Pond pump tripping RCD?

[staithes14]

at the end of the garden, about 40m from the house, is a pond. In the pond is a 320W pump that pumps water up to the top of the garden (4m head) which then runs down a stream to the bottom pond again.
From the CU at the house, the supply runs via a RCD through a buried armoured cable to a shed at the end of the garden. There, it goes through a second RCD which feeds two MCBs; one supplying the shed lights and the other the shed sockets plus, via a fused switch, an outdoor switch bank.
The outdoor switches (triple) feed an outside light, outside socket and the pond pump.
Recently the RCD up at the house tripped. It wouldn’t reset so I had to leave it like that for a few hours. When I had another look, it reset no problem. Everything ran ok then three days later, same happened. Again, after a ‘rest’ it took the reset.
It’s done this a few times now and the only thing that’s been drawing power on that circuit has been the pump. Yet the pump runs perfectly for days between trips (we run it constantly) so I’m not convinced the pump is faulty. I cleaned it out and the impeller was only replaced 18 months ago, and the pump motor is all sealed. If water got into the windings, would it show this sort of tripping pattern? I’d expect it’d trip and that would be that.
The only variation was the second time it happened, when the RCD in the shed tripped, rather than the one at the house end of the circuit. That suggests it’s not a faulty RCD.
It tripped just now, as I was typing this, after running fine since Saturday morning!

 

[winston1]

For a start you don't need two RCDs in line. You never know which "or both" will trip.

Anyway it seems it is not a faulty RCD so I suggest the pump. Cleaning it out and replacing the impeller won't fix an electrical leak.

 

[oldbutnotdead]

And not necessarily the pump causing the problem. 2 RCDs on 1 circuit is wasteful, but since the shed one tripped we can probably rule out a fault on the underground cable.
So you need to inspect (and test) all the circuits beyond the shed RCD.
That you've only been using the pump could be irrelevant- you still have live running from MCB to lightswitches (possibly to the fitting if its been wired that way) you still have live running from MCB to shed sockets, you still have neutral connected to all appliances. Remember RCDs can also trip on N-E faults

 

[Taylortwocities]

Whssign

• Taylortwocities
• 16 Jul 2016

What he said

 

[staithes14]

Yeah it was the sparky who put in the two RCDs. I thought it seemed unnecessary. I’ve visually inspected everything (with power off) and there’s no loose wires, no corrosion, damaged cable, nothing amiss.
I isolated the lights using the MCB and a trip still occurred with them isolated. So the fault is on the circuit that feeds the sockets and pump. Or the pump itself. After last nights trip, I reset the RCD this morning ok. Pump running fine at the moment.
I guess it’s down to testing each part of the circuit now.
A replacement pump is £180

 

[staithes14]

I’m not sure how to test the pump itself in situ (it’s hard-wired) but I guess an obvious thing to do is physically take it off-circuit and see if the trip still occurs.

 

[Taylortwocities]

It has been mentioned above, but I will repeat.
RCDs will trip if there is a problem on either the live or the neutral. Most switches, and MCBs only switch off the live conductor. To properly isolate a circuit you would need to switch off the MCB and physically remove the neutral conductor.

I’m not sure how to test the pump itself in situ (it’s hard-wired) but I guess an obvious thing to do is physically take it off-circuit and see if the trip still occurs.

To test in situ, you need an Insulation Resistance Tester. Indeed, that is the only proper way to find a fault like this. Disconnecting and waiting and hoping is a long, frustrating way of going about things.