Up front Rcd question

[Dirk boddington]

Surely the first principle is to ensure earthing and bonding is all correct and the earth rod/mat etc gives the lowest Ra possible so any touch voltages are as low as can be? Or am I missing something?

As an aside the only RCDs I've ever known to fail testing were MK and Wylex ones, so maybe ensure you have good quality equipment?

Mine is wylex, go figure.

This is where my knowledge is hazy so correct me if I'm wrong. Is it possible get touch voltage down to acceptable levels on a TT system?

 

[Ian1981]

Mine is wylex, go figure.

This is where my knowledge is hazy so correct me if I'm wrong. Is it possible get touch voltage down to acceptable levels on a TT system?

On a tt system the equation Ra x Idelta N =<50 volts must be satisfied so yes you can.

 

[Dirk boddington]

So the rcd itself plays a major role in allowing this to be satisfied?

 

[Ian1981]

So the rcd itself plays a major role in allowing this to be satisfied?

Yes , without an rcd it's virtually impossible to achieve your disconnection times with a live to earth fault.
E.g. Ra of 150 ohms x 0,03 = 4.5 volts.
Ra of 150 ohms x just 0.4 of an amp =60 volts
50/0.03 = 1667 ohms max resistance before 50v touch voltage exceeded

 

[Risteard]

An RCD will have no effect on the touch voltage. All it will do is ensure that the fault is disconnected in the required time.

 

[D Skelton]

Dirk, there's a simple answer here mate. Fit a 100mA time delayed RCD up front.

This is the problem with domestic TT systems. More often than not they rely on 30mA RCDs for both additional and fault protection, so if one fails, which they regularly do, and a fault occurs, this is a dangerous situation.

The time delayed RCD isn't there to provide additional or fault protection but is there merely as a back up if one of your 30mA RCDs fails and a fault occurs.

The discrimination comes in the form of the time delay. If a fault occurs and your 30mA RCDs are functioning as intended, you'll want the relevant one of these to trip knocking out at most half your board. You don't want an up front RCD tripping unecessarily and knocking out your whole board.

There is no requirement to fit back up protection to a TT system, but it's very good practice. No domestic TT system I ever install is complete without it.

 

[Dirk boddington]

Thanks D. It wouldn't matter to me if the up front RCD tripped as my board isnt a split load, if my RCD trips now it takes the whole board anyway.

One of the sparkys I spoke to said two 30ma RCD`s affect each other and cause nuisance tripping, ive never heard of this before though, has anyone else come across this?

 

[Ian1981]

An RCD will have no effect on the touch voltage. All it will do is ensure that the fault is disconnected in the required time.

Not sure I follow as the equation Ra x Idelta N <50 v needs satisfying in a tt .

 

[D Skelton]

Thanks D. It wouldn't matter to me if the up front RCD tripped as my board isnt a split load, if my RCD trips now it takes the whole board anyway.

What would matter though is easy identification of a fault. With a time delay up front, if that trips you almost certainly know the 30mA one is buggered.

One of the sparkys I spoke to said two 30ma RCD`s affect each other and cause nuisance tripping, ive never heard of this before though, has anyone else come across this?

On test they can do yes. Two RCDs in line won't affect each other under normal operating conditions, you just never know which one will go first, if not both at the same time.

If you haven't got a split load board, you're not creating a problem by adding a 30mA up front, but I'd still recommend a time delay. Cost wise there's not a lot in it.

 

[D Skelton]

Not sure I follow as the equation Ra x Idelta N <50 v needs satisfying in a tt .

50V is the constant in that equation