IET
Decrease font size
Increase font size
Topic Title: Back E.M.F. on the neutral.
Topic Summary:
Created On: 27 February 2019 07:20 pm
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.
 27 February 2019 07:20 pm
User is offline View Users Profile Print this message



Zoomup

Posts: 6117
Joined: 20 February 2014

From elsewhere. I found this to be very interesting.

This was an interesting problem and solution
I recently wired an community hall - the supply was a TPNE - one phase was being used for E7 heating. One phase was unused. And the third was to supply the lighting and power. I fitted a split board 12 way CU and ran two A2 radials for the sockets. The hall was fitted with around 30 wall mounted fluorescent lighting fittings via an 8 gang switchplate. I very wisely fitted a "lighting control" connection box (where all lighting cables terminated before going to the relevant switches and fittings) All circuits tested ok. But the RCD would trip occasionally when a light switch operated . The RCD would not always trip with the same switch . It could be any of the 8 - retested circuit - all ok, changed RCD - problem persisted . I took the lighting circuit off RCD protection (split board) . But the problem persisted. Fitted a separate CU for lights only with no RCD protection . Problem not solved. The RCD would still trip when light switches used (not every time , but now and again) I decided to put the lighting CU on its own phase - so utilised the unused phase. The problem still persisted. I eventually realised what the problem was ( Back emf on the neutral caused by inductive discharge lighting) . I solved the problem by fitted double pole light switches and switching both line and neutral conductors. First time in 40 yrs that I have come across this problem.

Z.
 27 February 2019 09:05 pm
User is offline View Users Profile Print this message



broadgage

Posts: 3170
Joined: 07 August 2007

Yes, the back EMF when the lights are turned off can result in leakage to earth that trips an RCD.

As you have found, a double pole switch can cure this, but not always. It depends on which pole opens first, and by how long, which is down to pot luck or manufacturing tolerances.
 27 February 2019 09:25 pm
User is offline View Users Profile Print this message



chrispearson

Posts: 1095
Joined: 15 February 2018

Hm, it's a long time ago, but I was taught that back EMF was to do with things like motors.
 27 February 2019 11:14 pm
User is offline View Users Profile Print this message



broadgage

Posts: 3170
Joined: 07 August 2007

And chokes in discharge lighting circuits.
 27 February 2019 11:47 pm
User is offline View Users Profile Print this message



mapj1

Posts: 12039
Joined: 22 July 2004

A change of switch is not a nice way to solve the problem however. Ideally a low pass filter would smooth the transient so that there was not a sudden voltage spike associated with disconnection. The same effect can sometimes fire PIR sensors, or glitch the TV picture.

The problem is that once current is flowing in an inductive series choke, if one attempts to open the supply suddenly all the magnetic energy has to go somewhere, and tries to keep the current flowing by increasing the voltage until something breaks over, usually the switch, so it arcs briefly with repeated short surges of current, rather than a clean break.
You may find a DP switch also helps even if both poles are in series in the line only - as in effect the gap is opening twice as fast.
A network of series R and C can be quite efficacious at rounding the edges off the spike, ideally across the choke itself, but in practice across the switch is almost as good. Or a gas discharge device of a rather more than mains, perhaps 600V breakdown, across the choke

-------------------------
regards Mike
 28 February 2019 08:30 am
User is offline View Users Profile Print this message



Zoomup

Posts: 6117
Joined: 20 February 2014

You may find a DP switch also helps even if both poles are in series in the line only - as in effect the gap is opening twice as fast.

A network of series R and C can be quite efficacious at rounding the edges off the spike, ideally across the choke itself, but in practice across the switch is almost as good. Or a gas discharge device of a rather more than mains, perhaps 600V breakdown, across the choke


Thanks Mike, those are gold nuggets of information to be stored away for future use.

Z.
 28 February 2019 01:13 pm
User is offline View Users Profile Print this message



chrispearson

Posts: 1095
Joined: 15 February 2018

Originally posted by: broadgage

And chokes in discharge lighting circuits.


I have to say that in this context "back EMF on the neutral" is a bit of a misnomer.

IIRC, it occurs in a motor because when it is spinning, it also acts as a generator. It is greatest when the motor is spinning fast (off load) and acts to reduce the current and the torque. If a motor is put on load and slows down, the back EMF is reduced, the current rises and the torque increases. At start up there is no back EMF and the current is highest.

I can see that equally well, a choke would produce a back EMF which would be absent on switching on (hence the greater starting current) and collapse on switching off.

However, what is happening here? The line supply to a lamp is disconnected and momentarily, the "back EMF" is still present in the N conductor so it travels back to the origin where it momentarily reduces the current in the N conductor and thus produces an imbalance in RCDs elsewhere.

What I am struggling with here is current flow in an open circuit.
 28 February 2019 02:01 pm
User is offline View Users Profile Print this message



mapj1

Posts: 12039
Joined: 22 July 2004

The thing is the circuit is not properly open, there is a short duration of 'snap crackle and pop' at the switch contacts, during which a rather nasty interrupted current flows though the cloud of ionised gas betwen the moving contacts that is being stretched, and the action of the coil is to aggravate this, by allowing the voltage to rise well above the normal (a kV or 2 pk to peak is quite possible with ringing occuring for some tens to hundreds of microseconds in a bad case) These very high rates of change of voltage (high dv/dt) cause induced currents in a radio like wayt that are troubles for some designs of RCD, and often other elecroncis too.
(a very primitive EMP weapon of the 1960s vintage to attack computers and so on might be a spark gap deliberatly excited to do just this.)

-------------------------
regards Mike
 28 February 2019 04:12 pm
User is offline View Users Profile Print this message



chrispearson

Posts: 1095
Joined: 15 February 2018

Originally posted by: mapj1

... there is a short duration of 'snap crackle and pop' at the switch contacts ...

So we do need AFDDs after all!

These very high rates of change of voltage (high dv/dt) cause induced currents in a radio like wayt that are troubles for some designs of RCD, and often other elecroncis too.

(a very primitive EMP weapon of the 1960s vintage to attack computers and so on might be a spark gap deliberatly excited to do just this.)


Thank you Mike. So is this "radio wave" travelling along the N conductors in much the same way as a signal might travel from a transmitter to an antenna?
 28 February 2019 09:42 pm
User is offline View Users Profile Print this message



mapj1

Posts: 12039
Joined: 22 July 2004

adding an AFDD to a circuit that trips the RCD because of this is perhaps also going to trip the AFDD as well.
It is not really 'on the neutral' the current when it flows is from L-N in the normal way, but fast moving charges on one wire cause other charges on other wires nearby to move in sympathy (that is also the miracle of Radio in essence, just the magnitudes are such that one is normally transmitting watts, and recieveing fractions of nanowatts.) In any case, some of the current gets induced in the wrong place, and as far as the RCD is concerned it looks like an imbalance.

It should be possible to make an RCD that ignores things that are very fast compared to 50Hz, but some designs don't.

-------------------------
regards Mike
 28 February 2019 10:15 pm
User is offline View Users Profile Print this message



wallywombat

Posts: 96
Joined: 19 October 2018

Is this a good use case for a type 3 SPD downstream of the RCD, - i.e. to supress switching transients?
 28 February 2019 11:35 pm
User is offline View Users Profile Print this message



mapj1

Posts: 12039
Joined: 22 July 2004

possibly - but it needs to be on the load side of the light switch, or accross the choke.

Given the enormous expense of an SPD that fits in a DIN box, I'd buy a bare one, and put it in any old metal box.

an example

that is £2.30 and has a monitor lead for a little neon to show the internal fuse has not blow.

-------------------------
regards Mike
Statistics

New here?

  • To participate in discussions, please log in and introduce yourself.

See Also:



FuseTalk Standard Edition v3.2 - © 1999-2021 FuseTalk Inc. All rights reserved.

 
..