Regarding the attached photographs: Will overload devices in a motor starter heat up and show significant temp rise when they are weakened or is this just terminal corrosion? I realize the temps within the heaters where I have my cursors are not accurate portrayals of the actual temperature within the heaters but are actually amplified by the "hall effect?", but I nevertheless take it as an indicator of some anomoly. I believe these are Square D of some bimetallic design, not sure. There is a 33 degree C differential between the upper heater terminals of A and C phases.
Also, the current draw was 115, 125 and 132 amps. Is that enough current imbalance to cause what we see here?

HSP9.doc (2,094 Kb, 63 downloads)

2 megabytes is too much for my slow dial-up. I'll take a look at your image tomorrow at work.

You mentioned hall effect - I believe you are thinking of cavity effect. The shiny metal heater has low emissivity which makes temperature measurement difficult. The one exception is looking into the recess of the heater where you have a "cavity effect" which makes the heater look very close to emissivity 1.0. In other words, this measurement is probably one of the most accurate measurements you can get of temperature of the heater.

Generally I don't pay much attention to the heater itself but more attention to the connections, especially the insulated cable attached (higher emissitivty).

Look for another simulular item with simular load as a baseline. The most likely cause is an internal problem. You should be able to verify this with a simple resistance or voltage measurement. The actual temperature is probibly higher than you read and not lower. With temperature levels like this I would flag this up for replacement as soon as can be scheduled. It should not be a costly item to replace anyway. Keep monitoring it untill it has been replaced. Also, what is it controlling. If if is a motor then your load imbalance could be an indication of some other problem within the motor.
Thermal Visionbob@thermalvision.ie

quote:

115, 125 and 132 amps. Is that enough current imbalance to cause what we see here?

Assuming that is the current reading from left to right, it is consistent with the temperatures shown on the lower attached cables (i.e. load current unbalance may be causing the temp differences on the lower cables themselves).

I'm still thinking about the connections and metallic parts.

It looks to me like broad areas of the cabinet are showing 70C. Is that a real temperature or an effect of a low emissivity setting. I suspect the latter.

What is your emissivity setting?

I also noticed the screw that appears the hottest (B phase) looks a little darker in the visual image. Two possible explanations for this:

Less likely explanation for this - actual heating causes darkening.

More likely explanation for this - that screw has a higher emissivity and only looks hotter.

Going again to the spot where you have your cursor Tref and Tmax - the cavity inside the heater always qualitatively looks like this. I don't really worrty about the difference between phases at that cavity. It may be partially due to viewing angle or load difference. But more imoportantly the heater is supposed to generate heat and since there are no connections in that cavity I don't worry about the temperature there.

I am almost ready to say this is one I would not worry about (other than current unbalance might deserve a little more investigation), but I'd like to see an image with emisisivty = 0.95 first. Also would be interested to hear other comments.

Regarding the current unbalance. It may be normal to see this 10% current unbalance on a motor load if facility voltage is unbalanced about 2% (max deviation from average). If facility voltage is balanced, there may be a concern you have a power system high-resistance connection contributing to your unbalance. Also note that this current unbalance places some extra thermal loading on the motor.

Pete and Bob, thanks for your input.
I always leave the camera emissivity setting at 1.0 but then try to stay conscious of shiney objects giving a false high temp, so I look for an angle which will minimize the reflection when I spot an apparent hot spot.
Unfortunately, this is a periodic customer at some distance so a re-shoot at .95 emissivity won’t be possible.
I have seen system-wide voltage imbalance before and it didn’t take long to suspect something common to all machines after I documented the same problem on the 2nd or 3rd motor in that MCC, but this one was one-of-a-kind. It could be that there is some voltage imbalance system-wide or as Bob alluded to, starter contacts or bucket stabs causing a voltage imbalance just for this machine, I didn’t check voltage. But I also didn’t see any infrared indication of the aforementioned. The MCE test that day showed a .56% resistive imbalance and an 8% inductive imbalance, right on par with the last 8 years of history.
So …, from all the feedback, I am coming to the conclusion that what we see is the result of a combination of some current imbalance and corroded connections at the heaters.

I didn't really recognize any evidence of a high-resistance/corroded connection myself.

I don't know what is typical for MCE test. Maybe someone else can comment.

Emissivity set to 1.0. Hmmm. So, the whole cabinet wall shown in purple there is at least 70C (160F)? Or even higher if actual emissivity is less than 1.0?

I was skeptical of that number (70C/160F) which is why I asked about emissivity. If it is the real cubicle temperature, it will be a tough environment for electrical equipment. I would want to understand why the cubicle is so hot. Maybe a very bad thermal design of the cabinet (too many componetns generating heat and not enough external surface area or ventilation). If it were here at the plant the first step would be to try to validate that temp with another measuremetn perhaps thermocoulple (with suitable electrical safety). May not be practical for you. Did the cabinet feel burning hot when you opened the MCC bucket door?

This message has been edited. Last edited by: electricpete,

I was on my third bottle of Gatorade that day, so ambient was very warm but I don't recall that the cabinet door was uncomfortably hot when I opened it. Upon entering the room, I do remember it feeling like my attic at the height of summer, so 140F+ is probably not out of the question.

Kurt,

It is always a challenge to interpret someone else's IR images, because of diffculting in knowing ambient condition, angles, etc. However, from what I see here, I am concerned that there are high resistance connection problems on center and right phases, at the top heater connection. My reasoning for this is that if the heating differential were due to load, then the source of heat (heater element) should dissipate both up the wires and down almost equally. There is a certain "chimney" effect, where components above the hotspot are hotter than those below because the heat rises and provides convective heating. However, there seems to be very little heating on the lower connection of the heaters, and I would expect equal conduction of the heating element heat on objects equidistant from the element. The discoloration of the screw also supports the idea that it is close to the source of heating. As far as the discoloration of the screw accounting for higher emissivity, I notice little difference in the slot of the screw, which also provides a cavity effect. You can see that cavity effect on the slot of the screw directly above it. The best emissivity in the image is the wiring insulation. The top-side wires on both center and right phase are signicantly warmer than either top-side left phase, or any of the wires below the heaters. Additionally, the entire upper half of the heater holder block is hotter on center and right phases. The lower half of all three phases, and the upper half of the left phase are much cooler. For these reasons, I believe that you have high-resistance connections at the top heater screws on center and right phases. I would suggest replacing the entire heater-holder block and all heaters. Inspect wires and connections as well for secondary damage. Please save the old one for inspection, and let us know what you find! Good luck.

Rich Wurzbach
ASNT Level III-Infrared and Thermal Testing
Maintenance Reliability Group

Hey Rich. Glad to see someone else comment. You may very well be right about the high resistance connections.

On the subject of emissivity, I don’t see how the slot refutes the presence of an emissivity difference between screws. Actually the opposite.. The slot is non-visible on thermal image on the hottest screw in center (indicating this one has high emissitivy). On some of the others the slot is visible indicating lower emissivity. Also note the upper left screw appears to be on a cool circuit but stands out as hotter. This clearly represents an emissivity difference rather than a temperature difference for this one (upper left). And this one is also slightly darker discolor from the visual image.

Also I am a little suspicious of emissivity difference on some of the other components. I think there is a mycarta board above the terminals in question and on the visual image it is lighter on the left-most (coolest) phase, suggesting possible different material (if the ones on right were darkened from heating, in general I wouldn’t think the color would be so uniform, but you never know). There are some congested areas next to the right two mycarta’s that tend to increase that temperature. But if I look in the open area to the right of the rightmost and left of the leftmost mycarta board, I see roughly the same temp.

If I look below at the cables I do see hotter cables on the right two phases indicative of the current imbalance higher on the right two phases.

As far as what’s going on above, the appearance of right two phases a little hotter I believe it may be just a combination of that known current imbalance which heats the right two phases, along with some emissivity differences (in particular that one screw on top center). I couldn’t say for sure.

I am fairly new at thermography and definitely may be wrong. I appreciate the opportunity to compare my analysis to others like Rich etc to see where I might be wrong.

Rich - What do you think about the ambient temperature? Is it normal or acceptable to have cabinets this hot?

Kurt,

If you have 8 years of history, you have a perfect baseline for this item. Has it changed in the 8 years, or was it always like this. If it has changed then simply replace it.

I have to say, I still go with an internal problem.

Also, on the issue of the discoloured screw. Emissivity is not related to colour, but is effected by surface structure, geometry, material, angle, wavelength and temperature. The discolouration we see in the image is a visual indication that a change has occoured. If a change in emissivity has occoured then it is more than likely due to a change in material or surface structure, caused more than likely by heat. Since the screws look like the same type and we must assume are all original (unless we can find out if one has been replaced) then we can say that some overheating has probibly taken place at some stage and from what I see in the image is still taking place.

Just to clarify I never said color=emissivity, but a difference in color is a clue that there may be a difference in emissivity.

I now tend to agree with Bob's comment that with these high tempeartures likely overheating is occuring. I still am pretty darned sure that some of what we see in differences between the hardware amounts to emissivity differences. Otherwise, how do you explain the hotter-appearing upper left screw on an otherwise cooler phase? And how do you explain the difference in screw slot appearance? (some slots appear hotter than the head and some appear the same). There is no question in my mind that there are emissivity differences present on these screws. They may be the result of overheating or material differences - it doesn't make that much difference - we know overheating is present.

An intersting thing - the coolest phase (left phase) is at least 120C (assuming no hotter reflecting source in the background), and probably higher if emissivity is low. Are you guys saying that phase is good?

In my opinion the thing that really stands out is the cabinet ambient temperature which causes everything here to be hotter.

I guess it's possible the heat might have affected to the connections (oxidation, relaxation) to the point that they have now degraded need to be replaced (replace the block). In theory those hot connections might be contributing to the high cabinet temperature and fixing them MIGHT reduce cabinet temperature but I doubt it. I'm guessing you will replace that block and still have ambient temperature just as high. Not sure whether you will have initial improvement in connection temperatures. But even if you initialloy improve the connection temperatures, if you don't reduce cabinet temperature you have just fixed a symptom (not a cause) and you will be back working on it again before long.

Just my opinion but interested to hear what else you guys think and what you disagree wit. I think I've learned a little here. (I never thought about looking in screw slots to determine emissivity differences before).

This message has been edited. Last edited by: electricpete,