Problems with a new solar install

[heliosfa]

The Dual MPPT inverter means that the voltage of each string can be entirely different.

The manual implies that this is a single MPPT?

 

[MCS Renewables]

The inverter Installation manual says minimum AC cable size for this inverter is 4mm.

So AC cable undersized , DC Isolator dubious at best as marked AC in photographs, DC wiring poor, cable through roof shoddy at best although Gavin A suggests some roof cooling benefit !, mounting rail fixing bracket screws not in meat of joists so shoddy, Cable terminations shoddy, no generation meter labelling, no electrical certification left on site despite adding new mains board and circuit.
Plus massive HSE issues and some major shortcomings in the MCS paperwork system.

This is after the installation company have been out to sort out the issues !

Sorry but they had their chance to do the project right first time and then the re visit chance to put it right. They believe they are finished and want payment.

Companies like this need reporting and sorting out by RECC, NAPIT, Trading Standards and the HSE.

 

[whinmoor]

I'd also suggest getting them to change the inverter, if possible. A Power One Aurora PVI-3.0 OUTD would be a more sensible choice for this in my view.

 

[Gavin A]

The manual implies that this is a single MPPT?

my mistake, I thought that was the dual MPPT model. In which case their should be the same number of panels on each string, and cables lengths should be as similar as possible.

 

[whinmoor]

It looks like there's a chimney that may cause at least some shading. A twin-tracker would be certainly be preferable.

 

[Gavin A]

The inverter Installation manual says minimum AC cable size for this inverter is 4mm.

but it's going to be peaking at around 3kW not 3.68kW due to the array size. The cable loss diagram on the 3300TL version (which is equivalent to the actual output here) puts 2.5mm2 at being below 1% cable losses until around 17m.

Either way, that;s not a safety issue.

So AC cable undersized , DC Isolator dubious at best as marked AC in photographs, DC wiring poor, cable through roof shoddy at best although Gavin A suggests some roof cooling benefit !, mounting rail fixing bracket screws not in meat of joists so shoddy, Cable terminations shoddy, no generation meter labelling, no electrical certification left on site despite adding new mains board and circuit.
Plus massive HSE issues and some major shortcomings in the MCS paperwork system.

This is after the installation company have been out to sort out the issues !

Sorry but they had their chance to do the project right first time and then the re visit chance to put it right. They believe they are finished and want payment.

Companies like this need reporting and sorting out by RECC, NAPIT, Trading Standards and the HSE.

It's pretty rough, but I don't see any immediate safety issues that would result in it needing to be locked off as you suggested - that's just causing unnecessary alarm for the customer IMO.

And there's really no point in involving anyone else until the companies been given a proper chance to correct all the issues, as they won't do anything until this has happened IME.

- - - Updated - - -

It looks like there's a chimney that may cause at least some shading. A twin-tracker would be certainly be preferable.

agreed - partly why I'd assumed that was what the 3680TL was. I'd obviously given them too much credit.

 

[Gavin A]

EDIT2: should both strings be the same length to have the same voltage? the installers said that they were running the top 4 as one string and the lower 6 as a second. I take it that the easiest way to check would be a volt meter?

if this is what they've done then they have properly messed it up if it is a single tracker.

yes a multimeter will tell you what's going on, but be careful as there can be quite a flash from the DC, only do it if competent to do so, and only when the isolator is off please.

 

[MCS Renewables]

GAVIN A ....are you saying that you would ignore the manufacturers instructions and install 2.5mm cable because you did some calculations !!

 

[heliosfa]

I've done some more digging and reading of spec sheets and MCS standards - if I have anything wrong, please shout as my speciality is Electronic Engineering, not electrical...

The DC isolator could well be rated to accept the relatively low voltage and ampage from the 2 strings of these panels with the links removed - we use some k&N isolators in this way with approval from K&N, and it will be low voltage and low ampage, so there would seem to be some conclusions being jumped to here without evidence to support them.

From the spec sheet for the "DC isolator" (here, page 11), the maximum voltage for purely resistive loads is 300V. For mixed loads, it is 250V or 600V for two poles in series.

The panels have a rated load voltge of 54.7V and an OC voltage of 64.9V, giving a total of 273.5V, 324.5V open circuit, if the pannels are connected as two strings of five or 328.2V, 389.4V open circuit, if the pannels are connected up as they said (6+4). From this, two poles in series are required?

The AC cable probably is UV rated for external use, and I'm not convinced it is particularly undersized for a 3kWp system - it's certainly not dangerously so, though it could result in the inverter tripping out due to overvoltage if the local voltage is relatively high.

From the MCS PV installation guide (here, page 44) the recommended voltage drop is 1%. Using an online calculator, a 15m run of 2.5mm^2 seems to give a voltge drop of about 2% at the peak AC output current of the inverter.

The choc block for the earth within the meter is pretty standard, it's allowed because it's within the extended cover of the meter, and there is no earth connection within those meters, so something along these lines has to be done.

Agreed, i was more commenting on its presence than suggesting it was a problem.

It's also the case that in a lot of houses the loft isn't adequately ventilated after having additional insulation fitted, so the extra ventilation through the hole is likely to be beneficial rather than a problem - most damp and mould issues in lofts is from condensation due to inadequate ventilation following loft insulation top up IME.

This loft has plenty of ventiation as appropriate gaps were left when top-up insulation was done.

 

[heliosfa]

It looks like there's a chimney that may cause at least some shading. A twin-tracker would be certainly be preferable.

Yes there is a chimney. The property is oriented pretty much South-West. The original suggested panel layout (with a mix of portrait/landscape pannels) looked to have more spacing between the chimney and the panels but would have been a bit ugly. The layout was changed to the all portait on the morning by the installers.

 

[Gavin A]

GAVIN A ....are you saying that you would ignore the manufacturers instructions and install 2.5mm cable because you did some calculations !!

scroll down the page a little and you'll see a section specifically stating that cable losses should be less than 1%, and a diagram showing the volt drop for different lengths of different CSA cables, which I referenced earlier.

OK so the diagram is for the 3300TL, but as I said, that's more appropriate here anyway as the system isn't going to be outputing significantly more than the panels rated output very often / at all.

Not that we'd be using 2.5mm2, I'm just pointing out that it's not dangerous to do so, and looks to be somewhere in the region of within manufacturers recommendations and below the 1% max volt drop MCS recommend. Certainly too close for me to want to call it so definitely over the internet as being undersized.

 

[whinmoor]

Assuming that the first photo was taken at about 10 or 11am, I'd say that the top right panel would suffer from shading for a period of time every morning. You need to reduce the effect of this by using a twin-tracker.

A 6-4 split on the Samil is far from best practice. It really needs a 5-5 but the shading will really affect output in the morning.

Get them to swap the Samil for a Power One 3.0 as suggested above.

 

[Gavin A]

I've done some more digging and reading of spec sheets and MCS standards - if I have anything wrong, please shout as my speciality is Electronic Engineering, not electrical...



From the spec sheet for the "DC isolator" (here, page 11), the maximum voltage for purely resistive loads is 300V. For mixed loads, it is 250V or 600V for two poles in series.

The panels have a rated load voltge of 54.7V and an OC voltage of 64.9V, giving a total of 273.5V, 324.5V open circuit, if the pannels are connected as two strings of five or 328.2V, 389.4V open circuit, if the pannels are connected up as they said (6+4). From this, two poles in series are required?

It looks to actually be an AC rated isolator to me from the code, which would be a concern and you'd definitely be within your rights to demand that be swapped out. Hard to say if the DC version would be ok with the links removed or not, could be, could not be, I'd want to check with the manufacturers.

From the MCS PV installation guide (here, page 44) the recommended voltage drop is 1%. Using an online calculator, a 15m run of 2.5mm^2 seems to give a voltage drop of about 2% at the peak AC output current of the inverter.

sunny design puts it at 1.13% with an equivalent inverter at 15m, or 0.98% at 13m. So it's borderline at best, but not actually a safety issue.

This loft has plenty of ventiation as appropriate gaps were left when top-up insulation was done.

maybe so, but a small bit extra isn't going to hurt in itself.

It's a common misconception that felt is there primarily to stop rain getting in - it's not, its main purpose is to prevent wind blown snow getting through into the loft, and help prevent the tiles getting sucked off. Water proofing is a secondary function, but really if it gets to the point that the felt has water running down it then you need to be fixing the tiles sharpish anyway or the battens will have rotted and you'll have to strip the lot instead of replacing the one original broken tile.

 

[heliosfa]

Assuming that the first photo was taken at about 10 or 11am, I'd say that the top right panel would suffer from shading for a period of time every morning.

THe first photo was taken at 12:16. Shading at 10:18 in the morning can be seen in new piccy, taken just before installation started. The return calculations were done by the company with no shading.




One of their "maintenance engineers" has been back once to fix the AC cabling issues and was asked to check the rest of the cabling for any other issues. So they have had an opportunity to identify and fix some of the issues. I also had to remind this "maintenance engineer" to isolate parts of the system before he started undoing cables and he had all of the covers off before isolating the main AC - I think this goes some way to justifying not having them back? I would rather not have to deal with a fried contractor in my cupboard under the stairs...


When they did the install, they cut the seals on both the main fuse and the normal meter (both were sealed by the local network operator about 3 weeks ago as they did a meter swap). As I understand, SSE permit the seals to be removed as long as the person doing it is a member of one of the trade bodies, that the person removing the seals contacts SSE before doing so and that temporary seals (issued by SSE) are put in place. Both the meter and the fuse have been left unsealed and the original installer would not contact SSE - he told me to do it.

When I mentioned this to the company's "contracts manager" he just said that they were allowed to cut the seals as they can't work live, but he would contact the DNO to tell them that the seals were cut and needed replacing.

 

[heliosfa]

Doing some more reading (not good, the more I read, the more I dispair...) what should the initial survey have consisted of? What happened with mine seemed to be mainly the sales lady taking photographs and spending more time talking about what the different panels were and how the Sunpoweres were the best.

Should I have seen any of the calculations they were meant to have done? especialy as the roof is hipped and they should be consulting a structural engineer (page 72 of the MCS guide).