Time delayed rcd’s and fault protection

dlt27 · Dec 5, 2019

Have been to a farm today that has a 100mA time delayed RCD main switch covering a DB that has MCB’s on majority of circuits ( less than 32A) and 30mA RCBO’s on sockets, lighting etc.
What I was thinking was does this comply with regs as circuits less than 32A should disconnect in less than 0.2s so because time delayed rcd fitted the circuits just on MCB’s don’t comply.
I thought to comply the RCD must trip within 200mS and before voltage rises above 50v.
Am I correct or am I being daft ( not for the first time )?

cliffed · Dec 6, 2019

Is it a tt system,if it is then mcbs may not comply,regarding the install of the cabling.
i would be incline to change mcbs to rcb’os on the mcbs.
The time delay would disconnect in 250 ms,I believe.

bill01803 · Dec 6, 2019

705.411.1 doesn’t mention time delayed

dlt27 · Dec 6, 2019

It is a TT system. Am I right in thinking if I I change time delayed to a normal 100mA then will be ok because it will trip within 0.2s, but if leave as time delayed I would have to fit RCBO’s to everything..
[automerge]1575618235[/automerge]
705.411.1 says just about values of RCD’s, but nothing about time delayed. Thanks

Wilko · Dec 6, 2019

dlt27 said:
What I was thinking was does this comply with regs as circuits less than 32A should disconnect in less than 0.2s so because time delayed rcd fitted the circuits just on MCB’s don’t comply.

Hi - check out BBB Appendix 3 Table 3A for maximum disconnection times. If I’ve understood correctly, you need 200ms and looking at the table some S types will do that.

loz2754 · Dec 7, 2019

dlt27 said:
It is a TT system. Am I right in thinking if I I change time delayed to a normal 100mA then will be ok because it will trip within 0.2s, but if leave as time delayed I would have to fit RCBO’s to everything..
[automerge]1575618235[/automerge]
705.411.1 says just about values of RCD’s, but nothing about time delayed. Thanks

If you change the RCD to a non time delayed type, you will find the RCBOs will not achieve selectivity with the RCD, and will cause nuisance tripping of the whole DB.
Bear in mind also that the disconnection time of 0.2 seconds relates to prospective residual fault currents significantly higher than the rated residual operating current of the RCD. Therefore, in line with Wilko's reference to table 3a of appendix 3, it is likely that in the event of a fault to earth, the 0.2 seconds disconnection time will be met.