IET logo
 
IET
Decrease font size
Increase font size
Topic Title: Unexpected dimming of lights.
Topic Summary: Why is it happening when the load is removed?
Created On: 22 September 2017 09:43 AM
Status: Read Only
Linear : Threading : Single : Branch
Search Topic Search Topic
Topic Tools Topic Tools
View similar topics View similar topics
View topic in raw text format. Print this topic.
 22 September 2017 09:43 AM
User is offline View Users Profile Print this message



potential

Posts: 1642
Joined: 01 February 2007

Yesterday I encountered an electrical phenomenon that I have not experienced before.
I wonder if anyone can explain how and why it is happening?

I'll start at the beginning.
I was working in my shed at the bottom of the garden.
At the same time my wife was clearing up the garden using an electric leaf blower.
The shed is fed by a 3-core 4 mm cable approximately 30 metres long supplied from a separate MCB + RCD supply taken off the house single phase meter.
5x standard 5ft fluorescents and 2x 80-watt metal halide lamps all with conventional ballasts light the shed.
Initially my wife used an outside socket on the shed.
I noticed she was switching the blower on and off quite quickly and all the lamps in the shed were dimming momentarily every time she switched it on.
Nothing odd in that I thought but made a mental note to tell my wife it wasn't a good idea to cycle the motor so rapidly.

Then she moved further towards the house and unplugged it from the shed socket to an outside socket supplied from one of the ring finals in the house.
That ring is connected to a RCBO in a different CU but supplied from the same meter.
(The house supply cable is underground and is supplied from a small substation just down the road.)

She continued to switch the blower on and off.
This time the lights in the shed hardly dimmed at all when she switched it on, as I'd expect.

However each time she switched the blower motor off all the lamps in my shed dimmed quite markedly approximately half a second afterwards.

What would cause that to happen?
Any ideas?
 22 September 2017 12:14 PM
User is offline View Users Profile Print this message



kellyselectric

Posts: 192
Joined: 22 July 2016

Ive seen a similar sort of effect when switching on a large 2 Kilowatt heater in my shed yes you expect dimming at the moment of switch ON but ive seen them dim a just briefly at switch OFF also have seen same effect in a house caused by a kettle I'm interested to know what mite cause such an effect. The supply to my shed is via a length of 1.5 mil flex from an outdoor 13 amp socket. The cable length is around 60 feet
 22 September 2017 01:00 PM
User is offline View Users Profile Print this message



nad

Posts: 399
Joined: 14 January 2005

Hello Potential and Kelly,

My first guess was the inductive reactance given back to the supply when the motor in the leaf blower is turned off causing the volt drop. But that wouldn't explain the purely resistive kettle. Maybe a high resistance ark is formed when the switch breaks the circuit causing the voltage to drop, dimming the lights?

Hopefully someone will explain it.

-------------------------
Nad

*Regularly edited due to spell cheque misdiagnosis
 23 September 2017 12:15 AM
User is offline View Users Profile Print this message



mapj1

Posts: 9717
Joined: 22 July 2004

This is the sort of thing best tracked down with an oscilloscope, or a low voltage transformer into the audio jack of a PC as a poor mains waveform monitor.
Firstly be aware that the eye is not good at distinguishing dimming and brightening in rapid succession, from the reverse, so it may not be quite as you imagine.
There are a couple of possibles. Iff the switch is bouncing, then as the motor spins down after connection it will act as a generator, but one that may be either in or out of phase with the mains when reconnected briefly, depending on the switch bounce period. This could either boost or reduce the voltage for a part cycle.
I have also had a an arcing double pole switch that placed a partial short across the mains as the L-N contact arcs briefly met, though it must be said, I have only ever seen that once, and it had a short and brutal life.
Then there are L-C transient effects, which become significant if the cable supply route is unusually inductive, this will be a function of substation rating and distance, type of electricity meter you have, and sometimes really bad because L and N travel different routes (an extreme case is a ring with a broken core, such that either neutral or live goes round the long way, while the other goes short .) In any case when a load goes off, which may be resistive, the change in current may lead to a change in voltage that overshoots. However this us usually in the tens of micro henry region, so only noticeable (i.e a few volts) with large loads and very fast switching - rate of change amps per micro-second.

-------------------------
regards Mike
 23 September 2017 10:16 AM
User is offline View Users Profile Print this message



potential

Posts: 1642
Joined: 01 February 2007

Thanks for the replies.
I've experimented a little more with the delayed dimming after switch off.
With the blower running, if the plug is pulled out of the external socket on the house the delayed dimming still occurs, so it won't be switch bounce or arc-over.
The delay period is quite substantial, not just a couple of cycles of the mains..... about the same length of time it takes to notice the dimming and then say a single syllable word.
I reckon it is about half a second.... and that is a lot of cycles.
The dimming is quite long too, about the same length of time.

It is as if there is something mechanical somewhere deliberately causing the drop in voltage.

Also remember that the dimming after switch off did not occur when the leaf blower was run from the socket on the shed.
 23 September 2017 12:32 PM
User is offline View Users Profile Print this message



mapj1

Posts: 9717
Joined: 22 July 2004

Odd.
Well assuming you don't have some sort of tap changing voltage optimiser in the meter cupboard, or its modern equivalent you have forgotten about, if yourealy want to know, then it is probably time to organise some sort of logging, either a scope with storage, or as I alluded earlier I have mis-used a computer sound card as to capture transients in conjunction with an old fashioned bell transformer to isolate and reduce the voltage.
I presume you or your same phase neighbours don't have an local generation, such as solar panels or wind, that might be fooled into pushing harder when seeing a lower impedance load, and then overrunning for a short time after its removal, and that the DNO have not got some clever voltage lift compensator at the substation. The fact it does not happen with an extra 30m of 4mm is odd - I make that only adding the odd tenth of ohm or so. Do lights at the house end on both CUs do this ? - if so the cause may be external. Also how much wire is common between supplies to one CU to the next - and I'm assuming its not split phase so one goes up as the other down.

-------------------------
regards Mike
 24 September 2017 02:12 PM
User is offline View Users Profile Print this message



OlympusMons

Posts: 71
Joined: 05 February 2016

The shed might be TTd, and not connected to the house earthing, so it might have something to do with earthing.
I am thinking the main house is TNC-S with a broken PEN, so the neutral current is flowing down the bonding, heating up the earthing conductor and raising the voltage on it. When the blower is switched off, the cable cools, the voltage changes and it is this which causes the lights to dim. I am probably completely wrong though.
What does the data plate on the blower say?
 24 September 2017 05:41 PM
User is offline View Users Profile Print this message



potential

Posts: 1642
Joined: 01 February 2007

The blower is marked 230 volts, 50 hz, 2.2 kW.
The house earth is exported to the shed via the 3rd core.
The bonding in the house is 10mm and very short, under a metre from the MET to the lead water pipe and gas meter supplied via a steel gas pipe.
The supply cable is very old and relatively small.
There is no connection within the house between the MET and the incoming neutral.
The MET has a 10mm wire running to the supply cable steel sheath.

Using a clamp meter there was less than 6 mA flowing between the sheath and MET.
The load in the house was small (as was the load when the phenomenon occurred) being only 1.65 amps.


I did a few more tests but my wife is out this afternoon so wasn't able to repeat the original scenario.
I did not see any dimming of lights in the house except on switch on. There was no dimming after switch off.
I couldn't check in the shed.
I did run an extension lead from the house socket to the shed but the lights did not dim when the blower was switched off.
(This was as I expected)

I checked what voltage exists between the house neutral and a true earth (screwdriver stuck in the lawn).
It varied as I'd expect, but at the time the voltage was less than 2 volts.
When I applied a 4 kW load via another circuit running directly to the CU this voltage increased to 3.6 volts.
I am aware that the supply voltage to the house varies a great deal, anything between 244 to 207 volts so I will monitor this in relation to the supply voltage in the next few days.
 24 September 2017 05:44 PM
User is offline View Users Profile Print this message



potential

Posts: 1642
Joined: 01 February 2007

I forgot to mention that it was not easy to see any dimming in the house because of light coming in the windows, unlike in the shed which has no windows.
 24 September 2017 06:11 PM
User is offline View Users Profile Print this message



alancapon

Posts: 6943
Joined: 27 December 2005

If you really do see 207V, then you need to contact the DNO. The statutory minimum in the U.K. Is 216.2V.

Regards,

Alan.
 24 September 2017 07:52 PM
User is offline View Users Profile Print this message



kellyselectric

Posts: 192
Joined: 22 July 2016

I thought ide.just add that I noticed this dimming just after switch off with our electric shower which is 8.5 kilowatts or 35 amps load I only noticed it a couple of times and just assumed it to be a weird coincidence but when I shower at nite next time il certainly look more closely. The shower is fed from the main consumer unit via a 40 amp RCD we dont havePME (thank goodness)

Edited: 03 October 2017 at 12:33 AM by kellyselectric
 24 September 2017 11:10 PM
User is offline View Users Profile Print this message



MWalker86

Posts: 99
Joined: 05 June 2017

Originally posted by: OlympusMons

The shed might be TTd, and not connected to the house earthing, so it might have something to do with earthing.

I am thinking the main house is TNC-S with a broken PEN, so the neutral current is flowing down the bonding, heating up the earthing conductor and raising the voltage on it. When the blower is switched off, the cable cools, the voltage changes and it is this which causes the lights to dim. I am probably completely wrong though.

What does the data plate on the blower say?


This would be more plausible if it wasn't such an instant and brief effect (as described)
 26 September 2017 12:48 PM
User is offline View Users Profile Print this message



kellyselectric

Posts: 192
Joined: 22 July 2016

BUMP!
 05 October 2017 11:12 AM
User is offline View Users Profile Print this message



kellyselectric

Posts: 192
Joined: 22 July 2016

So guys am I right in thinking then that the notable dip in lights when something like a 2Kw heater is switched off or when its own thermostat switches it off is simply due to the inductance of the supply lead creating a momentary voltage that pushes against the supply voltage and causes a brief dip? I have noticed this effect is more pronounced with supplys that are a bit weak for instance in a factory I worked in wehad several large ovens and they never made such an effect because we had a big TX just outside in the yard
 05 October 2017 01:57 PM
User is offline View Users Profile Print this message



potential

Posts: 1642
Joined: 01 February 2007

I think it may have something to do with the supply being "a bit weak" but I cannot work out in my head how inductance (or whatever) can produce such a long dip in the voltage.

A little dip, yes but the dip in voltage I'm talking about looks like a major short then cleared somewhere else on the distribution system.

So far I haven't seen any satisfactory explanation for it.
 05 October 2017 02:01 PM
User is offline View Users Profile Print this message



potential

Posts: 1642
Joined: 01 February 2007

Also it is a drop in voltage but not a loss of voltage which would cause my metal halide lamps to extinguish.
They remain lit.
 05 October 2017 04:41 PM
User is offline View Users Profile Print this message



kellyselectric

Posts: 192
Joined: 22 July 2016

I wonder if at the moment of switch off the high voltage spike created causes other equipment and the thing that caused it to draw more current for a short while maybe for 25 of 50 cycles? I know this cant explain the dip which occurs AFTER switching that one is weird I really wish I could find an explanation for that one
Statistics

New here?


See Also:



FuseTalk Standard Edition v3.2 - © 1999-2017 FuseTalk Inc. All rights reserved.

..