Tripping Patio Heater Circuit

[Kris_88]

Hi All, having an issue with what I thought would be a pretty straight forward problem at first if anyone can shed some light.

This particular circuit breaker trips when you turn it on via double pole switch, but only if it has been off for about 10 min or so. It only has to be reset by the breaker once and it's fine, you can then switch the double pole switch until your hearts content as long as it's not left off as explained above for over about 10 min. Details of circuit below:

Small restaurant has a 2.5mm Two Core & Earth FP200 Radial, clipped direct (15m run) via a 20A double pole switch feeding two 2kW dimples patio heaters fed from a 60898 Type C 20A breaker

Now for stuff I've tried:

Full Range of tests and visuals carried out to circuit and all satisfactory
New Breaker - Same Problem
New Double Pole Switch - Same Problem
Pick up a new heater and swapped with each heater one by one - Same Problem

Note: When one heater in disconnected out of the two, whichever one doesn't make a difference, circuits ok
Note: This does two heaters on one side of a restaurant, there is an identicle circuit which does the other side which is fine.

Id really appreciate any help if anyone has come across a similar problem, if it's something silly then I'm officially signing up to plumbing school.

Kris

 

[HandySparks]

What's the resistance of the heaters when dead cold?

Could be that as these are radiant heaters, which get pretty hot, the resistance of the element might be significantly lower than expected when cold. Maybe there's just enough start-up surge to trip the breaker (although I'm surprised for a type C).

Are the heaters on the other side on a longer run, which is adding just enough resistance to keep the start up current in check?

 

[Pete999]

Have you carried out any electrical tests on the circuit??

 

[Kris_88]

Thanks for the reply guys. Pete999, yes I disconnected heaters and done insulation resistance between all conductors, all above 999 throughout and for what it was worth a loop test as well. HandySparks, I was thinking along the same lines as originally this was on a B type breaker, so I changed it to a C type as EFLI was pretty good anyway. I'll be honest I haven't measured resistance of the lamps but having said that the other circuit is about the half the length and no issues :-S. I also amp clamped this circuit on switch on and it was only about 30-35A for a split second.

 

[Marvo]

...originally this was on a B type breaker, so I changed it to a C type

MCB or RCBO?

 

[Pete999]

MCB or RCBO?

Kris I'm sure I don't have to say this but you can't just go swapping types of CBs what if the disconnection times aren't right? the circuit may need redesigning to cope with the changes, I'm only saying, not trying to be clever.

 

[HandySparks]

Kris I'm sure I don't have to say this but you can't just go swapping types of CBs what if the disconnection times aren't right? the circuit may need redesigning to cope with the changes, I'm only saying, not trying to be clever.

I think you're teaching this guy to suck eggs. See below:

... originally this was on a B type breaker, so I changed it to a C type as EFLI was pretty good anyway.

I'll be honest I haven't measured resistance of the lamps but having said that the other circuit is about the half the length and no issues :-S. I also amp clamped this circuit on switch on and it was only about 30-35A for a split second.

Still think the cold resistance would be a useful value to measure.

Does the current clamp have a peak hold or were you just eyeballing the display?

 

[JameZZZ]

Could it be ingress of moisture in a connection near or in the heaters? The initial heat from the split second the circuit energises may be burning off the moisture?

 

[Kris_88]

The breaker was 60898 Type B20, measured the EFLI and was well within the limits to be a Type C. HandySparks, it was just a standard fluke meter with a clamp so I just had to eyeball it, didn't expect to get much from it but thought I'd compare the two. But as you mention above the longer I left the heaters off, the larger the inrush when I swithed it on but from what my eye could catch it was nothing major. The job may end up being passed to a collegue but if I do return I shall look into the resistance of the lamps, thanks for your input. JameZZZ, thanks for your reply, each heater had a pre-wired flex which was terminated into a waterproof adaptable box, from what I could see they were glanded well enough and didn't look like it's had any moisture inside, they are quite well covered as well. As its a relatively short run and about 90% of the circuit is visible I'm sure there's no other connection points. Thanks for your advise though lads!!

 

[Marvo]

Something's not adding up. If it's tripping a C curve breaker instantaneously then were talking about an inrush current of 5 x In which is around 100A.

I don't possibly see how you could get that kind of inrush current on a heater circuit which is to all intents and purposes a linear load.

 

[Kris_88]

Well that's the issue, i considered this as a possibility at first, but as you mentioned above it requires a hefty inrush so I kind of passed on the idea.

These are not on timers so generally I think the managers will just turn them on of an evening and leave them on until they close. I think the fact that the circuit breaker will only trip the once when it is first turned on, and can be reset with no problem and will not trip again no matter how many times you turn them off and on by the switch unless they are left off for a little while is telling me something, but don't know what. I timed it, it takes the heaters being off for about 10min for it to trip when turned back on.

 

[HandySparks]

I don't possibly see how you could get that kind of inrush current on a heater circuit which is to all intents and purposes a linear load.

Given that these patio heaters are generally a radiant type and can glow strongly (because the element gets so hot), I'd expect their characteristics to be approaching that of a filament lamp.

 

[Marvo]

He says there's an identical circuit on the other side of the restaurant that has identical heatersand isn't giving problems. I guess it's possible that in-rush is the problem but it's not normally a consideration for heaters to allow for start or in-rush current. It's also difficult to measure the current over the first couple of half cycles on start up unless you've got a scope or a tester with an extremely high sample rate and peak hold facility. I'd try swapping the 2 heaters on the tripping circuit for the 2 heaters on the well behaved circuit, maybe one of the heater tubes is rogue and this would indicate if it were. Otherwise I guess if the circuit is passing all bthe usual tests and the PEFC is high enough you could try a D curve MCB.

 

[Pete999]

Have you carried out any electrical tests on the circuit??

Not sure of the regulation number, but recall that it is recommended that any appliance 2KW or over should have it's own dedicated supply, question have you tried just using one heater at a time, and see if the fault vanishes with only one heater? a stab in the dark maybe, but worth investigating.

 

[Kris_88]

From memory I believe that the Zs was around the 0.40s so a D-Type could be a possibility, but would have to double check. Sorry I may not have mentioned originally but the fault does indeed vanish when any one of the two heaters in disconnected as I tried this thinking it may be a faulty heater, but as it was fine when I tried each on its own it didnt lead me to a faulty heater... I also had an identicle heater fresh out the pack from the office which I used to replace place each of the heaters one by one, testing them as it went and the problem persisted both times.