Wet UFH pumps causing RCD tripping

[ACL]

Since when did we start listening to wet pants?

We don't listen to them, but customers will listen to the last person in the building unfortunately.
It took a while but I finally convinced the customer and the plumber that the 100A RCD didn't need to be upgraded.

 

[UNG]

Turning on the UFH, the wiring centres go through a start up process then send power to the Neostats. They in turn go through a start up process and then turn on sending a signal back to the wiring centre to turn the boiler on. It's at this point that the RCD trips.

The RCD generally trips over night so other appliances are on but not being used.
Forgot to mention this is a new build and all appliances are new.
I will try to move to an RCBO to test.

One post seems to contradict the other here, you can't start the system but then it trips overnight

RCD ramp test indicates that it's tripping around 19mA.
1/2 test indicates no problems.

If a brand new RCD is tripping at that level I would be speaking to the manufacturer about a replacement, out of the box problems with RCD's seem to be getting more common in the last few years

 

[mattg4321]

Is that 19ma ramp test with all loads disconnected?

 

[brianmoooore]

IIRC, the acceptable range for a 30mA RCD is 15mA to 30mA. I.e. 30mA +0%/-50%.

 

[mattg4321]

Possibly it is. Imo most decent rcd I’ve ramp tested with all loads disconnected are about 24-27ma. 19 is low imo. Perhaps because of cumulative earth leakage, perhaps by design, perhaps faulty.

OP says it’s Hager. Usually these are around 25ma in my experience and reliable and good quality

 

[ACL]

One post seems to contradict the other here, you can't start the system but then it trips overnight


If a brand new RCD is tripping at that level I would be speaking to the manufacturer about a replacement, out of the box problems with RCD's seem to be getting more common in the last few years

The initial tripping was intermittent and would happen overnight. The system would operate happily for 3-4 days then trip out.
This led me to believe that a fault had developed on one or more of the heating components as the customer was turning off the 2 top floor heating system and no tripping would occur.

It was only when I was on site and tripping happened that I made the discovery that the tripping would happen when the UFH called for the boiler to fire and the pumps started up.
On Friday the system would always trip with all the UFH systems on so the 2 top floor UFH systems were switched off.
After my initial discussions on here I asked the customer 2 switch off the ovens and the hob, at the wall, on that side of the CU. They then turned on the 2 top floor UFH systems and no tripping occurred.
It was left like this from 11:00 until 16:00, then they switched the ovens and hob back on with no tripping.
Until they got up this morning and the RCD had tripped.
I'm now heading down the path of cumulative leakage current. More testing required.

 

[ACL]

Is that 19ma ramp test with all loads disconnected?

Yes all load MCBs switched off and tested at the CU.
I've also contacted Hager to check on my findings.

 

[brianmoooore]

You need to (temporarily, at least) move the heating to a different RCD, and see if the tripping follows it.

 

[ACL]

You need to (temporarily, at least) move the heating to a different RCD, and see if the tripping follows it.

That's my next plan, however I have ovens on both sides of the CU so need to work out installing an RCBO without too much disruption to the customer.

 

[ACL]

Thank you by the way to all of you.
Just being able to bounce things around has focused the mind.

 

[mattg4321]

19ma is low for a Hager RCD imo. I'd try moving it on another RCD.

It seems unlikely as I've fitted hundreds with extremely few problems, but maybe the RCD is faulty.

 

[Wilko]

Hi - just my quick thought - I have had more than one boiler pump causing trips so I’d work my way through all the pumps one by one doing dead testing of IR.

 

[Rockingit]

No leakage clamp meter as yet, any recommendations?

No leakage clamp meter as yet, any recommendations?
There is no N link between the boiler and the UFH as the wiring centres use a 0v built in relay to send the demand signal.
The remote Grundfos pumps and 2 port valves are controlled by the boiler spur, so I could separate the UFH and the boiler to 2 RCBOs

Good ones aren't cheap, sadly. I had a Metrel one (can't remember the model) that was amazing as it would even give you an indication of the nature of the current, lead/lag, pf etc etc which was genuinely helpful in understanding what was going on in the circuit, cost about £400ish ten years and some pikey stole it in a van break-in. Now you can get good ones for half that.

 

[ACL]

Good ones aren't cheap, sadly. I had a Metrel one (can't remember the model) that was amazing as it would even give you an indication of the nature of the current, lead/lag, pf etc etc which was genuinely helpful in understanding what was going on in the circuit, cost about £400ish ten years and some pikey stole it in a van break-in. Now you can get good ones for half that.

Thanks for that, I've been reading reviews and ordered the Megger DCM305E £220.
Hopefully it will help to get to the bottom of my problem.

 

[brianmoooore]

It's always good when we can prove beyond doubt the cause of a fault with our nice expensive test gear, but once in a while you'll be up against a truly intermittent fault that only shows very occasionally and for a split second.
In these cases you have to set up experiments, such as moving to a different RCD, and then wait for the fault to gradually give itself away .