M
Maff
When installing a 'garage cu' adjacent to the main ccu to accomadate an RCD for connecting a TL inverter, what cable size would you use to link the cu's , less than 400mm run
- Thread starter Maff
- Start date
-
- Tags
- additional cable cable size size
G
Guest123
How & where you intend connecting to the existing install will govern what size cable you will need to use i.e what rating is the overcurrent protection at the point you are connecting to....is it the suppliers fuse or an MCB etc??j
M
Maff
I was thinking of replacing the Henley block for an isolator, and wondered if I could feed it in there and bypass the CCU, using the same size tails (25mm/100A incomer)
M
malcolmsanford
What size KWp is this system. As it is domestic I assume you are staying within the 4kw range of things then 16amp protection is all that is needed.
So is there a spare way on the existing CU, if there is use that as it will be cheaper for you client. Take a supply from the spare way to the new enclosure, protected by a 16amp MCB. Fit a Double Pole RCD, as you said on a previous post it is transformless type Invertor so you need a B type RCD.
If there is not a spare way, why do you want to remove the Henley Block ? Why do you want to feed the enclosure that is going to be maximum 16amp with 25mm tails? Are you planning feeding both the existing CU and the enclosure via this isolator you intend to fit?
So is there a spare way on the existing CU, if there is use that as it will be cheaper for you client. Take a supply from the spare way to the new enclosure, protected by a 16amp MCB. Fit a Double Pole RCD, as you said on a previous post it is transformless type Invertor so you need a B type RCD.
If there is not a spare way, why do you want to remove the Henley Block ? Why do you want to feed the enclosure that is going to be maximum 16amp with 25mm tails? Are you planning feeding both the existing CU and the enclosure via this isolator you intend to fit?
Not necessarily. Most TL inverters that I have used don't require B type RCDs.
M
malcolmsanford
M
Maff
it a 2.6KWp system. We are doing a ccu change, so was going to put in an isolator switch, its an SMA inverter which according to its specs says a type A RCD, from reading the DTI guide I got the impression that a, you have to oversize the cabling, b) you must keep the ac supply seperate from any other circuits
M
malcolmsanford
So you now going to change the CU completely, and still fit the garage type CU?. So again are you going to fit the one isolator for both CU and garage type CU ? or are you going to fit Isolator then a set of Henley's and feed the 2 points from there.
Again yes agree if you wished to fit 25mm tails to the main CU, but you may not need 25mm tails to the Garage CU have a look at Reg 434.2.1
point 1 then if the manufacturer instructions say A type is fine then good. Point 2 ...............hmmmmmm had this before.
M
Maff
the design is thus- From the meter 25mm tails to iso switch then 25mm to henly, 25mm from henly to CCU, and 25mm from henly to RCD in seperate enclosure+ lockable ac isolater etc,
the proposed 25mm cable in total will not be greater than 2 metres
the proposed 25mm cable in total will not be greater than 2 metres
M
malcolmsanford
That is a good design and how I would do it, though you will only really need 4-6mm cable for the RCD enclosure, makes life much easier than terminating 2 x 25mm plus your CPC into it.
Also you need overload protection remember, not just RCD protection.
Also you need overload protection remember, not just RCD protection.
B
BruceB
You can probably miss out the henley; just take both cables from the output of the isolator. The second cable only needs to be 4mm/6mm which is likely to fit in the terminals with a 25mm. It is protected from fault by the DNO fuse (unless the DNO has specified a minimum cross sect area).
The SMA installation instructions do NOT require a rcd. They say if an rcd is required then it must be a minimum of 100mA. And type a rcd is fine for SMA TLs.
Also, as there is only the one device on the circuit, the inverter, then there is no need to protect against overload from the mains side of it, only fault.
Regards
Bruce
The SMA installation instructions do NOT require a rcd. They say if an rcd is required then it must be a minimum of 100mA. And type a rcd is fine for SMA TLs.
Also, as there is only the one device on the circuit, the inverter, then there is no need to protect against overload from the mains side of it, only fault.
Regards
Bruce
M
malcolmsanford
I'm sorry, but I don't think that is correct. I'm ready to stand corrected however. Why is another form of seperation required?
M
malcolmsanford
If you don't fit an RCD B type as fault protection then you have to install an Inverter that will not allow DC fault currents back into the AC side.
So the use of simple separation by a transformer was the way round that.
If the SMA Inverters are transformerless, and do not need a Type B RCD protection, then they must have either another form of separation, or the capability of not to allowing those DC fault currents to pass through.
So the use of simple separation by a transformer was the way round that.
If the SMA Inverters are transformerless, and do not need a Type B RCD protection, then they must have either another form of separation, or the capability of not to allowing those DC fault currents to pass through.
B
BruceB
It's the latter with SMA inverters. The document the states it has been cited here a few times.
http://files.sma.de/dl/7418/RCD-UEN103120.pdf page 8
"All SMA inverters with transformer and all transformerless SMA inverters mentioned in the following are not
capable of feeding DC residual currents to the grid due to the circuit design. They fulfill this requirement in
accordance with DIN VDE 0100-712 (IEC60364-7-712:2002)."
Regards
Bruce
M
malcolmsanford
Well as said obviously the TL inverter must accomplish sepration by another means as I posted.
What worries me about this is that SMA are advising on German installation procedures in a UK environment. Though we have some harmonisation additional protection in the UK for instance cable that is buried and not to Reg 522.6.6 needs 30mA protection and if your installing as to reg 551.7.2 on the load side then again you need 30mA protection.
So though I won't go as far as to say that the SMA TL Inverter should still be protected by a B type RCD, I would rather be reading a manufacturers instruction that is based on UK installation methods and regulations.
Seems there are a lot of practices within the PV industry that is taken directly from Germany, another is the rod procedure for earthing an array frame if the earthing system is TNC-S. Here in the UK there is no problem with our stable PME system and I personally can not see an advantage of fitting a rod when you have a much more stable earthing system available to you.
What worries me about this is that SMA are advising on German installation procedures in a UK environment. Though we have some harmonisation additional protection in the UK for instance cable that is buried and not to Reg 522.6.6 needs 30mA protection and if your installing as to reg 551.7.2 on the load side then again you need 30mA protection.
So though I won't go as far as to say that the SMA TL Inverter should still be protected by a B type RCD, I would rather be reading a manufacturers instruction that is based on UK installation methods and regulations.
Seems there are a lot of practices within the PV industry that is taken directly from Germany, another is the rod procedure for earthing an array frame if the earthing system is TNC-S. Here in the UK there is no problem with our stable PME system and I personally can not see an advantage of fitting a rod when you have a much more stable earthing system available to you.
B
BruceB
Well as I have stated in another thread I do not believe you should be connecting PV on the load side.
I am not sure you should blame the Germans - it was our dear DTI that published their guide. Personally I agree with you and think connecting TL framework to TNCS MET is arguably less risky than putting an additional rod in. But I would not do it that way whilst I have MCS assessors looking at my work.
Regards
Bruce
I am not sure you should blame the Germans - it was our dear DTI that published their guide. Personally I agree with you and think connecting TL framework to TNCS MET is arguably less risky than putting an additional rod in. But I would not do it that way whilst I have MCS assessors looking at my work.
Regards
Bruce
M
malcolmsanford
The DTI seems to have literally copied the German practices and fair do's if that is the way they want to go and you need to dance to that tune you have no option.
I'm not blaming the Germans they are doing the installs to their standards, I expect if the German Department of Trade and Industry were forcing UK standards on German electricians there would be a couple of Germans berating this as we are, an such a forum.
It still seems to me that it is the wrong thing to do, unless of course we intend to adopt more European practices, or because the industry is still in it's infancy, they may review installation practices at a later date, and for now adopted their methods.
What is worrying me is that it is putting this magical 100mA RCD in installers minds as an accepted norm, where here in the UK additional protection is by a 30mA RCD. I'm just surmising that perhaps not so savvy installers will not realise that they are installing systems that require additional protection by 30mA devices and fitting just 100mA as the SMA catalogue advises.
I'm not blaming the Germans they are doing the installs to their standards, I expect if the German Department of Trade and Industry were forcing UK standards on German electricians there would be a couple of Germans berating this as we are, an such a forum.
It still seems to me that it is the wrong thing to do, unless of course we intend to adopt more European practices, or because the industry is still in it's infancy, they may review installation practices at a later date, and for now adopted their methods.
What is worrying me is that it is putting this magical 100mA RCD in installers minds as an accepted norm, where here in the UK additional protection is by a 30mA RCD. I'm just surmising that perhaps not so savvy installers will not realise that they are installing systems that require additional protection by 30mA devices and fitting just 100mA as the SMA catalogue advises.
I think that's correct. Seperation isn't needed by some kind of mechanism to prevent DC fault currents certainly is.
Is the German PME system less stable than ours? I didn't know that.
The grey area with PME/PV systems is a nightmare. I can see both sides of the argument but currently we are advised to install a seperate earth rod and it's what we've been doing until we get told otherwise.
G
Graeme Harrold
The G83 test sets limits on DC injection back into the AC system and is set at 20mA, which is aptly detailed on the type test report.
The German grid is considerably better than in the Uk the inbalance limit is 4.6kw over there
M
Mark Davey
We have installed a Fronius IG35 PLUS today (8 x HIT Sanyo 235W) does anybody happen to know whether this also has built in protection that means a type B RCD is not required??
Also I read and hear differing views on the bonding of the array mounting system & module frames.
In no uncertain terms the tutor on the Eco-skies course I attended stated that only ground mount systems require the bonding and that this is achieved using a seperate CPC to a dedicated earth rod, he suggested that roof mount systems do not require bonding due to "placing out of reach", a definative answer on this would be great.
My final question is a very simple one, we have designed the installation in 2 strings and would like to know if the string fuses required or do the individual modules have satisfactory protection so that the solid links supplied from the factory OK.
I felt that my training was really not comprehensive enough and left so many basic questions unanswered so I have enroled on the C&G2399, but until such time a big thanks in advance to all the other forum members and site administer as your information which is much appreciated
Also I read and hear differing views on the bonding of the array mounting system & module frames.
In no uncertain terms the tutor on the Eco-skies course I attended stated that only ground mount systems require the bonding and that this is achieved using a seperate CPC to a dedicated earth rod, he suggested that roof mount systems do not require bonding due to "placing out of reach", a definative answer on this would be great.
My final question is a very simple one, we have designed the installation in 2 strings and would like to know if the string fuses required or do the individual modules have satisfactory protection so that the solid links supplied from the factory OK.
I felt that my training was really not comprehensive enough and left so many basic questions unanswered so I have enroled on the C&G2399, but until such time a big thanks in advance to all the other forum members and site administer as your information which is much appreciated
W
Worcester
Check the DTi Guide, with only 2 strings it would be very unusual to need to fit string fuses so long as the cable is rated high enough.
A type B RCD would not be required as the inverter already offers electrical seperation between the array and the AC side of the installation due to the transformer in the IG35 PLUS. You need only worry about type B RCDs when using transformerless inverters. Generally, the same is true when considering the bonding of an array. An array installed with a transformerless inverter must be bonded - using an earth rod if the incoming is a TN-C-S system.
R
Ramjam
I just mined my way through this issue and this is what I discovered, I can only speak for the SB4000TL......
The Germans mostly don't earth array frames....
All panels induce some form of current in the array frame and in a transformerless inverter this current leaks over the the AC side.
The total leakage from a 4kWp system in never likely to exceed 80mA
The SB4000TL inverter has a 100mA type B RCD contained within it, should the DC leakage to the AC side exceed 100mA then the RCD will trip with a fault code on the inverter which can be inspected by an installer using Sunny Explorer.
The DTI guide demands that the array frames of systems with transfomerless inverters be earthed, in the case of PME systems this means to a dedicated earth spike rather than the MET (for me this has always been the case).
BS7671 requires that additional protection is provided by a 30mA RCD, in germany 100mA is acceptable, not here!
If you fit a 30mA RCD or connect to a CU with a 30mA RCD you run the risk of leakage from the array frame making it's way through the inverter without tripping the type B RCD in the inverter and nuisance tripping the 30mA RCD. To avoid the nuisance tripping fit a good earth rod and connect it well to the array frame (every rail).
SMA struggle to accept that their inverter should be protected by a 30mA RCD as required by BS7671 as their solution to nuisance tripping is to fit a 100mA RCD instead. They are just learning about the UK market and this is one of their grey areas.
The SB4000TL is a brilliant inverter, get a 1m SDS drill bit, an earth rod, inspection pit and some 10mm earth and lugs and way to go. Remember there are some simple ways to test earth rods, I shan't mention them because TT experts want you to go to Hogwarts before you can fit earth rods but it's pretty easy once you get the hang of it and with a dedicated PV earth rod the cost of TT failure is nuisance tripping not death so you don't need to be terribly scared.
The Germans mostly don't earth array frames....
All panels induce some form of current in the array frame and in a transformerless inverter this current leaks over the the AC side.
The total leakage from a 4kWp system in never likely to exceed 80mA
The SB4000TL inverter has a 100mA type B RCD contained within it, should the DC leakage to the AC side exceed 100mA then the RCD will trip with a fault code on the inverter which can be inspected by an installer using Sunny Explorer.
The DTI guide demands that the array frames of systems with transfomerless inverters be earthed, in the case of PME systems this means to a dedicated earth spike rather than the MET (for me this has always been the case).
BS7671 requires that additional protection is provided by a 30mA RCD, in germany 100mA is acceptable, not here!
If you fit a 30mA RCD or connect to a CU with a 30mA RCD you run the risk of leakage from the array frame making it's way through the inverter without tripping the type B RCD in the inverter and nuisance tripping the 30mA RCD. To avoid the nuisance tripping fit a good earth rod and connect it well to the array frame (every rail).
SMA struggle to accept that their inverter should be protected by a 30mA RCD as required by BS7671 as their solution to nuisance tripping is to fit a 100mA RCD instead. They are just learning about the UK market and this is one of their grey areas.
The SB4000TL is a brilliant inverter, get a 1m SDS drill bit, an earth rod, inspection pit and some 10mm earth and lugs and way to go. Remember there are some simple ways to test earth rods, I shan't mention them because TT experts want you to go to Hogwarts before you can fit earth rods but it's pretty easy once you get the hang of it and with a dedicated PV earth rod the cost of TT failure is nuisance tripping not death so you don't need to be terribly scared.
M
malcolmsanford
Great post Ramjam and it outlines what I'm saying, the BS 7671-2008 tells us that the manufacturers instructions takes precedence, but for PV inverters most of these instructions are geared at the German electrical system not our UK one.
The 30mA RCD is one and the use of a rod is another. Because of the use of the 30mA RCD for additional protection then in the UK I would be very wary of using an Inverter that actually feeds DC current back into the AC system and if they leak in excess of 20mA again I don't think we should be using them, unless you can ensure that the installation that as been erected does not need 30mA protection. I can see now why the DTI are insisting Inverters be installed on a dedicated circuit, but surely if you use TX type there is no need for this as the separation will not allow DC currents back into the system. So I gather that have blanketed the installation method.
The point I want to also raise with the earthing is the use of a rod for a PME system. There is no concern these days with the exporting of the earth in a PME situation in the UK. Many years ago when the Electrical safety Regulations were in force it was advised to contact the DNO and seek permission, but today if you called the DNO to ask they would most likely be incredulous about your request. With the advice now on also earthing TV aerial masts, granted it is recommended more for communal installations, but many people are advocating all masts to be earthed, you could get a very dangerous situation of an earthed mast of a PME/TN-S internal system and an earthed array of a TT system, within touching distance of each other and IMO that is introducung a far more dangerous scenario.
The 30mA RCD is one and the use of a rod is another. Because of the use of the 30mA RCD for additional protection then in the UK I would be very wary of using an Inverter that actually feeds DC current back into the AC system and if they leak in excess of 20mA again I don't think we should be using them, unless you can ensure that the installation that as been erected does not need 30mA protection. I can see now why the DTI are insisting Inverters be installed on a dedicated circuit, but surely if you use TX type there is no need for this as the separation will not allow DC currents back into the system. So I gather that have blanketed the installation method.
The point I want to also raise with the earthing is the use of a rod for a PME system. There is no concern these days with the exporting of the earth in a PME situation in the UK. Many years ago when the Electrical safety Regulations were in force it was advised to contact the DNO and seek permission, but today if you called the DNO to ask they would most likely be incredulous about your request. With the advice now on also earthing TV aerial masts, granted it is recommended more for communal installations, but many people are advocating all masts to be earthed, you could get a very dangerous situation of an earthed mast of a PME/TN-S internal system and an earthed array of a TT system, within touching distance of each other and IMO that is introducung a far more dangerous scenario.
We phoned the DNO on one of our first TL installs to ask them about exporting the PME onto the roof. They had no idea what I was talking about and one guy advised me to ask an electrician's advice!
D
dansk
M
malcolmsanford
Yes but seemingly no according to the DTI.
I can only relate to normal electrical practices and if you were in a domestic installation and taking a sub main out to an out building, if that building have metallic parts to it either a metal service such as a water pipe, or the construction of the building was metal, then you would need to bond the extraneous conductive part back to the MET if you were exporting the PME earth. If you were not exporting the earth perhaps because the building was a long distance from the origin of the supply, then most likely you would TT system.
There is obviously a reason the German electrical industry do not export earths on a PME, to which I don't know, and so it seems they insist on the array being bonded via a separate rod. I have no problems with this as such, as long as you are not having other metal work at a different potential that can be touched while your working or touch the array frame, and I used the example of perhaps an aerial mast that is earthed via a PME/TN-S from within the equipotential zone.
I can only relate to normal electrical practices and if you were in a domestic installation and taking a sub main out to an out building, if that building have metallic parts to it either a metal service such as a water pipe, or the construction of the building was metal, then you would need to bond the extraneous conductive part back to the MET if you were exporting the PME earth. If you were not exporting the earth perhaps because the building was a long distance from the origin of the supply, then most likely you would TT system.
There is obviously a reason the German electrical industry do not export earths on a PME, to which I don't know, and so it seems they insist on the array being bonded via a separate rod. I have no problems with this as such, as long as you are not having other metal work at a different potential that can be touched while your working or touch the array frame, and I used the example of perhaps an aerial mast that is earthed via a PME/TN-S from within the equipotential zone.
R
Ramjam
I'd suggest that the most common earthed metalic object on a roof is the metal flue of a central heating boiler.
D
dansk
I find I a bit of a headache trying to the do the right thing, I spent hours last night researching pme and bonding, it seems a very grey area to me, but my thoughts are if it's roof mounted take the bond to the met rather than a rod, let's say somehow you could reach the array from inside the zone it would have to be bonding to the met anyway. But did come across this on my travels last night Geosolar DC Earthing which explains in simple terms
D
dansk
What if you was holding onto the scaffold and the array at the same time and it was bonded to pme? Would that cause a shock?
D
dansk
Sure - But isnt PME "multiple earth" so the neutral conductor is earthed in stages along the line from the supply? But if broken neutral in CU then is a problem
Last edited by a moderator:
M
malcolmsanford
In section 9 of the ESQCR-2002 covers PME in LV installations. The DTI impose stringent guidelines that the distributor must adhere to when supplying the PME system, and one of these is the loss of a neutral. The distributor take measures against the loss of the neutral conductor, and it is therefore seen as a low risk occurrence, which is why exporting an earth is not seen as potentially dangerous as it once was.
D
dansk
M
Mark Davey
[h=2]Re: additional cu cable size[/h]
The issue is, as I see it, that in the event of a fault which took the neutral out at supply, the array would become 'live' and potentially lethal to anyone working outside of the property.
not only the array but also all bonded exposed conuctive parts?
Similar threads
OFFICIAL SPONSORS
These Official Forum Sponsors May Provide Discounts to Regular Forum Members - If you would like to sponsor us then CLICK HERE and post a thread with who you are, and we'll send you some stats etc
Advert
Thread Information
Thread Tags
- Tags
- additional cable cable size size