Electricpete, I think you may be mis interpeting John Snell.

A Thermal Gradient is a gradual change in temperture over a distance, not another term for indirect heating. However, a thermal gradient will be present when looking indirect measurements as the source of the heating will be some distance away.

Although the matrix system described by Mr Snell is better than other methods of evaluating anamolies it is not perfect and I think you need to treat results obtained with this method carefully.

Despite John says in the paper, I think the matrix could stand for some development. I used the matrix on this anamoly, and I ended up pretty much where I started.

That is, the severity depends variable of the unknowns.

MMM

“Electricpete, I think you may be mis interpeting John Snell.”

I agree with your definition of thermal gradient in general but I don't think it's how John Snell is using it.
See
http://snellinfrared.com/tt/news0703/July-2003_TT.pdf
Page 3, lower left hand corner
Stage 1 question
“Are thermal gradients large or moderate? [5]”
Note 5 as follows:
“5. Such as a bus stab, enclosed bus connection, motor terminal box, massive
connection, or connections inside any oil-filled device.”

It is crystal clear in this particular instance he uses the term thermal gradient to refer to indirect heating.

Hope I'm not sounding overly argumentative here Bob. I don't think the terminology is overly important but I'm glad you brought up the clarifiction about other possible uses of the term. In this particular case I'm sure he is referring to thermal heating and using that question to boost the severity in the presence of indirect heating.

“In your experience do most buses have porcelain insulators?”

Porcelain or Mycarta or some other non-conductor so support the bus within the duct. If you look at your image, wherever the bolts are is hottest. I think this is where the support insulators are that conduct the heat from the bus to the duct. Even though we consider them a poor enough conductor that we are concerned about indirect heating, they are still a better heat conductor than air. In between the insulator stations the temperature is cooler where you have the insulating effect of air. I don’t have any experience with hotspots in this type of bus. My gut feel is it could be a lot more than 20C hotter than external temperature. But probably the OEM knows more than me.


I agree the matrix is not perfect. It is a helluva lot better than using temperature alone. And depending on the organizational dynamics, sometimes a repeatable predefined structured approach to evaluating and communicating anomalies is preferred by customers over an expert opinion. (that is my environment)

For this particular anomaly, if you use the version at http://snellinfrared.com/tt/news0703/July-2003_TT.pdf , the answer “Yes” to the stage 1 question “Are thermal gradients large or moderate?” puts you into the highest severity category “high probability of failure.”).

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

Pete:
I think obviously "thermal gradient " can refer to any differential temperature.
However in the case of indirect heating, since the internal temperature is unknown, a gradient can only be estimated, and those estimates can vary greatly.

The Thermal gradient between two phases of a 3 phase system can also be misleading, as inevitably some of the heat from the anamoly travels to adjacent phases.

The thing I was thinking about, was to estimate what the temperature of a given conductor might be in the abscence of an anamoly.
If we have 3 identical conductors, with roughly equals loads, exposed to the same environmental conditions, what is the most useful reference point?
Doesn't this reference need to be at a point on one of the conductors where it is evident in the thermograph that all of the energy from the anamoly has been expended? The only remaining energy would be that of the load on the conductor.
This was why, in my 1st thermograph, my reference point for the DT was up near the top of the bus, on the center phase.. Same conductor, same load, same environmental conditions, yet out of the range to which the heat from the anamoly conducts....

Once again I agree in on other definitions of thermal gradient in general.

In this particular context of "Stage One Questions" my opinion remains unchanged. Snell made himself clear in note 5 that he is using the term thermal gradient here to refer to indirect heating.

http://snellinfrared.com/tt/news0703/July-2003_TT.pdf
Page 3, lower left hand corner
Stage 1 question
“Are thermal gradients large or moderate? [5]”
Note 5 as follows:
“5. Such as a bus stab, enclosed bus connection, motor terminal box, massive connection, or connections inside any oil-filled device.”

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

Mr Snell is refering to a thermal gradient, a gradual change over a distance.

When we look at the exterior of the bus ductwork and it is being heated by an internal problem then the it is reasonable to assume that the internal fault is significantly hotter this can be either by conduction through all insulating supports or by radiation across the air in the bus ductwork. As we are looking at the resulting heating effect and not the source we must assume that a significant thermal gradient exists between what we see and the source of the problem. This is based on experiance and assumption, it is quite different to trying to quantify results based solely on external surfaces heated indirectly.

It is also worth pointing out that the paper you just posted a link to was published after the matrix paper posted earlier. This is I assume because Mr Snell himself sees there are flaws in the matrix idea.

And no, you do not sound over argumentative

That is not the way I read it.

But I am happy to agree to disagree on what Mr. Snell is saying. At least the important point I think we all agree that a given observed temperature rise should be viewed with much more caution if it results from indirect heating.

There has been a parallel thread on Snells website. In his latest post he referred to the difference between the anamoly and surface temperature as thermal gradient. I have not asked him this question directly, and he may have been using the term in the general sense.

But you can go to the website and see the thread.

http://snellinfrared.com/messageboards/wwwboard1/index.asp

From your link John said:

"In any regard in this type of situation, an enclosed bus duct, as well as in other similar situations with a large thermal gradient, an internal anomoly will often not have a very significant looking thermal signature. So even if we see a subtle signature, it can point to serious internal heating."

It looks to me like he is using thermal gradient to refer to indirect heating. i.e. thermal difference from the internal contact area to the external area that we have access to.

You will have about as much luck changing my mind as you will have getting a stubborn mule to walk backwards ;-)

Personally, I don't know: second opinions can be - well, questionable. Went to my doctor and looking back over my shoulder noticed he was rubbing his fingers with lubricant and crossing two - I asked, "Doc, what's this all about". He said just in case you want a second opinion. So I only paid half the bill and went to another doctor. After the exam I said I think I need a second opinion - he said, "OK, you're ugly too".

Sam,

Nothing like some "thermal gradient" humor!

Walt

Pete: This way you stubborn mule!
In order to answer the question in the matrix "Is the thermal gradient large?" you must know both numbers. It is like a delta.

agree or disagree?

You cannot quantify the gradient in the abscence of the second number, as in this case of ***hidden*** indirect heating.

In this case, the indirect heating is an instance where the source of the heat cannot be observed directly. Since it cannot be observed, it cannot be measured or quantified, and if it cannot be quantified it cannot be used in an equation to establish a gradient or delta.

In fact the delta T's availble in any case are those in which both values are observable and measurable..no?

What is the difference between a gradient and a delta?

A gradient requires even more data.. the rate of change in the variable, as in part of some function.

So it seems to me the use of the word "Gradient" or even "delta" in reference to hidden indirect heating is improper, no matter who uses it in this way.

So we can establish a delta or gradient in the case of indirect heating, but only if we measure the heat of the source, which I cannot do in this case.

No mas. We have to agree to disagree. Not on the definition or proper useage of the term gradient (I agree with you guys there), but on the way John Snell is using it in the Stage One question.

Greetings all -
I am jumping into this conversation late, since I don't usually hit this board.

I have done some sample analyses of the heat transfer related to indirect vs. direct measurements and reported them in published papers. In one instance, a contactor with a hot contact surface and an sealed box around the contacts, I estimated that an approximate 27F rise on the outside indicated a 200F interior rise for a vacuum filled component and a 20F rise meant about 100F rise interior for an air filled component.

I have tried attaching a pdf of one of the papers. See the second section.

Also look at the iPresentation on Direct vs. Indirect Temperature Measurements, on the ReliabilityWeb. I discuss these issues there.

The primary factors in the relationship between the observable hot spot and the source hot spot are the sizes, materials, and heat transfer paths available for the heat to get out. This will vary from case to case.

If you have the design information for the system - dimensions and materials, then directly applicable calculations can be done. Feel free to contact me to have them done.
Jack

Using_FEA-based_heat_transfer-_extend_scopeIR.pdf (1,073 Kb, 5 downloads) Heat Transfer Calc Paper

Great to see you here Jack - we were beginning to think you had something against reliability! Thanks for the contribution.

I hope you will stick around.

Stay warm up north!

Terry O

Thanks Terry -
Mostly just keep forgetting the forum is here. I have moved it up higher on my bookmarks list -- and turned notifications on. (I'll be buried now!
Jack