despite there being a big chimney right in the middle). i suggested he complain that they changed a dual steca to a single mppt which was not suitable for his install. he did and now has a power one. he's been monitoring it a bit,not crazy, but just keeping an eye, and he swears blind he got best production from his fronius.... and looking as his scribblings, hes actually right. the higher voltage by having all the panels on one string was actually better.... no real point to that story other than you can do all the computer simulations you want in real life it doesn't always work out.

 

[Glen Powell]

I would check out the total tech spec of the optimum start up wattage before choosing solar edge, you may not be happy with the results, see if you can get a sale or return deal.

 

[LightGEN]

I haven't checked to see if the panels already installed are in their compatibillity list, but how about Enecsys Micro-Inverters? Enecsys had some production issues last year but these seem to have been resolved - we installed a system, based on these, at the beginning of December and we're happy with it so far.
SMA is launching a micro-inverter some time soon.

 

[Dave-geo]

A further update!

Since connected to one string to increase the voltage to suit the MPPT the increase in performance has been tremendous. The voltage is now fine at all times even when shading occurs however we still need to decide on a permanent solution.

I am really surprised at the problem the shading causes. I have placed below a picture of the complete roof and chimney to show our system. It is 14 Bosch 245w panels on a 50° roof facing due South.



With the three panels shaded as shown below the system will only generate 0.35Kwh



As soon as we reach the stage of shading shown below (less than 10 minutes after the previous photo) the generation shoots up to 1.95Kwh. Does that seem correct? I have been able to replicate this on every sunny day so it is not a one off occurrence.



There is a further progressive increase to a peak around solar noon of 2.85Kwh as the shade clears the final panel.

Is it possible to use any computer modelling to see which panels should be put in a separate string? It is ok looking at them now but more difficult for me to predict what will happen at other times of the year.

We are also still looking at SolarEdge as a way to overcome the problem but are concerned at the point Glen made earlier regarding the start-up wattage, I haven’t been able to find much information on this.

Any further help or comments would be really appreciated.

 

[LightGEN]

I've now checked and see four, apparently different, Bosch 245W panels listed in the Enecsys compatability list. All are model number c-Si M60 - M245 3BB followed by either 11, 13, 14 or 17 to differentiate the four different models.

 

[Gary K]

Do Bosch allow clamping on the short side? If not, I suggest you get your installer do do the job correctly so that you do not have warranty issues and whilst he has got the scaffold up( at his cost), you can fit the micro inverters as required.

 

[Dave-geo]

I have checked the specification sheet and as Gary says clamping is not allowed on the short side for these panels. What would be the reason for this and is it really something that should be corrected? To alter this now would be a large job and mean ripping the roof apart again which I am not keen on.

 

[SolarCity]

The reason is generally that the frame hasn't been designed to take the flex when clamped on the short sides. Some manufacturers have a stronger frame and they have passed their panels to be installed in this way - In reality, I doubt that you would have too much issue. However, I would say that if you left them this way then your warranty is effectively over.

 

[Earthstore]

I believe it is called fiscal loadings, ie will the panel break when it is covered with snow, also as pointed out it may well flex with the wind and break anyway.

The shading is not good to be honest, but as the sun gets higher in the sky the effects will be less, of course they will be worse all winter.
I hope this helps.

 

[Gary K]

I'm sure its a daft question but why couldn't you put the 2 panels that are on the right, to the right of the chimney under the solar water panels? The shading would then effect fewer panels as the shade obviously can't be to the left and right of the chimney at the same time.

 

[Dave-geo]

I'm sure its a daft question but why couldn't you put the 2 panels that are on the right, to the right of the chimney under the solar water panels? The shading would then effect fewer panels as the shade obviously can't be to the left and right of the chimney at the same time.

Gary, Thanks for your reply. I had thought aout that but as the roof faces Due South this might give shading to these panels in the afternoon/evening. I could then end up with a problem at both ends of the day!

The installer is calling around tomorrow morning so I am hopeful of working out a solution then. I will let you all know how we get on.

 

[Dave-geo]

I am still working with the installer to find a satisfactory solution to the shading problem, looks like we are going down the SolarEdge route.


I am not sure what to do regarding the panels being clamped on the short side. I really don’t want my roof ripped apart again! Would it be possible to add additional rails vertically attached to the horizontal rails which would then allow the panels to be clamped in the correct positions? This would seem to be an easy and cost effective solution.


Any comments good or bad would be appreciated.

 

[SolarCity]

I am still working with the installer to find a satisfactory solution to the shading problem, looks like we are going down the SolarEdge route.


I am not sure what to do regarding the panels being clamped on the short side. I really don’t want my roof ripped apart again! Would it be possible to add additional rails vertically attached to the horizontal rails which would then allow the panels to be clamped in the correct positions? This would seem to be an easy and cost effective solution.


Any comments good or bad would be appreciated.

Adding vertical rails is what we do on our installs. I don't know if your mounting system allows it but I'd be surprised if it doesn't. What is the make?

 

[Dave-geo]

Thanks for the reply Solar City.

I don't know the make but it is 40mm x 40mm and I have put a photo below of it. Do you reognise it?

Thanks for your help.

 

[SolarCity]

I could be wrong but that looks like Schuco to me. If it is then you will have no problem going for the cruciform method.

 

[Dave-geo]

I could be wrong but that looks like Schuco to me. If it is then you will have no problem going for the cruciform method.

That is good news. I have looked through a Schuco brochure but can not see that rail however looking at the accessories they list cruciform clips so I guess that is all they will need plus lots of extra rail!

Thanks again for your help.