Using armour of SWA as CPC ;)

[disinfo]

This is the incoming supply to a power press that I've been called to a fault on this morning.

I've had time to write this post because i'm sat in the canteen until the maintenance electrician sorts it. I'll be here a while from experience.

Untitled

• disinfo
• 12 Sep 2013
• 1

A grand total of five armour strands are connected to the connector block, then 1.5mm tri-rated thereafter.

The line conductors are 6mm and the supply is fed from a C40 MCCB set at 1.0

 

[Taylortwocities]

Lazy little whotsit, whoever did that.....

 

[ericmark]

Unfortunately not uncommon. I have many times argued about the use of earth clips and hose clips to attach copper cables to SWA as often the shells which should be inserted to stop the cable being crushed are missing.

So often I have come across work like what you show and when it has been corrected it is no better. I hope in your case it was corrected satisfactory?

 

[disinfo]

It was corrected, it was glanded into the panel like it should have been.

When I looked up the Cu equivalent of the armour, divided it by 40 and multiplied it by five I think it worked out at about 0.9mm from memory.

Zs at the panel was 0.11, I realise there may have been parallel paths.

Would the 5 strands of armour withstand a 2kA+ fault current should there have being a L-E fault?

 

[ericmark]

Would the 5 strands of armour withstand a 2kA+ fault current should there have being a L-E fault?

Glad to hear corrected but as to 2kA the problem is time. Depending on the trip of fuse feeding the time would be very short so instead of looking at the PSC we look at the let through value. Often the let through value of a fuse is less than a MCB. It is rare to have to worry about the let through value but on one job where we wanted to supply a radio battery charger from a tower crane supply we had to feed the RCBO with a fuse to reduce the let through value to less than the 4.7kA the RCBO was rated at.

So although wrong, likely the 5 strands would have been enough, but still not right.

 

[Taylortwocities]

A shark joint maybe?
http://www.alertelectrical.com/prodvar/4097/shark-416-ip68-2-4-cores-x-1016mm-gel-cable-joint

 

[mfarrow]

A shark joint maybe?

Not in this thread TTC!

I'm surprised that 5x strands is only 0.9sqmm though.

 

[Taylortwocities]

Whoops, sorry. Too much Merlot last night....

 

[JohnW2]

When I looked up the Cu equivalent of the armour, divided it by 40 and multiplied it by five I think it worked out at about 0.9mm from memory. Zs at the panel was 0.11, I realise there may have been parallel paths. Would the 5 strands of armour withstand a 2kA+ fault current should there have being a L-E fault?

As you say, the issue is whether the very short run of 5-strands would be able to stand a very high fault current. I would presume that most of the 'parallel paths' would be the the SWA armour itself - once in the cable,those 5 connected strands are probably in pretty intimate contact with the other 35, so all 40 are probably 'active' along the length of the SWA.

Kind Regards, John