Solar pv batteries distance to CU

[binky]

the inverters are usually sealed against moisture and the fan forces air over cooling fins, rather than through the electronics section itself.Most now work on 'natural convection' ie they have more cooling fins and no fan. Your right though, the fan would probably help with condensation, and as the inverter will off load some heat, probably enough to keep it dry. Ventilation for the box, i would try and make the box large and airy, vents at top and bottom, but fit vents with anti-insect mesh or you will have every spider, woodlouse and snail making it their home. 

 

[revor]

Have now got Q- cells panels lined up the G6+ and G8+ are made in Korea. Europe is supplied from China Korea and Malaysia. Will have to settle for 340w ones the USA factory produce even greater output 450W think is the biggest.  My "box" is a mini insulated metal clad shed 2M x 1.2 x  1.4 High. No ventilation yet as not worked out how much I need but have got the vents to do it. Had a max min thermometer in there and min been 14 deg and max 25deg. Now got everything lined up to procure. (Whilst I have asked for replies to be notified, since part way through the thread not been getting them, is there an issue somewhere. Checked my spam box nothing in there.)

 

[binky]

 Now got everything lined up to procure. (Whilst I have asked for replies to be notified, since part way through the thread not been getting them, is there an issue somewhere. Checked my spam box nothing in there.)

Don't know about the website, it may be playing up a bit as all things computer do!

Q Cells are nice panels, German cell design which they used to sell on to panle assemblers. They were bought out by the Koreans a few years ago to assemble complete panels. Buils quality is superior to Chinese and performance seems reliable/ good without resorting to expensive stuff - lot of 'bang for your buck'. 

LOcate shed in shade of panel array of you can, and I suspect a few eaves vents top and bottom will work fine. 

 

[revor]

Notification working again for me. Have ordered the Q panels and Renusol ground mount system. Was looking in detail at the inverter manual and the voltage drop spec is very tight at 1 % so by my reckoning it would mean I would need a 35 mm SWA from the incoming grid to the inverter. Not going to be possible. Had been informed that 10 mm would be more than adequate.  I presume that this extract from the manual means voltage drop. The power dissipated on the line must be lower than 1% of the nominal power. There  is a table of length vs cable size  and for 6 mm it is 15 M.

Something I had not appreciate and something that I thought would need to be done as an extra is that in the event of power failure from the grid it switches automatically to battery and auto disconnects from the grid and also has facility to switch on a generator. I do not know all that much about inverters but have looked at some but don't recall this feature on any of them the only off shelf system am aware of is the Tessla back up Gateway, so makes the inverter good value after all.

 

[binky]

biggest issue with smaller size cables is it can cause voltage to rise, so if you are near top limit on your grid supply, then it can push the voltages higher and cause the inverter to shut down.

I had a job where the supply is 16mm over a 110m run, the customer wasn't keen on shelling out £3k for the new cable or digging long trenches on the access track to his barn, so we fitted the system anyway.  The system was underperforming forecast, so we checked the grid voltages and they were at top limit,  so we complained to Western Power. They have since replaced the pole transformer and dropped the set voltages down to a much lower level, and everything has worked fine since. The customer now wants another set of panels on the same barn roof, which I wasn't keen on fitting, but, I calculated the resistances and the subsequent likely rise in voltages. I was surprised how little difference it made, and it would keep the system voltage well within spec, so we wil be fitting a second set of panels sometime soon. Energy dissipation was around 80/100W over the length of the cable - a light bulb basically. I doubt it would actually see that full energy loss, as the customer has a plant room about half way for his bore hole, and from this room there is also a supply to his jacuzzi, which will be programmed to take advantage of the solar.

So whilst we like to avoid more than 1% losses over the cable, in reality it's OK in the right circumstances. At normal 3% 16mm would be fine at 100m run

 

[revor]

Thanks for that I will read in detail and discuss with the UK distributor of the Inverter. I am awaiting reply to an earlier email. It may be an Italian thing rather than an UK thing. I will probably be able to get a 16 mm down the 50 mm duct  there are 2 90 deg bends to negotiate but could dig down to these if becomes an issue. I had just come on here now to say that all the effort I have put in and help given by others particularly you Binky that despite my best endeavours the panels that arrived today are made in China and not Korea as I was told.  I keep finding these can of worms.

 

[binky]

Thanks for that I will read in detail and discuss with the UK distributor of the Inverter. I am awaiting reply to an earlier email. It may be an Italian thing rather than an UK thing. I will probably be able to get a 16 mm down the 50 mm duct  there are 2 90 deg bends to negotiate but could dig down to these if becomes an issue. I had just come on here now to say that all the effort I have put in and help given by others particularly you Binky that despite my best endeavours the panels that arrived today are made in China and not Korea as I was told.  I keep finding these can of worms.

 hope ypu have a pull rope n the duct?, If the duct is smooth bore, it shouldn't be too much proplem, Fairy liquid helps.

Q Cells from China! That's those off my xmas card list - disappointing after all the blurnb about being Korean., especially given the current climate and backlash against Chinese products.

Your welcome for all the help. 2 thoughts:-

1/ use 2 core cable and earth rod the array so you can use 25mm SWA

2/ use split concentric cable, but that is a bit naughty as only yhe DNOs are really allowed to use that stuff (and my garage becuase I didn't know any better at the time) . Physically this cable is much a smaller diameter. 

 

[revor]

I have been in touch with the guy who told me what panels were made where, and suggested he arranges a swap. Think he was a bit defensive as he suggested I should have asked the supplier where they were made! But fingers crossed he sorts it. Morally and even legally I think he should then all can see on this forum how good they are at responding to a customer good press better than bad press.

 We still need to properly spec the cable so your thoughts here are another input. The split concentric is out of my league (at moment anyway not sure want to down that learning curve particularly if DNO's are only allowed to use it. )

The wiring of the inverter shows, feed in from the grid and feed out back to the distribution unit which would be 4 core I guess. As mentioned  before on a different thread one installer that I was really keen on and I really liked him (which helps) had used SY cable to carry DC from the array to the inverter in a shed some 40 m away on an install he showed me.

Beginning to think now would be better to house the inverter and battery nearer the house. I have a reasonably sized outside WC which once had a washer and dryer in there it also has a convenient CU  but would need to upgrade the cable to the house approx 10M easy to do as in a duct. That would mean bringing DC from the array some 60 m.  I came across SMA Sunny boy designer which has a solar pv "calculator" looks pretty nifty have loaded my data into it but will need to study the manual to understand the answers properly. Looks like a super tool to anyone designing solar PV systems and easy and free to download from SMA. Have you used it Binky? The cable sizing part of the tool is what I am interested in. Having the inverter nearer the house removes that uncertainty w.r.t voltage drop and sizing particularly as got the feed from grid to inverter and back again to the CU. Have 3 CU in the house (did a lot of radials) so could reconfigure them so battery feeds the important circuits. However as we only use 5-6 KWh a day and power cuts never last 2 hours might leave it as is and in prolonged cut need not use the cooker etc. All lighting is LED and all appliances A or A+ rated. The solar pv is the last of our contribution to minimise carbon foot print have stuffed the house with insulation. MVHR, solar gain, thermal mass, solar thermal panels (had free hot water since end of February) thermal store have LPG though but boiler hardly on but no brainer considering cost and maintenance of pellet boiler and heat pumps.

I have got rope through the duct will change it though as are not rot proof, never tried fairy liquid as lubricant in this way isn't it messy?. Find talcum powder / French chalk good any damp around that might cause stiction it dries it off.

 

[binky]

SMA design tool is pretty nifty. I normally use PV-sol, but that isn't free! 

Don't use SY - it's  not designed to be buried, unless you can gurantee no water in the ducting, it's also dearer than SWA.

Should only need a 3 core for single phase, just share battery and inverter on the same cct straight back to board, or it's own board after the house meter, the inverter doesn't use separate cabling to feed in and out

 

[revor]

Apologies for not being back sooner have had emergency health issues family wise.  Not intending at all to spec SY only mentioned it as I had come across a potential installer who had used it either deliberately or erroneously  although I was told  it was used because it being very flexible it threaded through ducting easily.

Q cells agreed to swop the panels and the originals have gone back to the wholesaler but not had replacements yet. I keep chasing.

Having trouble spec'ing the DC feed cable from the panels to the inverters due to my lack of knowledge on DC.  Seeing as the fore mentioned installer had used an unapproved multicore to carry the DC I am thinking a properly specified SWA would do the job. I will have 2 strings of 9 x 335 W  40. 5 V panels 60 M from the inverter so thought a 4 core SWA of the correct size would do the job. In trying to gen up on DC keep coming across a formula I do not quite understand the which includes a factor of 2 and some other bits in order to correctly size the CSA of the cable. So am at a loss on this so far. What would you use in that situation?

As for the inverter AC in out feed I was thinking same as you but the instruction manual clearly shows the wiring with separate feeds. Think it may be due to the fact that this inverter will switch automatically to battery in the event of a grid failure and disconnect from the grid at same time.

Shame you are not near to me will need someone to connect me up once got all the stuff installed.

 

[binky]

cable sizing is  a bit different on DC systems, as few articles refer to it except for car/ leisure battery systems at much lower voltages. You are working at 360V and around 8 Amps per set of panels,  so even a cable as small as 1.5mm will handle the ampage. (general rule, the higher the voltage the smaller the cable) SWA is usually rated at around 1000V or 600V, so can easily handle 360V (albeit under stalled conditions the voltage can be higher, but not by much - look at Voc figure). so really we are just looking at voltage drop / cable losses, for which this fomulae is standard, but based on 240V so will oversize cable https://the-regs.co.uk/blog/?p=481. I'll dig out my regs book later and do some numbers. 

 This tool seems fine https://www.solar-wind.co.uk/info/dc-cable-wire-sizing-tool-low-voltage-drop-calculator  and would suggest 6mm is fine. Up it to 10mm if you like as any reduction in voltage drop improves performance. 

 

[revor]

Thanks very much that is great. I visited the sites you sent and used the solar wind one. Coincidentally I had found that one the other day but did not work for me as was inputting the wrong info. Played around with it and yes 6 mm will be a good size for both the DC to the inverter and for the inverter to the house. Both at the 1% voltage drop level and having measured more accurately the length from the array to the inverter it is 55 M so a bit of advantage gained there. Won't get any advantage upping to 10 mm there is  a cost and bit  more work possibly pulling through the duct. There are sets of values in the spec sheet which I need to  understand a bit more on and go back to the calculator to double am check using correct values. Have got 2 values for amps Isc Impp in each of 2 tables presumably for either end of the performance curve . There is quite a difference from 10.47 to 7.84 so makes a difference to the cable rating.

 

[binky]

most of the solar farms are wired in 6mm on the DC cabling running at 600-800V  on runs of 100s of metres. OK DC cable is a different spec, buts it's still only 6mm . 6mm is a lot easier to pull than 10mm, 10 Amps is nothing to that cable, the tables in the back or the regs book state cable ratings are for DC and AC. 

 

[revor]

Thanks for that Binky did not think of looking in the tables have got the yellow edition for 17th edition (table 4d4) and have calculated for 55m for the DC run at 10 A it is 1.1% voltage drop and .8% with 10 mm cable. So as concluded earlier not much point going down the 10 mm route. The 15 M from the inverter to the CU at  26 A 230 V comes out at 1.2%. Had in my head that DC needed higher spec cable don't know where it came from. In my teens once worked on fairground rides and remember being told to keep away from the big DC motors that powered the rides as DC more dangerous than AC and I could get "fried". Perhaps came from that time. Got everything spec' d now just need to have the panels as promised and order inverter and battery. My gut tells me could have issue of getting them as promised as not getting my chasing messages returned. Many thanks for your much valued guidance in steering me through this Maze. Much appreciated.

 

[revor]

Been ages making any progress and have got the array frame erected and by next week should have all I need to do the job. I have quite a dilemma on what the earthing arrangement should be. The IET code of practice for grid connected PV says for a  TN-C-S earthed house the panels and frame should have a earth rod and that connected to the TN-C-S earth. (No mention of TT or TN-S) From what I can make out the target impedance is quite different for both so what  happens when they are connected together. I spoke to the technical authority on the inverter and he explained that there is quite a lot of confusion around PV earthing with many different approaches. (In the guidance notes there is an acknowledgement that there are complex arguments as to whether or not and array should be earthed but don't go into them.) He recommended doing the earthing to the ground rod but not earth the armour of the SWA carrying the DC to the inverter but to ground the inverter to the house in normal manner. He also mentioned that many installations are not earthed at all which might explain why I have been finding difficulty getting hold of earth ancillaries for solar PV. (Anyone know of source of Stainless Steel crimp terminals without having them especially made). Any other views on this earthing conundrum?

 

[binky]

OK, there's never been a requirement to earth panels or framework except where you have Micro-inverters behind the panels, on TN-C-S, which means you have AC cabling behind the array itself, and the metal framework represents an 'extraneuos conductive part' ie a short could make it live. Guidance may have changed since I last read that, but no real explanation was ever given as to why. 

As you have a ground array (I'm assuming it's steel) you have lots of earth rods in the ground already. If you wanted to make sure you have a good earth, I would use copper strip, as per ligtning conductors rather than a rod, and galvanised self tapping screws to connect to the frame.  I would connect steel armour of the SWA to the frame, but not the inverter end. Your inverter is on the side of your house, so connect that to the house earth. With TN-C-S we are not allowed to 'export' the earth to another building away from the property, which is where TTing a shed or garage comes from. I wouldn't be looking to connect house earth and framework (earth rod).  Given a faulty Neutral in your house supply, you could find your frame becomes live. Your inverter will look for DC faults every morning as it starts up, in particular it looks for shorts to earth, known as RISO in the PV industry. Anything less than 20 MOhms and it won't start up. 

I've worked on many solar farms, these are not earthed on the DC side. I would install a DC isolator at the array end of the SWA, (you need something to terminate the SWA to anyway) so in the unlikely event of damage to the SWA, you can isolate it to work on. 

 

[revor]

Many Thanks.  Proving very interesting have had about 4 takes on this so far. The ground array is of aluminium bolted with marine grade SS to concrete blocks concreted into place and then back filled with stone. (See photo) I have driven in 4ft of rod but don't have a suitable 3 prong meter to test  it, my Metrel multi-tester unlike some does not have the facility. A local installer will be doing the important bits I need for warranty and MCS so will get him to test and if needs be drive another rod. I am intrigued how you would use a lightning conductor strip. I am not convinced just burying would work the ground at the surface is stony but below that you hit clay which doesn't half grab. So how is the strip used? I did pick up that DNO don't like you using their earth yet the recommendation in the IET code of practice is to do that. The consensus of opinion is not to connect to the TN-C-S which to me also makes sense but all opinion is contrary to that of the guidance notes. Each of the strings will have isolators close to the array housed in a mini shed. 

 

[binky]

the rod doesn't need to be 'brilliant' it's only real function is to remove static / induced voltages from the frame, so that you don't get a tingle off the frame if you touch it. Any such charge would probably dissipate through the concrete blocks anyway. Lighting conductor strip is usually buried in the ground, ie dig hole / trench and bury it. As you have done the gorund works already, stick with the rod. I've yet to see any ground array with a dedicated earth, it's not that critical / unnecessary / standards are unclear   :^O .   The ones I've fitted used ground anchors - frame sat direct on the ground, and what is effectively  large corkscrews driven into  the earth.

 

[revor]

Thank Binky most helpful now sorted can get on with job this week weather permitting.( Don't know why photo is there again.)

View attachment 11139

 

[revor]

By now hoped to have been up and running all is connected up apart from a meter  and to the grid. Been let down  empty promises etc. However there is something I am not clear about. The inverter instructions mentions it would be advisable to be connected to a Type A 100 mA RCBO  presumably at the  the rating of the unit i.e 32 A.  I find it confusing in that the letters could mean the curve type or the RCD type and have not been able to get a satisfactory answer. Besides can't see listed 100mA RCBO' type A anywhere

The exact wording in the manual is.

If in compliance with local standards or in in specific cases the use of an external circuit breaker is required we recommend that a type A switch  with fault current of at least 100 ma is used.

DELIOS s.r.l. declares that the construction of the DLS systems does not produce earth leakage direct currents and therefore, in compliance with article 712.413.1.1.1.2 of section 712 of Standard CEI 64-8/7 a type B circuit breaker with standard IEC 60755/A 2 is not required.

What does this mean?

Thanks