Array frame earthing

[edexlab]

I would disagree with the statement ''The chances of the frames becoming live with AC is very slim"" and suggest you check the voltage from the array frame to earth rod/Met when using a TL inverter
I've checked this on most of the Tls I've installed which is a fair amount and found that on good days the voltage has been upto 126v down to 56v Ac (this does seem to vary with different makes of panels )being the lowest on an overcast day so in my opinion it's a good idea to follow the manufacturers guidance and bond Tl's ,and I also advise customers to turn the array off when cleaning just in case the bonding connection should ever be compromised as the Ac potential vanishes as soon as the system is turned off

Also most Tl's will normally have residual currents upto 50mA and with some over ,this is why you should'nt use 30mA Rcd's

 

[powervan]

u got more chance of killing urself keep messing with ur array!! best stop reading this and leave it to the EXPERTS!!!! let me no when u find one)))))))))))))))))) u makes ur choice takes ur chances!! or do what some senior members do in there spare time wrap urself in PVC (safe as houses) if not bit swetttty

 

[SibertSolar]

Interesting thread....

"Also most Tl's will normally have residual currents upto 50mA and with some over ,this is why you should'nt use 30mA Rcd's" - important observation. As is Edexlab's comments about PV arrays having a potential difference present when using TL inverters.

PV arrays will generate parasitic capacitive leakage currents (AC) naturally, especially when there is dew/moisture present in the mornings and things start to warm up. Combine this leakage current with anything inherently produced by the TL inverter and you can often "nuisance" trip a 30mA RCD. This is why SMA, Fronius, PowerOne etc recommend the use of a minimum 100mA trip limit RCD (300mA for PowerOne I think) when fitting their TL inverters. Whether this requirement conflicts with a pre-determined need to fit a 30mA RCD in compliance with wiring regs/BS standards (indirect contact fault protection, for example) is a debatable point.

Interestingly enough, I did hear that bonding your array to earth can exascerbate the generation of capacitive leakage currents by the array.

Btw, these slowly rising capacitive leakage currents can often not be detected by an internal RCMU (residual current monitoring unit) in the inverter as RCMU's typically only identify sudden rises in leakage current and will often not result in disconnection until max 300mA is reached. Relying on the inverter's built-in RCMU to act as a circuit protective (safety) device should not be recommended.

In addition to these naturally generated capacitive leakage currents, if a component of that leakage current is pure DC, with any greater than 6mA amplitude, a Type-A RCD manufacturer cannot guarantee that the RCD will continue to operate within it's required trip characteristics (required as a safety device to conform to its own standard/Reg 133.1.3). Some Type-A RCDs have been seen to have operated up to 30% over their intended trip limit when such DC leakage current components exceed 6mA pure DC. Type-AC RCDs will simply find themselves with a saturated trip coil and will NEVER trip, under fault conditions.

Anyway, I digress somewhat....my take on it is somewhat like this: If you have a TT supply, don't use a TL inverter - or if you can't mitigate against the BS7671 requirement for a 30mA RCD, don't use a TL inverter unless you can ensure that a 30mA Type-B RCD will function without nuisance tripping. If you can use a TL inverter, earth the array and fit a Type-B RCD for AC and DC circuit leakage fault protection. If you're concerned about the risk of lightning strikes then follow the risk assessment (as part of the recent BS7671 443 ammendment) and fit Type2 SPDs to the DC and AC circuits. If an LPS (lightning protection system) is already in place (or you have a TT supply) then fit Type1/Type2 SPDs accordingly depending on whether adequate separation distances are respected from the LPS and the array frame.

See here and here and here for some info that you may (or may not) find useful,

Andy

(don't shoot me if I've completely hijacked the thread, my apologies, it's been a long day and I'm a wee bit tired so apologies if the above makes no sense at all!)

 

[powervan]

WOW if ur tired hate to get lecture offfff u in morning

 

[moggy1968]

Aye, he's an intersting fella, knows his acorns! had a long chat with him the other day about RCDs and in particular about TT systems and RCDs IIf you remember andy we discussed wether a 100ma RCD was adequate on a TT supply if there was no other requirement for a 30ma RCD)

 

[SibertSolar]

Lol, indeed, I missed that point but was reminded by Bruce's comment on another thread, cheers

 

[Rw1001]

All mcs approved panels are class 2 double insulated. The dc cabling is also double insulated. For this reason I wouldn't earth the mounting system any more than I would earth metal guttering or tv aerial bracketry.​

 

[edexlab]

All mcs approved panels are class 2 double insulated. The dc cabling is also double insulated. For this reason I wouldn't earth the mounting system any more than I would earth metal guttering or tv aerial bracketry.​

I would suggest you read some of the threads regarding this subject if you have'nt already done so ,and the DTI guide and BS7671 is worth a look, and Siberts post above
and instead of assuming that it'll be ok you can test a TL array to earth and if you're happy not earthing it then good luck, and you'd better hope no one ever climbs a aluminium ladder to fix/replace the aerial (or clean the panels) and makes contact simultaneously with the array and ground