Circuit breaker load bolt torque

[cryomark]

Has anyone the installation specs for an old Westinghouse MC3800F circuit breaker? I'd like to know the recommended torque for the 1/2-13 bolt that mounts the 2 hole lug to the load side terminals.
The reason- I've a MCC that has exhibited the same symptom of spontaneous circuit breaker trips through 2 changes of the circuit breaker. (MC3800F thermal magnetic; MC800 rating plug; shunt trip). The machine (400Hp, 460V, 3Ø driven compressor) is soft started and will run lightly loaded for 1/2 to 10 hours before the circuit breaker will simply open up. The RVMS has been replaced as part of a general upgrade and both its internal monitoring and our external data-logging show no changes in the load(~150KW) before the trips.
When I removed the lugs from the load terminals, it took excessive effort to break the bolts loose. I remember another installation where an excessively tightened ØB bolt caused a malfunction of the push to trip button but that is not the case here.
Can the torque of the external connections significantly affect the thermal characteristics of a CB such as this?

 

[dpc]

Have you tried asking Cutler Hammer? They've always been really good at getting me data on old Westinghouse gear.

A very loose connection could be generating excess heat - an infrared check would spot that pretty easily, I would think.

You might check the conductor size. For an 800A breaker, the minimum size per UL testing is 3-300 kcmil copper per phase or 3-400 kcmil aluminum per phase The breaker's thermal element depends on the conductors as a heat sink. If the wire is too small, the thermal element can operate below normal trip point.

And, as you probably know, a standard 800A molded case breaker can only carry 640A on a continuous basis.

High ambient temperature is also a possible suspect.

Good luck.

 

[busbar]

 
Lacking specific manufacturer guidance, a default NETA torque value for ½-inch plated-steel hardware is 30-74 lb?ft, depending upon SAE bolt grade.

A very basic millivolt-drop test might be conducted. See

http://ecmweb.com/mag/electric_testing_contact_quality/index.html

 

[PWR]

The MC800 is an early generation (early 1970s) Westinghouse "Seltronic" solid state trip molded case circuit breaker. It is not a thermal magnetic unit and heat generated at the lug is unlikely to be the issue. More plausible is that some incompatibility exists between its method of sensing overcurrent/short circuit conditions and the current waveform being supplied to your drive/load.

 

[mc5w]

You also might need to remove metal oxides or sulfides from the busbar surfaces using silicon carbide paper. The Consumer Product Safety Commission specifies the use of #220 or #240 silicon carbide abrasive paper for cleaning copper of aluminum wire strands which is really a variation of the old Signal Corps method. For flat surfaces and fine wire strands #800 silicon carbide paper would be more appropriate. You would then need to use Ilsco Deox(R) on your threads and flat-to-flat surfaces. Silver plating also needs to be protected against corrosion with Ilsco Deox(R) as well so that it does not build up the black nonconductive silver sulfide tarnish.

I have found from experience that if semiconductive grit from silicon carbide paper would contaminate insulating surfaces then garnet abrasive paper will work but you will need to use more elbow grease before electrical grease.

You might also have to do this with the internal parts of the circuit breaker. I one time had to rebuild an ancient Federal Pacific Busway plug.

Mike Cole, mc5w@earthlink.net