Electrical power connection failure

[Bronzeado]

thread238-237001

Dear desertfox an prc,

Regarding the thread above, please, find attached a photo of the connection that has failed during a short-circuit.

As you can see, some wires of the cable has been melted due to the connection temperature. My question is: if the previous thermography inspection did not find any problem in that connection, why it failed?

Do you think that failure was due to connection ageing?

Best regards,

Herivelto Bronzeado

We use thermography inspection for long time and we trust in it. We only do maintenace in a connection if its temperature is above the nearby cable. That connection did not present high temperature.

Do you know if somebody do maintenance in electrical connection even it appears to be OK?

 

[desertfox]

Hi Bronzeado

Thanks for the picture,looking at it the connector looks in relatively good condition and the cables as you say have melted right at the joint interface.
I don't know what your reference temperature is, which says the joints okay or not okay, I assume the cable must have a normal ambient temperature rating and an acceptable temperature rise during short circuit conditions.
What I will say is that during a short circuit the temperature rise on the conductor will be extremely high and even higher if the joint has increased electrical resistance, due to the joint connector losing bolt tension with time and thus resulting in failure.
If you look at this site it gives some formula to calculate the temperature rise of a conductor under short circuit.

http://www.copperinfo.co.uk/busbars/pub22-copper-for-busbars/sec6.htm

Finally are there any mechanical stresses placed on the cable, like bends or twists etc because if there are, then maximum stresses would occur just at the joint face ie where the cable as melted and over time with normal temperature fluctuations these stresses could cause a problem when the cables under short circuit.

desertfox

 

[waross]

As you can see, some wires of the cable has been melted due to the connection temperature. My question is: if the previous thermography inspection did not find any problem in that connection, why it failed?

If he thermographic testing was done under conditions of light load, it may have missed a poor connection.

Bill
--------------------
"Why not the best?"
Jimmy Carter

 

[Bronzeado]

Thank you desertfox, I will take a look in the material you have posted.

Regarding the forces caused by the short-circuit current, I did a quick calculation about the forces acting on the strands of a cable considering a current of 15kA and a distance of 0.1mm between the strands of a single cable with 15 strands (for simplification)

F=2x(IxI)x0.0001/s

I= 15000/1=1000A, s=0.1mm, then

F=2x1000x1000x0.0001/0.1=2000N=2kN

Do you think this force will twist the cable strands if they are not in a good symmetry? If so, what would be the effect in the connection?

Bill, the failures occured in three connection which were in "series" in the same cable.
Also, after we fixed the problem and made the infrared inspection in all connections, after another short-circuit (in this case a three-phase short-circuit) three connections failed, melted the cable in all the phases in a way simlar to that in the photo.

I am trying to understand what is happening.

I would very much appreciate your thought.

Best regards,

Herivelto Bronzeado

 

[MLPeacock]

It may have been lightly loaded when thermal scanned. Was the clamp on amperage reading taken at the time and followed with torgue test?

 

[desertfox]

hi Bronzeado

I was suggesting you look at the temperature rise for the short circuit current not short circuit forces, I was meaning any mechanical stress created on installation.
Further if you check a connection what torque or preload do you put on the bolts.
Look at this formula on the link I posted and further down that page

t=0.051(A/I)^2 * ((1+0.0076*theta)^0.5-1) (this formula).

Incidently your thermography only gives surface temperature, although that might be okay most times.
Whats your base reference temperature?

desertfox

 

[electricpete]

Since there has been some discussions of limitations of thermography, I'll measure another important limitation that can apply to overhead lines which are viewed from a distance: The temperature is averaged over a "spot-size" or more correctly "Measurement Field of View". As you increase distance from the object, the measurement field of view represents a bigger and bigger area. If you are averaging the hot-spot in with cooler surroundings (worst case blue sky), the reading will be low. Telephoto lens is sometimes needed. If you have a competent thermographer, he is aware of all this.

=====================================
Eng-tips forums: The best place on the web for engineering discussions.

 

[desertfox]

Hi Bronzeado

Have a look at this paper not because it uses spring washers but it underlines the principle of how joints can come loose:-

http://docs.google.com/viewer?a=v&q=cache:61Gs7K7GzqEJ:

http://www.solonmfg.com/springs/pdfs/ElecConnWhitePaper.pdf+failure+of+copper+bus+bars+due+

to+over+tightening+bolts&hl=en&gl=uk&pid=bl&srcid=ADGEESgG3UNt-d5FNtCUSmgDGC2Q2LJ77V8vEANc7XGFO_56VEcYxtutJFQNksVNAyApIrTI09ykptKghv4NpJzr82S65Um_e_7oMdp_BTOcW4ybLL-pEUl77abNCEnzBgOoejf8e8D7&sig=AHIEtbQfilD48sXsWE2Vf9aVx_Wa8HLunQ

 

[desertfox]

Hi again

try this link

http://www.solonmfg.com/springs/pdfs/ElecConnWhitePaper.pdf.

 

[Bronzeado]

Dear friends,

What I am trying to do is to understand what is happening "physically" with our connections.

Which physic phenomena occur in a connection during a short-circuit?
There is only heating (RxI^2) or it is possible to have electric arcs between the metal surfaces inside the conector that do not really make electric contact?
Can these arcs (if exist) trigger the process of connection failure?
Does this make sense for you?

Best regards,

H. Bronzeado

 

[desertfox]

Hi Bronzeado

Your connection problem cannot just be put down to one thing it may be several thats why we keep asking for information.
Clearly your joint is over heating, one of the main causes of that are joint the bolts loosing off over a period of time which increases the joint resistance, this in turn increases the I^2R at the joint and hence the temperature rises even for normal loading.
Now as someone as already stated at low loading or even normal current loading this increase in temperature may not be picked up, however at a short circuit condition the poorly connected joint and the huge increase in temperature rise are to much for the connector.
It could also be age of the cable and the above and we can't be sure without more information, a joint thats been overtightenend can produce a similiar failure over a period of time to a joint thats not tight to start with.

desertfox

 

[jghrist]

I would not use this type of bolted connection on aluminum conductors. When the aluminum heats and expands during normal loading, it permanently deforms. When the loading and temperature decreases the connection become loose, increaseing contact resistance and allowing oxidation of the conductor.

Belleville washers can be used for pad connections with compression terminals on the aluminum conductor, as shown in desertfox's link, but I don't think they would be effective in a connection where the aluminum conductor is directly bolted.

 

[Marmite]

I agree with Jghrist. The connector looks unsuitable for the conductor. I would suggest you change them for compression type tap connectors, rather than bolted. If the tap needs to be disconnectable then you could use a flag type fitting and a bolted palm compression lug on the tap.
Regards
Marmite

 

[BJC]

What's the catalog number of that connector??
What is the catalog number of the connector in that picture?
Go here and look on page 479. What you have looks like a type NSNT/NHNT connector.

http://portal.fciconnect.com/Comergent/fci/documentation/2008_master_catalog.pdf

This is the Burndy master catalog.

IF you do have the NSNT/NHNT connector so that would explain your problem. Bolted connections with aluminum connectors with aluminum allways need carefull consideration.
Teh fitting above is silicon bronze and will corode aluminum just about where you see it.
There is lots of application information in the Burndy catalog. Call you local rep if your in doubt.
For the configuration your using I would used cripped aluminum lugs ( with no-ox compound inside).