Anyone do thermal imaging ?

[Warrior]

Had a request from one of my customers

Can you spend half a day, walking round with the thermal imaging guy, opening panels etc, as i know the plant like the back of my hand.

Anyway i did it, interesting !

It turns out that this company ( who only do thermal imaging ), do this twice a year for this customer.

The customer insists it is done twice yearly, and it clearly works well, as they have an excellent predictive maintenance regime.

Now the interesting part, it costs

 

[sprocketflup]

so what is it used for?

and I think your dog is boiling lol

 

[revjames]

I have assisted a thermal imaging survey many times at the aluminium smelter. The guy would do a comprehensive survey with a camera worth

 

[Warrior]

As a site report, mainly preventative maintenance, looking for hot spots, like loose connections, phases out of balance, overloading, bearings running hotter than normal.

The idea is to find it before it breaks, therefore it can be rectified at a convinient time rather than it failing halfway through a production shift.

The software produces a report, and you can analyse images, checking pinpoint temperatures anywhere in the image.

End result is that the customer has a report with images, of items to address before they become problems.

Im sure this tool will now help with a few more other things aswell now

Another sample image below

 

[Warrior]

I have assisted a thermal imaging survey many times at the aluminium smelter. The guy would do a comprehensive survey with a camera worth

 

[revjames]

The key is having someone who knows the plant and the potential problem areas then. We had a crazy policy on H&S. Because someone in the same group (in Australia)had an arc flash incident. it was decided that anyone opening an MCC panel or switching, had to wear full arc flash gear (hood, gloves and overmitts, or a whole suit if it was HV)

This meant it was almost impossible to do the thermal imaging survey as the guy with the camera would have all the gear on as would the person opening the panels. I think they conveniently made an exception when this was pointed out!

 

[Larnacaman]

Most of the bigger continuous process companies will have ongoing in house thermal imaging programs in place via there maintenance departments.

It's a proven technology for locating weaknesses and faults in there infancy, thus allowing a timely replacement of components etc, with the least amount of disruption.

Oil companies have been using them for years....

Been involved with the use of them on a number of occasions, when major plant has been installed and had cause for concern on there performance. One such time was with a 3 MVA MV Stand-by gen set, where it found more than just the one fault that was originally suspected...

5000 is quite an initial investment, but if you have the customers, then good luck to you. Brilliant for checking ACB's GCB's and the like, for high resistance contacts, and newly made off/repaired or in-use buried cable joints, ...before being back filled!!!

Great bits of kit Warrior... What make and model did you go for??

 

[m4tty]

Why does the middle one look hotter mate. Is it because the outside phases can dissapate heat easier? Cheers

 

[Andy™]

probably because L2 has a bigger load then L1 & 2 (i know he said 'balanced', but i very much doubt it would be exactly the same per phase. unless the only thing there is 3 phase motors etc)

 

[Sidewinder]

Canoey,

Have you got the optical image of this so I can "have a go" at the diagnosis?

So you have 800A rated incomer.

Drawing 460A per phase +/-10A?

Ye?

That is between 450 & 470A? A 4% imbalance?

Also, there is a Delta T of 6.3 Deg C, a 20% imbalance?

They look like copper bus bars, though not having seen the "job" it is difficult to tell from the pic.

Been a while since I have done any imanging fault diagnosis now.

Do you have any Ductor readings from the customer from the install?

You can tell the load is pretty balanced the N bar is running at 24 deg C only 7 deg above ambient (roughly!)

Well, Boaty Boy how am I doing so far?

 

[Sidewinder]

OK,

Not bad then for a first guess me thinks.

Any chance of the optical image then?

Pop off to work & pick it up for us will you!

I took the imbalance at the hottest point to the corresponding coldest point.

I have a customer with what appears to be a very similar incomer but this is a 1600A job.

Looks like the lower ones are the DNO incomers terminated into a switch panel with an insulating horizontal mounting bar?

Are the incomers Al or Cu?

Are they the DNO lugs with 6 holes? 2 for mounting and 4 "spare" round the corners?

Or 6 for mounting?

The outgoers are Cu bus bar?

Are the lugs on the incomer tinned Cu or Al?

What is the overall environment?

Dry, damp, cold, heated, indoors, in an external switch room etc. please?

1 invoice down...

 

[Larnacaman]

A ductor tester will confirm a higher resistance at the L2 joint. It looks as if it could do with a little maintenance.

Aways better to check such areas with a ductor test, but that calls for down time!!! which is where thermal imaging comes into it's own , ...being able to show up problems while in actual use...

 

[septiclecky]

so what is it used for?and I think your dog is boiling lol

Clearly you don'r know much about Thermal Imaging the scale at the bottom shows you the differences between temps red areas shown are only 13 degrees

 

[Sidewinder]

Larnacaman,

That was my point in asking.

That is did the L2 have a higher installed R, or has this developed since.

I agree that it looks like maintenance is required, however, it should be possible with suitable info that should be obtained during such an inspection to identify the possible fault in ore detail than "the L2 connection needs maintenance"!

Not having a go, but this is what TI fault identification is all about!!!

Paul

---------- AUTO MERGE Post added at 16:10 ---------- Previous post was at 16:09 ----------

septic,

I thought the poor mutt was quite cold actually poor thing!

 

[Larnacaman]

Larnacaman,That was my point in asking.

That is did the L2 have a higher installed R, or has this developed since.

I agree that it looks like maintenance is required, however, it should be possible with suitable info that should be obtained during such an inspection to identify the possible fault in ore detail than "the L2 connection needs maintenance"!

Not having a go, but this is what TI fault identification is all about!!!

I was referring that a ductor test would be far more accurate than the thermal imager, but that needs a system ''downtime''!! I didn't mean that a ductor test needs to be conducted to confirm the thermal imager. The imager quite clearly shows there's a higher resistance at L2.

 

[Warrior]

Larnacaman - Went for the Flir i60

Sidewinder - That pic was took in the back garden, when he was out to do his business, so yes fairly cold

This thermal imaging should be a nice learning curve for me, i can see on here theres a few that know it well.

Like i say, at the minute its just for my regular customers, whose plant i know very well, as you need a fair understanding of the equipment to question....is it too warm or is that what should be expected of this.

Ive learned that its not just looking for hot spots, its about looking at differences in temperature, like the 800a pic in this thread.

 

[Sidewinder]

OK,

Greysale is a bit grainey.

I was hoping you might have had a true optical image rather then a virtual IR image of the joint.

I know the 6 hole lugs you mean.

With the info I have here I'd have to go for 1st loose DNO lug on the middle phase, second would be poor crimp on DNO lug, 3rd would be something to do with the link bar mounting.

Where were you clamping the I?

DNO side or customer side of this joint?

If customer side, was there "leakage" at the mid point mounting, rare though this would be.

i.e would the I on the DNO side be higher than the I on the consumer side of this joint?

Not the sort of job where you can wiggle the wire to see if its loose either!

Well...

 

[Sidewinder]

Canoe,

I mean a visible optical image not the IR one, i.e. like that tken with a digital camera.

Anyway, I'll stick by my ideas above as I can't glean any more from the pics really.

Only one other thing I can think of could be a cracked link bar though I would have expected a very hot line along the crack which I can't make out.

Leakage is out as the current values are DNO side of the links.

Another thing that has just come to mind is a foriegn body between the lug & the bar?

Failing that the only other thing I can think of is a bad bar, i.e. high R, poor copper etc.

Did you get the fault fixed? If so what was it?

Cheers

 

[Larnacaman]

Bus bar arrangement looks as if it's site fabricated, and not very well either!!

Alignment, is poor and therefore maybe possible stress at the bar insulator connection.

But from the thermal image, my guess would be the crimping on that middle phase, coupled with some sort of problem at the bar insulator position...

What i don't have an answer for, is the single circular hot spot above the supposed bend on the bar of L2, and a much fainter one over on the neutral bar in the same location???

 

[Sidewinder]

Larnacaman,

This is the problem with IR images, IMHO you NEED the true optical image to go with them.

I would say the bars are in a panel board?

Thus just not well made?

Most seem to be a bit rough if you ask me!

The ideas you have are along the same lines as I have as you can see.

However, I'm now not totally convinced!

It's the insulator bit over the link bar that get me really.

The hot spot could be several things, but if it were a big issue surely the bar around it would be hot too?