Not known this before. Friend's cassette deck is chewing tapes, so I said I'd take a look. Plugged it in and flash/bang from the plug, blew the 13A fuse in the extension, and tripped the socket circuit on the main board. Anyone know the cause?:
Given the ablation on the earth and live terminals, I'd suggest something bridged between them, that screw wasn't rattling around loose in there, I presume?
I would be tempted to test the fuse.
A LE short at that point may not have taken out the fuse.
ARW snipped-for-privacy@blueyonder.co.uk> wrote in news:Nd4HH.155906$GUR9.144313 @fx16.ams4:
/Or had the screw been missing and it had been arcing for ages?
Looks to me like it was tracking between L and E *before* the fuse.
My money is on the fuse in the plug itself still being OK - apart from the layer of carbon on its surface. It is scrap now.
I reckon the plug has been soaking wet at some time in the past and was just damp enough to allow a carbonising arc path to form. The damage is consistent with a fault inside the plug and before the fuse inside. Brass doesn't melt all that easily so there was a high current flowing.
Last time I saw one similar a CH leak was dripping water onto it. It went with a hell of a bang when switched on (took two goes to find it).
It looks more like a short than a lot of arcing.
ie too much melted metal and soot as opposed to a overheat that would normally be caused by arcing.
The live screw was rattling about inside, half the thread stripped and charred (you can just see it in the pic). To my eye it looks forced out. But maybe it was cross-threaded and worked loose? Cable was securely clamped.
Fuse (13A) had blown.
FWIW, the cassette deck wasn't harmed. I've replaced the plug/3A fuse.
Thanks for the update and thanks for posting the photo. I found it interesting.
Agreed and how would it short to the earth terminal with the plug cover properly fitted? The earh terminal is enclosed by the mouldings in the cover.
And the metal is ablated on the earth terminal, metal that looks like it might be deposited on the plug base between the earth and live terminals. With burnt plastic (soot) being blasted against the top left side of the fuse. There is also a "shadow" in the soot mark below the lower right corner of live wire terminal.
Looks like a short between the live wire terminal and earth wire terminal. The live wire terminal is post fuse so that may be a goner as well. I'm a bit stuck for why or how that short occured. I don't think that water would be conductive enough or hang around long enough with the sort of power levels required to blast molten brass about the place...
Agree with the rest of posts, but would also note that the transverse "crack" in the insulation of the brown suggests that this may have been overheating for a while, and hence become embrittled. A slack screw (and long-term mild arcing within the live terminal) could have caused this.
One guess would be that the live wire was fitted to the terminal when removed from the plug, and then the terminal replaced and the fuse inserted.
If they had left some long strands on the wire end poking out the back of the life terminal, they would be folded up and sandwiched between the end of the live terminal and the plastic moulding.
Putting the top on the plug could have folded a strand towards the earth terminal. Then it would only take a small shift in the wire, with a knock or something, to create a L&E short via the strand. Once power was applied, it would blow the strand and start an arc, which then established between the edge of the earth terminal, and the screw in the live terminal. The vaporised metal and soot being the end result.
It's not, actually.
The missing screw might be a candidate but unless the plug cover is broken leaving a gap a loose strand of the brown wire is likelier.
Er, wall socket live > plug pin live > fuse > live wire terminal
Live wire terminal post (aka after) fuse...
Screw isn't missing it's in the photo(*) and was apparently rattling around inside, the screw shows signs of melting. The mouldings in the plug cover look undamaged, well apart from the results of the failure.
Don't think a single strand would survive long enough to do that much damage to the terminals. I suppose it could have formed an ionised path to sustain an arc between them.
(*) You don't get to see the full image by default, scroll around abit.
Yes, sorry you're right. Not thinking straight.
The damage to the head and threads of the loose screw make me wonder if that screw was involved in the L/E short.
Is there an RCD on that circuit? The damage suggests a high fault current, but I'd have expected something in the 30mA RCD + MCB + 13A Fuse to have limited the fault current & duration.
I agree 100%.
Yes. again 100%. That should NOT have happened on an RCD protected circuit. Worrying
There was actually two stages to this event. (I ams speaking as a doctoral level materials scientist here who has done failure analysis using scanning electron microscopes with EDS/WDS onboard and XRF)
An arc/plasma has formed between the earth terminal and the live wire's terminal. As suggested by othe, this may have been formed via a stray strand.
It requires less energy to sustain an arc than to create the arc in the first place.
This produces a vapourised & ionised metal arc plasma.
The clue is in that the earth pin has molten metal on the bottom right hand corner and bottom edge only and you can see the molten metal also on the top left hand corner of the live wire's terminal. Both regions are pointing to each other and the closest physically.
A layer of brass (copper & tin) metal has been deposited on the insulative area between the earth terminal and the live terminal.
This deposited brass will be electrically conductive......
So that live wire terminal actually becomes part of earth once that 13A fuse blows.
(In PVD, PLasma vapour deposition, a plasma is used to deposit metal onto a non-conducting surface.)
you then get a second flashover event from the now earthed live terminal to the top of the plug's live pin which is also the bottom cap on the now open circuit fuse
Then there was an explosion which then deposited carbon all the way down to the top of the plugs live pin and the bottom cap on the fuse and it is this event that popped the 32A breaker. I seem to recall that the real current achieved will be several times higher than that 32A rating (John Rumm and ARW can probably work this out from the Adiabatic equation! :-) along with the time it took for the 32A MCB to open and break the arc..... )
You can see the carbon deposited all around the plug cover's fuse "shell"
So what initially started as a molten copper-tin arc plasma between then became a carbon arc plasma. (don't forget, if you already have hot "air" with metal particles in it, its easier to sustain an arc than to initiate it !)
EDS/WDS in a SEM will confirm the surface region between the earth pin and the live wire pin to be coated in copper and tin, and that the region between the two fuse ends is carbon.
To the OP:
Perhaps you have that Wylex board where the rewirable carts were replaced with those plug in MCBs? In this case you have no RCD.....
P.S.
The distance between the two fuse caps is approx 4 x the distance between the bottom fuse cap and the "now earthed" live wire terminal.
So once the fuse wire had melted, it would have formed an arc within the fuse casing. There is ceramic powder within the fuse casing which act as an arc quencher.
It would have quickly transferred itself to the outside of the fuse as the ditance is 4x shorter..... (the energy required to initiate an arc reduces as the seperation distance decreases)
Yup, hence my guess at a stand of flex, and as you say the erosion of the metal on the edges of the terminals does suggest those as the initial flashover point.
Yup in fault current scenarios the magnetic part of the MCB trip mechanism will activate. That is at a nominal 5x In for a type B MCB. So a current of 160A or more will do it.
Difficult to know without also knowing the loop impedance at the socket. However since it was at the socket and not the appliance end of the flex, it will presumably be within the normal design parameters for a
32A ring circuit - so no more than 1.37 Ohms. Chances are (looking at the damage done) that it was a good deal lower than that. So you could have ~500A of fault current to play with. If the MCB opens in the nominal 0.1 sec "instant" part of the trip, that would suggest a let through energy I^2t or 250K x 0.1 = 25,000 J, which (to put it mildly) ought to get things toasty pretty sharpish! :-)In reality it would be cheaper to replace than test - since there is not much testing you can easily do other than see if it will operate again. Chances are that it will be fine - a few hundred amps is not going to stress a device with a 6000A breaking capacity.
However if its a "high energy" installation - i.e. close to a substation, or fed from a more industrial sized feed, and so with very low external Ze, then you could get close to or even exceed the breaking capacity of the MCB and damage it. (particularly so if its one of the "plug in" wylex MCBs designed for the old BS3036 rewireable fuse boards, since many of these have lower maximum breaking capacities.
Chances are a normal RCD may not be fast enough to trip in these kinds of circumstance. It will normally take at least one mains cycle (20ms) to detect the imbalance, and the MCB will have it all done and dusted long before that.
+1
and MCBs with less breaking capacity...
Well the fact that the plug is still in one piece and the house is not on fire, one might argue it all worked as it should.
The RCD is possibly a diversion since they have no ability to limit fault current - only fault duration.
In high current fault scenarios the fuse and MCB would operate before the RCD will get a look in generally, since RCDs need to see a fault for a good proportion of a mains cycle before tripping.
As others have suggested the carbon deposits on the fuse would suggest that ultimately there was a flashover between the earth and neutral
*pins* (i.e. not just the screw terminals[1]) of the plug - that basically puts the plug fuse out of the picture as well, once the arc is established. So now its all down to the circuit MCB. [1] even though the flash over between terminals started the ball rolling.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.