Three phase equipment RCD protection on a TT system

[IAmSparkytus!]

TT system, existing three phase fuse board recently installed by another spark which only has single phase circuits terminated into individual RCBOs

Required to install three phase circuits for new equipment, is there a better way of doing this than fitting a 4pole RCD ahead of the main switch?

Conscious that this is going to cause total loss of power if a problem ever occurs.

No space nearby to install a seperate DB for the three phase equipment only.

 

[Risteard]

TT system, existing three phase fuse board recently installed by another spark which only has single phase circuits terminated into individual RCBOs

Required to install three phase circuits for new equipment, is there a better way of doing this than fitting a 4pole RCD ahead of the main switch?

Conscious that this is going to cause total loss of power if a problem ever occurs.

No space nearby to install a seperate DB for the three phase equipment only.

Go from the triple pole circuit breaker into an adjacent DIN rail enclosure with your four pole RCCB.

I assume that there's upstream fault protection, e.g. an S-type RCCB?

 

[davesparks]

What make/model of DB?

Some have 3 phase RCBO or RCD conversion options

 

[IAmSparkytus!]

Go from the triple pole circuit breaker into an adjacent DIN rail enclosure with your four pole RCCB.

I assume that there's upstream fault protection, e.g. an S-type RCCB?

No room adjacent to the distribution board to be able to do this as there are 3 x three phase circuits to install so 3 x RCCBs and enclosures would be required.

No upstream RCD but didn't think this was a requirement of BS7671 if meter tails are visible and proper glands and termination are in place?

Easy enough to add a 100mA time delay RCD upstream of the board if required but would still need 30mA RCCBs for each three phase circuit.

Would an upstream 30mA 4P RCCB be ridiculous?

 

[IAmSparkytus!]

What make/model of DB?

Some have 3 phase RCBO or RCD conversion options

Lewden .. dont think they do an RCD conversion kit unfortunately!

 

[davesparks]

Would an upstream 30mA 4P RCCB be ridiculous?

Yes, that would be a non-compliant design.

 

[davesparks]

Easy enough to add a 100mA time delay RCD upstream of the board if required but would still need 30mA RCCBs for each three phase circuit.

What are the 30mA RCCBs required for?

 

[IAmSparkytus!]

What are the 30mA RCCBs required for?

My understanding is that all circuits on a TT system are required to be protected by a 30mA RCD .. is this wrong?

 

[davesparks]

My understanding is that all circuits on a TT system are required to be protected by a 30mA RCD .. is this wrong?

This is indeed wrong.

Installations with TT earthing often cannot guarantee a suitable Zs to be able to meet the requirements for fault protection.
So in order to achieve adequate fault protection RCDs are used.

If an RCD is required for fault protection it does not necessarily need to be 30mA and it may also be acceptable to have a time delay.

All other requirements for RCD protection, additional protection etc, still apply.

 

[pc1966]

What is the nature of the 3-phase circuit you are installing?

If it is for a socket outlet up to 32A or buried cables less than 50mm from wall surfaces then it would need 30mA "additional protection", but if not (above 32A sockets or hard-wired) then often not but obviously depends on what it is. Then the use of a something 100mA delay RCD up-front would allow fault protection, but with the risk of everything going off should that happen.

 

[IAmSparkytus!]

Opted for a 100mA RCD to be fitted before the board after looking at table 41.5 of the BBB.

30mA additional protection not required for the nature of the three phase circuits. Cables not hidden anywhere and hard wired, no sockets etc.

With Ze value changing due to ground conditions, how close would you allow your Ze reading be to 167ohms before deciding that a 300mA RCD would be fine to use?

For example, if you were getting a reading of 165ohms for earth electrode resistance, would you be happy to install a 300mA RCD? What would be the figure you'd be OK with?

 

[pc1966]

Opted for a 100mA RCD to be fitted before the board after looking at table 41.5 of the BBB.

Hopefully time-delay "type S" as in series with the RCBO?

With Ze value changing due to ground conditions, how close would you allow your Ze reading be to 167ohms before deciding that a 300mA RCD would be fine to use?

For example, if you were getting a reading of 165ohms for earth electrode resistance, would you be happy to install a 300mA RCD? What would be the figure you'd be OK with?

That is a good question!

Personally it would depend on the conditions under which it was measured. For example if it was close to the limit during a dry spell of several days I would not be concerned, but if a similar value was measured when ground is soaking wet after recent rain then I would not be happy as it could well be double that in dry weather.

 

[IAmSparkytus!]

Hopefully time-delay "type S" as in series with the RCBO?

That is a good question!

Personally it would depend on the conditions under which it was measured. For example if it was close to the limit during a dry spell of several days I would not be concerned, but if a similar value was measured when ground is soaking wet after recent rain then I would not be happy as it could well be double that in dry weather.

The RCD provided was a Type A time delay which threw me a bit. It does have an S in a box shown on the details on the front of the device so I'm assuming it's a Type S as well if someone with more knowledge can clarify before works are carried out? Have managed to dodge a lot of work on TT systems so not particularly well versed in them!

In regards to the second part of your reply, the previous test certificate for the installation shows a Ze reading of around 70ohms. I'd say this would be more than acceptable for a 300mA RCD, would you agree?

 

[westward10]

If your circuits are 32A or less you must achieve a 0.2 disconnection time, not always possible with time delayed devices.

 

[pc1966]

The RCD provided was a Type A time delay which threw me a bit. It does have an S in a box shown on the details on the front of the device so I'm assuming it's a Type S as well if someone with more knowledge can clarify before works are carried out? Have managed to dodge a lot of work on TT systems so not particularly well versed in them!

Yes, the nomenclature related to RCDs is bonkers and gets worse with RCBO (is 'B' the magnetic trip or DC sensitive aspect?) so worth a check. Usually the RCD itself will have the symbol for Type A (both AC and half-wave rectified AC waveforms) and some indication of delay or not.

In regards to the second part of your reply, the previous test certificate for the installation shows a Ze reading of around 70ohms. I'd say this would be more than acceptable for a 300mA RCD, would you agree?

If it is 70 ohm then fine, but I would always check the circuit Zs anyway. You might find it is well below that due to bonding of service pipes (which cannot be relied upon for earth) in which case a separate Ze check would be best to verify the rod is still present & correct.

 

[pc1966]

If your circuits are 32A or less you must achieve a 0.2 disconnection time, not always possible with time delayed devices.

True, though the disconnect time is for typical fault currents which would usually be much higher than In for the RCD.

Still, not many MFT can test at high currents, I think mine maxes out at 500mA so OK for checking ultimate disconnect time for a 100mA device but not really for a 300mA device.