IET
Decrease font size
Increase font size
Topic Title: loss of neutral on 3 phase system
Topic Summary: Damaged accessories
Created On: 24 February 2019 10:02 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.
 24 February 2019 10:02 am
User is offline View Users Profile Print this message



tillie

Posts: 937
Joined: 03 May 2006

Hi , I was called out on Friday to a kitchen where an Rcd socket had failed with a rather large bang even though nothing was plugged into it.

The dishwasher had stopped working and some of the lights had gone out.

When I got there hours later I was not too sure what was going on.

Tested voltage at 3 phase distribution board and found the following

L1 to N = 365v

L2 to N = 55V

L3 to N = 65V

All correct voltages across lines and lines to earth.

Traced fault out to 3 phase switchfuse located in the main switchroom where the neutral had burnt away.

Repaired neutral and switched back on.

Total damage was 4x rcd sockets , two of which failed with large bangs.

Also the kitchen has 8x 5ft twin 58w fluorescents with electronic ballasts and 6x cfl emergency bulkheads of which 4 had failed.

I have come across a few burnt out Neutals in my time so I am familiar with what happens to the voltage across lines etc.

What I cannot understand is that the lighting is all on one circuit and I found the PCBs in 4 of the emergency lights fried and 2 of the electronic ballasts in the fluorescents fittings had fried.

Why didnt the high voltage damage all the 8x fittings that are on the same circuit and also all of the 6x emergency bulkheads which are on the same circuit.

They all would have seen the 365v.

The high voltge destroyed the electronics inside the 4x rcd sockets.

All the 3 phase equipment survived ie Rational cookers , dishwasher and extract fans.

Is it just pure luck that most of the lights survived or something I am not understanding.

All the lights are the same make and model .

Regards
 24 February 2019 11:03 am
User is offline View Users Profile Print this message



mapj1

Posts: 12039
Joined: 22 July 2004

A large element of luck- although the electronics will be designed to withstand 230V RMS (plus at least 10%) all day, the amount of actual margin before failure will be down to manufacturing tolerances. Quite possibly the ones that failed first became a low impedance once faulted and either operated an MCB, or just pulled enough extra current to drag down the voltage and redress the balance slightly, and reduce the stress on the other units.
I'd not be too surprised if the apparently undamaged ones have been 'lifed' to an extent, and actually you get called back to change them in a few weeks or months.

Conventional transformers of the iron and copper kind are normally close enough to the wire in their designs that the over-voltage will saturate the core, and the transformer primary draws too much current, but the secondary voltage is limited, and a rather odd wave-shape. If things are correctly sized either a fuse or thermal cut-out operates.
Modern switching supplies actually rectify the mains and store charge on an electrolytic capacitor at about the peak of the mains voltage, so normally about 330V or so - with 400 V in place of 230 this becomes nearer 550V. The electrolytic cap may just electrolyse a bit but then grow a thicker oxide, and hold off the extra for a few hours, or it may sizzle and fill with hydrogen/ oxygen mix from the electrolysis and keep passing a high current - soon after that pop goes the weasel, and hot conductive goo exudes over the rest of the circuit. Bulging capacitor bodies are a good early warning to turn off and take cover while it cools !
Or the rectifier diodes may fail, either due to over voltage, or due to over current if the capacitor does go for the sizzling electrolysis of doom with a vengeance.

Again, if things are well designed, either an on-pcb fuse or fusible resistor, or thermal link should open quickly and quietly, but in the more "value engineered" designs there may be no such features, and the first you know about it is a loud bang.

-------------------------
regards Mike
 24 February 2019 11:18 am
User is offline View Users Profile Print this message



tillie

Posts: 937
Joined: 03 May 2006

Mike , thank you very much , absolutely brilliant explanation.

I am going to recommend to change all the ballasts in the 8x fluorescents.

I changed the damaged 2 yesterday and pick up 4x emergecies tomorrow.

I shall also recommend changing the working emergencies.

The rcd sockets I replaced were totally blackened on the reverse side of the accessory , they were the type with the red neon around the reset button.

Regards

Funny thing is when reading up on lost Neutrals on Google I came across two seperate posts I done a few years ago about this very subject that my ageing memory has completely forgotten about.

Regards
 24 February 2019 02:59 pm
User is offline View Users Profile Print this message



alancapon

Posts: 7492
Joined: 27 December 2005

Although you quote the voltage, what you have here is the final steady state values, once everything that was going to go bang, has. Whilst things were failing, it is likely that the voltages were swinging round quite violently due to the failing items drawing significantly more current than they should. The effect of this is to raise the voltage on the phases that that particular item is not connected to.

This is one place where SPDs can help, but they do need to be across all three phases and rated correctly, rather than some of the "toy" ones that are designed to fit in a CU. Hopefully, one or more of the supply fuses will operate before the SPD itself self-destructs.

Regards,

Alan.
 24 February 2019 05:00 pm
User is offline View Users Profile Print this message



wallywombat

Posts: 96
Joined: 19 October 2018

I doubt that SPDs would have helped. They're designed to handle transients measured in kV and lasting micro-secconds. An over-voltage due to loss of N may well not be high enough to trigger the SPD; and even if it did, the prolonged short would probably trigger the SPD's OCPD before the supply's main fuses failed.
 24 February 2019 05:32 pm
User is offline View Users Profile Print this message



alancapon

Posts: 7492
Joined: 27 December 2005

Yes if you are talking about SPDs at the origin. If you have SPDs nearer to the load as well, these are likely to clamp the voltage closer to what most "well constructed" kit can take. I do know of an SPD extension lead that saved a broadband router and telephone plugged into it in a "broken neutral" scenario. The SPD operating tripped the single front-end 30mA RCD, and probably saved a lot more equipment inside the property. I don't recall whether the RCD survived the experience, but the SPD extension lead did not.

Regards,

Alan.
 19 March 2019 08:21 am
User is offline View Users Profile Print this message



Nedryerson

Posts: 142
Joined: 12 December 2009

Ah loss of Neutral!

A year or two back I was called to a local Hospice who were experiencing problems when using their standby generator (For 'problems' read wall lights jumping off the walls etc).

They had employed a consultant who came from 'London' no less who declared the neutrals weren't connected up properly (in a nice neat row) and I they told me the sparks was fired also.

Removing the cover from the generator panel revealed that the neutral had never been connected or earthed so the phase voltages would have been all over the shop. A piece of wire sorted the problem.

I also had crazy voltages (Circa 350V) in my house when someone pinched the neutral conductor from the overhead line. After the DNO had replaced the neutral a chap turned up and asked me sign a form certifying none of my connected equipment had been damaged by over-voltage. I declined to sign.

Finally, can anyone remember the formula for working out the phase to phase voltages on a three phase system when the neutral becomes disconnected. I vaguely recall you need to know the load.

Ned
 19 March 2019 09:32 am
User is offline View Users Profile Print this message



Jaymack

Posts: 5571
Joined: 07 April 2004

Originally posted by: Nedryerson
Finally, can anyone remember the formula for working out the phase to phase voltages on a three phase system when the neutral becomes disconnected. I vaguely recall you need to know the load.

Search for Millman's theorem. One year, our lady instructor set the final paper in theory. The week before this national exam, she gave us a problem in an unbalanced 3 phase load with no neutral, it came up in the exam and was worth 20 marks.

Regards
 19 March 2019 12:09 pm
User is offline View Users Profile Print this message



mapj1

Posts: 12039
Joined: 22 July 2004

here is an example

you need the 3 voltages and the three load resistances.
If loads are reactive or negative resistance like SPMS can be then I tend to use LT spice these days.
If you ask me on the new forum I can probably attach a typical simulation run to a reply tonight, but you'll need to have LT spice on your machine to display the waveforms.

-------------------------
regards Mike
 19 March 2019 12:29 pm
User is offline View Users Profile Print this message



Jaymack

Posts: 5571
Joined: 07 April 2004

Originally posted by: mapj1
here is an example
you need the 3 voltages and the three load resistances.

That's the simpler method with resistances only. It's more difficult when complex numbers are involved. As I remember it took well over an A4 sheet and using log tables with a slide rule!

Regards
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.

 
..