Type D RCBOs

[johnnycash126]

Dear all,

I have to protect a 13 amp ring main which is currently protected by a type b rcbo.The circuit is going to be used by step down transformers from the US.Notorious for tripping on start up.I thought a change to a type D rcbo would cure this,however neither MEM or Crabtree make a type D RCBO.This makes me think that perhaps im not allowed to do this.

Anyone any experience of this?

 

[Darkwood]

The advice is similar from IET as NICIEC they confirm the use of the regulation but all give caution to using it as a easy option ...RCD's are affected by a number of mains transients and harmonics issues that can blind and mask the sensing coils plus RCD's are not fail safe and have a much larger failure rate by a wide margin compared to mcb's... these are just a few reasons to ensure where possible that Zs requirements are always met to 41.3 with regards to this discussion.

Im probably a bit ahrd minded about this regulation because it has caused quite a stir when introduced and had months of backlashes but at the end of the day its introduction comes down to the consumer is safer where installation don't meet regulations due to nothing more than poor design and incompetence of the installer and as this is becoming even more of an issue with DI wonder courses its implementation into the regs is to back up poor design where it should never have been needed in the first place.

In essence on a TN system this regulation was introduced as a plaster to repair an already bad wound but safety of consumer has to come first regardless of the reasons it exists.

 

[spark 68]

For TN-S or TN-C-S systems a Zs value significantly above that required by 41.3 could indicate a high resistance fault somewhere in the circuit which, in turn, could pose a fire risk.

Hi DW, that is (the bit colored black) is a different scenario where the Zs is marginally higher for the OCPD in question, and more to the point where it is totally expected by virtue of the correct (R1+R2) + Ze, and not a faulty circuit.

I agree with you in that I would not design a new circuit that way, unless as you say due to a unavoidably high Ze where you have no control.

Provided of course that the cable meets all the other requirements of BS7671 (ie. L-N loop, PFC etc.)

 

[Deleted member 9648]

This debate is NOT about using an RCD on a TN system to get over poor design or a fault on a circuit causing a high Zs. The circuit the OP has meets the requirements for a type B,and presumably has a Zs reading which would be expected for the circuit.All he wants to do is use an OCPD which will slightly exceed the max Zs for disconnection within 0.4s.
This debate has come up before and at no point have I ever suggested an RCD could be used to get over a poorly designed or faulty circuit.

 

[Darkwood]

Home now and I dont have the BGB with me,but unless it has changed since the BRB the above statement is not true.
The reg in the BRB that applies if I recall is 411.4.9

If anyone has access to the BGB and can check if 411.4.9 is still there it would clear this up. If 411.4.9 has been removed since the BRB I will stand corrected.

My bad Wirepuller was badly phrased when i printed-

Like we have been saying throughout this thread the 1667 limit is applicable to TT only and if your on a tn system then your rcbo has to meet requirements of table 41.3.

I meant to come across that you should always design a TN system to 41.3 the 'has to' was meant to say 'Should'

 

[spark 68]

TBF I wanted to know more about the VD theory in post #30 with RCDs and RCBOs on TN systems

I have difficulty with understanding that statement.

I would agree with DW that if you initially supplied the RCD/RCBO with a lower than specified Voltage that it may not operate in the expected way.

Assuming that the supply is normal to start with, and neglecting the test button/resistor arrangement for the moment, we know that all the conductors in the circuit have some resistance however low.

If we had a bolted L-E fault with negligible impedance the voltage would attempt to drop across the various resistances in the circuit, while this voltage is attempting to fall the current would also be rising proportionately at the same time, this has to happen by virtue of ohms law, this out of balance current and magnetic field through the RCD would surely have to trip the device ?

 

[Deleted member 9648]

My bad Wirepuller was badly phrased when i printed-

Like we have been saying throughout this thread the 1667 limit is applicable to TT only and if your on a tn system then your rcbo has to meet requirements of table 41.3.

I meant to come across that you should always design a TN system to 41.3 the 'has to' was meant to say 'Should'

Thanks for clarifying.....although the part that I take issue with is that the 1667 limit is applicable to TT only. Not according to 411.4.9 it isnt.That said at no point would I ever suggest that such values should be accepted on a TN. Clearly R1+R2 + Ze applies and Zs readings way off the expected would not be acceptable. But bear in mind that the max Zs for a 32a type C is 0.72....and for a type D 0.36. There are inumerable perfectly healthy TN fed RFC's which exceed these values and to which high inrush current equipment is required to be connected.

 

[spark 68]

Just to throw another grenade into this debate,

Something which DW alluded to earlier, although not specific or exclusive to TN systems, is the fact that certain items of equipment can "blind" the RCD, DC imposed on the lines for example, this saturates the magnetic core so it effectively cannot see the imbalance due to a fault, this is how the old D-Lok no-trip ELI meter works, you are then reliant on the OC part of the RCBO, which is another reason to aim as close as possible to the required Zs.

The RCD is no magic bullet, it may only mitigate a certain risk, and may be unreliable.

I once worked for a panel building company building a large number of DBs, (I was testing for them) the schedules for these various DBs used quite a few RCBOs, several hundred in fact, about 1 in 5 never met the full range RCD tests, either not meeting the times, or not tripping at all, some not turning on with no power applied. These were brand new out of the pack, and were made by a well known manufacturer, so not a cheap no-mark brand.

 

[Deleted member 9648]

In some ways,while this is a good debate, it has gone a little off piste. As far as I'm concerned the OP's question is simply can an RCD be used to meet disconnection times where the OCPD wont on a TN system. The answer,regardless of the technicalities,is that yes it can in full compliance with Bs7671. 411.4.9 applies.
The question of whether it should be permitted is another matter entirely,and is where this thread has gone....I'll leave that to those with more technical knowlege than myself.

 

[Darkwood]

Sorry for the disappearance it was that Valentines crap ... gotta keep her happy but yes Wirepuller we have gone off on a large tangent here so good point to end it maybe for another time and debate closer to the direction the thread went in.

 

[Engineer54]

The RCD is no magic bullet, it may only mitigate a certain risk, and may be unreliable.

I once worked for a panel building company building a large number of DBs, (I was testing for them) the schedules for these various DBs used quite a few RCBOs, several hundred in fact, about 1 in 5 never met the full range RCD tests, either not meeting the times, or not tripping at all, some not turning on with no power applied. These were brand new out of the pack, and were made by a well known manufacturer, so not a cheap no-mark brand.

No need to convince me about the reliability of RCD devices, and as you have found, only become noticeable when you are dealing with large numbers, rather than the one or two, or a couple more for most Domestic/small commercial installations. On projects the size i'm used to dealing with, we are talking in the thousands of RCD devices, and of all types and sizes. Like yours, ours are also from top end manufacturers....

 

[Engineer54]

Sorry for the disappearance it was that Valentines crap ... gotta keep her happy but yes Wirepuller we have gone off on a large tangent here so good point to end it maybe for another time and debate closer to the direction the thread went in.

You Romantic old so and so!! lol!!