Strange Co-ax Fault

[securespark]

Hi guys.

Got a weird one today.

Customer explains that Virgin came to fit a new cable box. The only installing they did was to swap the box over. No drilling, etc. They offered to attach the old analogue aerial so, they said, he could watch some channels without using the Virgin box. Well, the Virgin guy attached the co-ax to the TV and there was an almighty flash & the lighting fuse went.

I checked the co-ax. It did not light up with a volt stick. Scratching my head, I went round turning lights on. Sure enough, with the lounge light on, the co ax. became live, showing 224V between the outer braid and the MET of the installation. My immediate thought was a cable short between co-ax and the switched feed to the lounge light, but this seemed too simplistic. The only co-ax near the lounge light wiring is to another independent set and that does not become live when the lounge light is powered up.

The other co-ax points in the house are all live. Apparently, they are jointed under the floor, but are not routed via an amp. That was my second thought, prompted by a colleague. However, if it was a faulty amp, why is it only live when the lounge light is on?

The only other possibly relevant point is that the cpc to the lights, although present, is showing >1999 ohms on an EFLI test.

I thought maybe there's an earth fault on the lounge light and the fault current has somehow found its way onto the co-ax braid.

The fuse remains intact with the lounge light on (and therefore the co-ax at 224V referenced to earth), but if that co-ax is connected to a co-ax socket on a TV (I presume these are grounded?), then the fuse blows.

Any bright ideas?

As another colleague said, "You do get 'em, Simon...!"

Thanks in advance...

 

[electronicsuk]

My immediate thought was a cable short between co-ax and the switched feed to the lounge light, but this seemed too simplistic. The only co-ax near the lounge light wiring is to another independent set and that does not become live when the lounge light is powered up.

Sometimes the simplest explanations can be correct. Have you traced all the wiring back to be 100% certain it doesn't run anywhere near the coax cables, or is that just an assumption based on the position of the lights/TV/switch points?

The other co-ax points in the house are all live. Apparently, they are jointed under the floor, but are not routed via an amp. That was my second thought, prompted by a colleague. However, if it was a faulty amp, why is it only live when the lounge light is on?

If there is an amp, the supply could easily have been taken from the lighting circuit. The majority of amps I've seen are double insulated, but I did once have one that required an earth connection. So, amp fed from lighting circuit, circuit has L-E fault, but as you state below, the circuit fails EFLI test, so fuse would never blow and anything connected to the amp would become live.

The only other possibly relevant point is that the cpc to the lights, although present, is showing >1999 ohms on an EFLI test.

I thought maybe there's an earth fault on the lounge light and the fault current has somehow found its way onto the co-ax braid.

Yup, that sounds sensible. Do any other metal fittings related to the lighting circuit become live when the lounge light is on?

Another remote possibility would be an earth fault in a backbox shared with a TV outlet. For example, dual backbox with TV outlet and table lamp socket, L to backbox earth fault, and no earth continuity on lighting circuit so invisible to circuit overload protective device.

 

[RF Lighting]

Did you take a Ze and Zdb reading? Were they ok?

Just seen table lamp socket mentioned above. Might be a good call!

 

[DCC]

If there is an amp, the supply could easily have been taken from the lighting circuit. The majority of amps I've seen are double insulated, but I did once have one that required an earth connection.

So, amp fed from lighting circuit, circuit has L-E fault, but as you state below, the circuit fails EFLI test, so fuse would never blow and anything connected to the amp would become live.

Sounds just strange enough to be the answer, sure there is no buried amp hanging around?

 

[securespark]

Thanks for the responses so far.

I have not yet found an amp. I had a limited time there today (around 60 mins), so did not get a whole lot done. I cannot see any co-ax anywhere near the lounge lighting wiring, save the co-ax gpoing to an independently-fed (aerial wise at least) set in the litchen, which is not live.

No Zdb's to worry about, Rob. Ze OK, Zs at tv socket 0.72. Zs at lounge light >1999, though.

All co-ax outlets are either the old surface-mounted style, or just co-ax cable with a co-ax plug on the end. No box-mounted outlets as such.

Have not had time to trace stuff seriously yet. Will be back there tomorrow AM for a good root.

Have not yet found any fittings live when lounge light is on.

 

[Spark123]

You'll probably find some dipstick has wired the coax to the lighting circuit.

 

[Steve]

 

[securespark]

Strange thing is, the customer said the aerial worked sweet as a nut on his old TV. He now has a flatscreen. Ooops, forgot to mention that.

 

[Steve]

flatscreens sometimes / usually? have an IEC lead - the chassis is earthed. CRTs have no earth. Clue there perhaps.

 

[ColJack]

flat screen?... you mean a plasma or an LCD?
most CRT's over at least the last couple of years have been flat screen tubes.. gone are the days of curved glass on the telly..

anyway.. a plasma or LCD sometimes means a wall mount... has he gone through the cables in the wall?

 

[flameport]

I thought maybe there's an earth fault on the lounge light and the fault current has somehow found its way onto the co-ax braid.

Some idiot in the past has installed a class I light fitting, but when doing so, could not find the earth (cpc).
Or worse, they tested the circuit, found the cpc was defective and looked for an alternative.
They located the coax nearby and 'borrowed' the outer braid as the earth.

Subsequently, a fault occurs on the light fitting, causing the coax to be live.

 

[securespark]

Yeah, that's one of my photies! Forgotten about that one...

Steve, yes, it does have an IEC lead and you're right, there's every chance the old set was double insulated on a two core flex.

 

[aptsys]

A lot of the IEC leads used on LCD tv's are actually 2 pin versions, so it still may not be earthed.

 

[Steve]

and thinking on, my Toshiba LCD set came with a fixed 2 core flex.

 

[Adam_151]

So whats the answer Simon?