Garage RCD

[muffking]

I don't know, but I wonder if being locked up at home is making me more critical... But this makes no sense to me. How does the impedance come into it? The same current difference is seen in each RCD.

Feel free to tell me my thought process is wrong, but does a higher ZS at the garage RCD not affect the amount of current flowing through the CPC compared to the house RCD in the event of earth leakage in the garage?

 

[EFLImpudence]

The RCD knows nothing about the CPC.

It only measures the difference in current between of the Line and Neutral.

 

[Taylortwocities]

Whssign

• Taylortwocities
• 16 Jul 2016

What he said

 

[muffking]

I should have worded that differently as I referred to the ZS, but I really mean the impedance created by the longer circuit overall.
In this case, if there's earth leakage at the garage does the impedance of the circuit not affect the L-N imbalance?

 

[EFLImpudence]

Yes, but No.

Plus the impedance of the cable from house to garage will affect the load on the house RCD differently to the garage RCD.

I see what you are getting at.

A high impedance will affect the current throughout the cable. It won't be higher at the house than the garage.

A fault downstream of the garage RCD will look the same to both both RCDs.

 

[muffking]

Thanks.

I think there's some mileage in working this out to clarify why as I have it in my head that circuit impedance after the RCD would affect the current flow to earth.
I'll draw a diagram to try and work this out.

 

[EFLImpudence]

I think there's some mileage in working this out to clarify why as I have it in my head that circuit impedance after the RCD would affect the current flow to earth.

It would - but not relevant as both RCDs will see the same if the fault is downstream of the garage RCD

If the fault is between the RCDs then the current might be higher but the garage RCD will not see anything.

 

[flameport]

Circuit impedance will affect the current flow to earth at any given point - a fault to earth near the origin will have a higher current than a fault a significant distance away at the end of a cable.

However that doesn't affect which RCD trips - regardless of the distance, both RCDs will have the same current flowing through them.

 

[muffking]

Thanks guys.
Just spent the last 20 mins drawing this out and frying my brain trying to understand how 30mA missing from the Neutral would be affected as the r1+r2 changes.

I think I got it when I realised that the second drawing (B) didn't take account of the total impedance to the load being the same where ever the RCD's are situated, so current draw would be the same.
You've got to love diagrams

 

[SUNRAY]

are you making the mistake that an earth-leakage can be "cancelled out" - balanced - by the additional current or a similar leakage on the 'other' conductor?
That is, of course, if the garage is two kilometres away.

Many years ago a farmer friend moved a wooden shed for every picking season and used a thin twin cable (I'll guess it was 1mm²) along the edge of the track from where it was in the winter to the new location, this was a 20 minute walk so I'll guess it was around 2/3 mile. It had a CU with RCD as main switch feeding 1 socket and 2 lights (one inside, one outside) via cartridge fuses but the socket fuse had long since blown and not been replaced. He simply connected the twin to the CU and an earth spike as he had done every year. The lights worked which was all that mattered to him.
The inside light was on all day and the outside bulkhead all night.

One year the RCD in the shed tripped if it rained but not the RCD at source. When he returned the shed to the winter location he reconnected it as usual including original earth spike. Now when it rained it randomly tripped either or both RCD's but only if a light was on. The actual fault was the socket wiring (run externally in plastic conduit) had got damaged in the first move and the rain simply filled the tube and socket and shorted neutral to earth.

I can't remember the exact details but I did do loads of measurements and calculations and got to the place where I found it can happen. I think I replaced the source RCD but it was too far back for me to remember.

 

[ajohn]

The usual answer when an electrician fits a garage cu with and rcd in it and there is also one on the feed in the house cu is to remove the garage rcd as there in no point in having it.

The answer is a feed from the house cu via a fuse or mcb that isn't protected with an rcd but that means the feed must run in swa in the most recent regs. It's effectively a sub main.

It's also done by splitting the meter tails to power a separate switched fuse unit. That is the more usual arrangement. Sort of traditional.

The fuse or mcb then needs a rating greater than the sum of mcb's in the garage cu sized to trip independently and sums need to be done to make sure it trips fast enough if there is a feed cable fault. One of them can be tricky - live to armour. That can mean an rcd is needed as well. Delayed or greater than a 30ma rating - ok as it's for cable protection. Live to neutral is usually no problem. The 30ma one is in the garage cu then and it will work independently.

It all probably works out a lot easier when the house cu is fitted with all rcbo's - easier to get a feed without rcd protection. Just needs an mcb for the garage and also not much of a problem to add an rcd to that circuit if needed.

Or feed the garage via an rcbo fitted bypassing the house rcd.

I think there is another option but it's rather contentious.