I feel that you have a duty of care to get a professional in to locate the problem. That is unless you are a competent electrician and appropriately insuredfor the work you will be carrying out.
Regards
Steve Dawson
In article , Stephen Dawson writes
Does it trip when the "extra" switch is turned on, or just with 75% running? Fans and extractors can have quite a switch on surge, as can fluorescent and halogens. I would guess that it is the surge which is the problem and Steve is right - you need an expert.
Yeah, I was being generous, there's a lot of detail we don't know.
FWIW, I thought electronic halogen transformers were pretty good on PF, certainly the ones I played with a while back just rectified the mains and chopped the power across a small toroidal ferrite transformer at HF. No storage capacitor to bugger up the input waveform.
CFL's are crap, due to a rectifier-cap input circuit. ISTR a 20W Philips one was PF=.515, 11W was 0.575 when I measured some (Voltech power analyser).
Proper electronic ballasts for linear FL's should have active power factor correction, and be pretty close to unity.
Anyway, we should have added that lot up vector wise, accounted for the low PFs due to distortion rather than phase shift, etc, etc. But for a "back of the envelope" calculation, we've shown that there may be an overload problem that the original sparks toting up the bare watts figures didn't account for.
It's difficult to find specs of 2D lamps, but I suspect the tube voltage will be around 50V, which would make the uncorrected PF more like 0.25, which is 810 VA.
Again, tube voltage is something like 110V, which would make uncorrected PF less than 0.5, so VA could be more like 150.
One possibility could be progressive failure of PF correction capacitors (if the fittings ever had them), resulting in a progressive increase in the circuit current. It's not uncommon to find PF correction capacitors going open circuit in strings of fluorescent luminares as they age (or more strictly, they short out and an internal fuse opens).
Yep, from the sound of the problem it doesn't look like its a simple one. I suspect the electricians that did all the work a few years back weren't "competent".
It'll just have to wait until people are back at work next week.
What am I missing:
AFAICS there is no reason to believe anything is unsafely amiss, beyond switch on surges popping the breaker.
OTOH I forget what the level of expertise of the OP is.
NT
While I can happily accept the numbers, I'm not so sure I'd interpret them the same way. 7.6A of loads on a 5A circuit would not normally be a problem, due to diversity. The problem I think is less liekly the continuous ratings, more likely the surges, and the poor surge tolerance of type Bs.
NT
There's no diversity allowed within a lighting circuit. You need to expect it can handle all lights on at the same time.
What surges are you talking about? Type B's (and type C's) can trip if a mains filament lamp flashes over when the filament breaks, but there's been no suggestion this is the problem here.
Depends on if they are all switched on, which is a possibility (hence why you don't use diversity for loading calce on lighting circuits).
Note also however that for a 6A type B you need >8.5A to get onto the start of the thermal trip part of the curve.
Where is your youth hostel??
I'm not sure it's quite that bad. I have a Thorn Lighting data book (early 90s vintage) which gives the following data for the 16 W 2D:
tube volts 97 V running current 0.2 A circuit watts 21 W total VA 48 VA circuit power factor 0.44 (uncorrected, lagging) 3rd harmonic current 17%
This brings into focus the fact that the sums should be based on circuit watts (not tube watts) as the ballast losses are not negligible, especially for these small lamps. 5 watts ballast loss for a 16 W lamp is not exactly high efficiency.
Anyway, 13 luminaires at 48 VA each is 624 VA total, or 2.6 amps.
Same Thorn book says 103 volts (tube), 0.43 A, 47 circuit watts. If uncorrected that would be ~100 VA each, 200 VA total (PF = 0.455).
The 13 2Ds are very unlikely to be corrected. The default fittings aren't, you have to select the HPF special order option (high power factor, not highpass filter!) to get a capacitor.
Certainly a possibility, although I think that's less common with modern polypropylene capacitors, compared to the old impregnated paper things.
I can't see any problem with the OP at least diagnosing and locating the problem, even if he leaves fixing it to a "pro". So long as he is confident that he will not cause harm to himself (or others) in the process, there is plenty of advice available here from suitably skilled people.
IIRC its not just lighting on this circuit.
filament bulbs, CFLs and motors all eat current surges when switched on, for a different reason in each case.
NT
Indeed. A 6A type C should solve the problem. The mcb would be rated in accordance with todays regs, the cable is all >7.6A rated... yes its a close one but not a danger in any way, and should solve the problem. But no way is there any room on that circuit for anything else to be added!
NT
I would personally only be happy with that solution once I had convinced myself that there was nothing on the circuit that was actually causing the problem as a result of higher than normal power dissipation.
Especially since this is a circuit that used to run ok and has not been altered recently.
It's still a lighting circuit.
The problem is not at switchon, but up to 30 minutes later. In any case, the surges you are talking of are totally insignificant with regard to the fault current trip element on even a type B breaker.
Assuming that you check the final earth loop impedence value is in accordance with the manufacturers value. It is always nice to know that it wil attempt to trip in the event of a fault.Zs value for type B BSEN60898 is
8 ohms, for a Type C is it 4 ohm ( according to BS Devices selection from Amtech Single Cable). This may be an issue if the supply is TT or the Ze is high.Steve Dawson
HomeOwnersHub website is not affiliated with any of the manufacturers or service providers discussed here. All logos and trade names are the property of their respective owners.