TN-C-S/Adiabatic help!

[frontier219]

Afternoon all,
Got a bit of a strange one (by my usual standards anyway).
Job is a large barn conversion, done years ago, there's loads of work to do in the outbuildings as they are renovating those, but they aren't touching the house just yet. Main intake is a brick built building about 2-300m from the house at the end of the drive. The system is three phase, on a TT earth, and there is a 300mA RCD (no time delay) protecting the submain.
The cable from the submain is a 95mm 4 core SWA that is buried and then reappears inside one of the outbuildings, where it's terminated into a busbar chamber and then feeds various switch fuses and consumer units.
I was wondering about the possibility of asking UK Power Networks to move the system over to a TN-C-S so that I can remove the upfront RCD, and avoid nuisance tripping. Apparently there have been a couple of instances over the years of faults tripping the main RCD, and the owner having to go all the way to the end of the drive (a pretty steep hill with no lighting) to reset it.
The only potential issue I can see with this is that the SWA would then be required to carry any fault current, and I haven't looked at the Adiabatic since college! I've found a few snippets of information that have been helpful but they have generally been based on exam questions rather than the real world.
Would anyone be happy to give me a couple of pointers and check my workings? Or am I barking up the wrong tree completely on this one?

 

[pc1966]

The system is three phase, on a TT earth, and there is a 300mA RCD (no time delay) protecting the submain.

Well that is your first problem. Without a delay on the up-front RCD you won't get good selectivity with and down-stream RCD on any sort of "hard" fault that causes more then 300mA to flow.

The cable from the submain is a 95mm 4 core SWA that is buried and then reappears inside one of the outbuildings, where it's terminated into a busbar chamber and then feeds various switch fuses and consumer units.

According to the data sheet I have to hand that has 147mm of steel in armour, so about 18mm copper equivalent. That is enough for PME bonding up to a certain size (don't have regs here to say quite what) but for sure a 100A 3P supply is fine on that aspect.

Would anyone be happy to give me a couple of pointers and check my workings? Or am I barking up the wrong tree completely on this one?

If you post your workings folks will help out. You have consider:

• Will the Zs at end of cable be sufficient to disconnect in under 5s on a sub-main fault? You need to know the feed fuse to check that as well as the R1+R2 for this cable, and a guess at what TN-C-S earthing might offer (usually below 0.35 ohms for 100A sort of circuits). If you have the existing measured PSCC at the supply point you should be able to estimate that Ze as your TN-C-S earthing ought to be the same.
• Based on the supply fuse/MCCB and Zs, what sort of fault I2t are you seeing and then if you meet the adiabatic limits.
• Do you have enough conductor to satisfy PME bonding conditions? This also then applies to feeds to other buildings and in turn all of their extraneous bonding conductors.
• Are any of the buildings from this needing TT earthing due to livestock, etc, special areas?

The lack of delay/selectivity on the up-front RCD is your key TT issue just now. That and possibly a lack of other thought-through RCD protection on other circuits. Depending on what you are looking at you might even want to change the supply up-front RCD for an adjustable one that can be set to something like 300mA & 0.5s, and then have selective 100mA incomers on all sub-DB, and finally 30mA protection for anything needing it (which is most stuff these days).

 

[davesparks]

At 300m that puts the installation a long way from the nearest PME electrode, this may be something to consider.

I think you would be better off sorting out the selectivity of the RCD's installed rather than trying to change the earthing system.
A variable RCD at the start and some sensibly placed 100mA time delayed RCDs should sort it out.

 

[westward10]

Have you done a line to neutral Zs, this will give you a Ze for a change to TN-C-S and ADS. Much to consider as already mentioned.

 

[frontier219]

Thanks for all the advice.

I've only taken a L-N Zs at the consumer unit closest to the busbar at the moment and that came in at 0.35 ohms on the nose so I'm guessing it will easily be good enough at the main intake.

I'm assuming I need to take one at the main intake to work out my PEFC for the adiabatic, would that be correct?

I agree that sorting out the selectivity would be a good solution too, but there are plenty of circuits in the house and other parts of the building that aren't yet being renovated (the guy has a three year plan, it's a big place) that it would be prudent to remove the need for an upfront RCD if at all possible.

 

[bennie]

Thanks for all the advice.

I've only taken a L-N Zs at the consumer unit closest to the busbar at the moment and that came in at 0.35 ohms on the nose so I'm guessing it will easily be good enough at the main intake.

I'm assuming I need to take one at the main intake to work out my PEFC for the adiabatic, would that be correct?

I agree that sorting out the selectivity would be a good solution too, but there are plenty of circuits in the house and other parts of the building that aren't yet being renovated (the guy has a three year plan, it's a big place) that it would be prudent to remove the need for an upfront RCD if at all possible.

You need to take your reading from where the mains enters the property, if your taking it from a sub DB your probably measuring parallel earths. You need to know your Ze not Zdb. Also if the earthing system is changed you'll probably have to upgrade the size of the main bonds.

 

[LastManOnline]

At 300m that puts the installation a long way from the nearest PME electrode, this may be something to consider.

Why would the distance to the PME,s electrode matter?.The general mass of earth is a conductor of negligable resistance.The usual considerations taken in to account are the resistances of each individual electrode (installation and PME,s) to the immediate ground they are sunk in to.Whether the distance between them is 300 or 3000 meters should,nt really matter.