hot spot on inverter output breaker

[electricpete]

Since I am brand new to infrared, please be patient with all my questions.

Here is IR plus visual image:

http://reliability-magazine.com/pub/inverteroutput.jpg

If visual image is too dark, it may be easier to see here:

http://reliability-magazine.com/pub/inv3.jpg

This is a 120 vac breaker on output of an inverter.

In lower left hand corner of the picture you see the top of a molded case breaker. It is a 2-pole breaker with right-angle metal busbars bolted on, and lugged cable bolted to the busbar.

1 - Cabinet ambient temperature is 24.6C
2 - The upper left connection (on top of bottom left breaker) is 24.6C
3 - The upper right connection (on top of bottom left breaker) is 44.2C.
(We checked sister unit and items 1 and 2 were identical).

The infrared shows green strip surrounded by two red bands. I believe the lower-temp green strip is caused by the thermal insulating effect of the lug electrical insulating sleeve. Since the terminal stud is not particularly hot, I think the heat comes from the crimp wire-to-lug, rather than bolted connection lug-to terminal.

Emissivity set to 0.95 throughout the image.

What is your opinion?

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

 

[electricpete]

Correction: #2 should read:
2 - The upper left connection (on top of bottom left breaker) is 27.7C

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

 

[IRstuff]

Is the surface to which the lug is bolted metal? It looks shiny.

Based on the temperature distribution, the crimp appears to be the hottest area and most of the other high temperature appear to flow from that, with one exception. Like your other posting, the wire that's attached to the crimp appears to have heat over a much longer length than one migh expect.

Does the greenish wire in the foreground get crimped to another high current wire at the lug? It looks very similar to your other pictures, where there are hot wires seemingly connected with other hot wires at some junction, which is hotter than the wires.

TTFN

 

[electricpete]

Hi IRStuff. Thanks for responding. I now have a directory on my computer hard drive which is named in your honor ;-)

To keep the discussion focused, I hope you don't mind if I snip pieces of your message to respond to.

"Is the surface to which the lug is bolted metal? It looks shiny."

Yes, definitely metal and looks shiney (low emissivity) to me as well. I didn't think about it before but maybe abrupt change from green to blue at interface of washer to metal busbar is emissivity difference?

"Like your other posting, the wire that's attached to the crimp appears to have heat over a much longer length than one migh expect."
Hmmm. 2-pole breaker both leads must have same current (especially ungrounded system). I'm pretty sure there are no further terminations for many feet. I can't understand what that would be telling me.

"Does the greenish wire in the foreground get crimped to another high current wire at the lug?"
I assume your talking green in the infrared picture. I'm pretty sure it is breaker bolted to busbar, busbar bolted to lug, lug crimped to wire, wire runs many feet Only one lug landed here. (in contract, the fused disconnect in other message had two lugs landed on the same terminal).

"It looks very similar to your other pictures, where there are hot wires seemingly connected with other hot wires at some junction, which is hotter than the wires."
In the other picture (fused disconnect) the wire associated with the lug which is hidden from view is hotter. So the hotter high-resistance crimp lug is hidden from view. The cooler low-resistance lug is on top and it does get it's heat from the terminal (lug below).




















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

 

[electricpete]

btw - thanks for the comments. That gives me a few things to double-check tomorrow.

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

 

[electricpete]

We did a follow-up inspection today and found the same temperatures.

Load is currently 30A. Inverter capacity is 15kva at 120vac - i.e. 125A.

I believe if the inverter were loaded to capacity, the existing temperature rise of 20C above ambient would increase to 320C which clearly is a destructive temperature.

This is based on assumptions:
ResistiveWatts = I^2*R
TemperatureRise ~ Resistive watts
Combining the above: TemperatureRise~I^2.
So a 4x increase in current gives a 16x increase in temperature rise above ambient?
Would you agree with this conclusion?

Any more comments?
Thx.

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

 

[Glave]

electricpete

One thing you may wnat to check is chaning the emissivity value for a spot on both the crimp and the bolted portion of the connection. From the IR image most, if not all, of the heat you see seems to me on the conductor itself. This being said the comparative low E value on the bolted tab on the connector and the crimp can mask themselves as "cooler" temps when the could be as warm if not warmer. Tyr playing with the E value for a spot and see what kind of temps you get at the connection end also. Suggest E of between .65 and .5 for the connection tab. It is also possible to conduct an emissivity test of the tab to determine exactly what the value is. Ahhh another good reason for a class