Infrared Scan - Transformer Radiator Fins

[Shells]

Hey guys

So we had an infrared scan done on our radiator fins. It would appear that one side does not have any flow. The transformer doesn't seem to have any overheating issues but the transformer is also lightly loaded. There is no valve or plug to argue that maybe it is preventing the flow. I am assuming there is a circulation problem but other than low oil level or some other obstruction, what else can cause that? Any other suggestions?

Regards
Shells

 

[zog]

My first thought would be low level, it looks from the photo like the left side is higher than the right so if the transformer is at a slight angle the top of the fins on the left may be above the oil level preventing any flow.

 

[Shells]

My first thought would be low level, it looks from the photo like the left side is higher than the right so if the transformer is at a slight angle the top of the fins on the left may be above the oil level preventing any flow.

Thanks. That's what I was thinking also so will check that today.

 

[Chamuit]

You need to adjust your emissivity on the next scan. It shouldn't be higher than .97 for an IR scan.

 

[Shells]

You need to adjust your emissivity on the next scan. It shouldn't be higher than .97 for an IR scan.

Hmm.. ok, I didn't even look at that. Didn't do the scan but will definitely ask the service provider to look at that! Thanks!

 

[Chamuit]

I saw this in a class. I was trying to remember who it was that had a picture like this. Anyway... The transformer I am talking about had a manufacturer defect and only half of the fins were open (internally) to allow for flow. The orifices were not there to allow for the oil to flow through them.

 

[Shells]

I saw this in a class. I was trying to remember who it was that had a picture like this. Anyway... The transformer I am talking about had a manufacturer defect and only half of the fins were open (internally) to allow for flow. The orifices were not there to allow for the oil to flow through them.

Thanks. Will try to contact the manufacturer also

 

[iwire]

You need to adjust your emissivity on the next scan. It shouldn't be higher than .97 for an IR scan.

Isn't that largely irrelevant in this case?

We don't need to know the true temp to see a problem in many cases.

 

[Chamuit]

Isn't that largely irrelevant in this case?

We don't need to know the true temp to see a problem in many cases.

More than likely, yes. That is assuming that the surfaces on the transformer are all painted or coated with the same material.

It does show that the operator of the camera, more than likely, is not trained in its use (FWIW.) Black bodies are the only 1.00 emissive things measured.

 

[iwire]

More than likely, yes. That is assuming that the surfaces on the transformer are all painted or coated with the same material.

Changing the E setting is not going to take care of the problem of shooting two different surfaces at the same time.

It does show that the operator of the camera, more than likely, is not trained in its use

Maybe, or it could show they know exactly how it works

The last Flir class I went to the trainer was of the opinion we should not be reporting temperatures anyway, just comparisons and recommendations.

 

[Chamuit]

Changing the E setting is not going to take care of the problem of shooting two different surfaces at the same time.

We don't need to know the true temp to see a problem in many cases.

You are correct, but, we were talking about seeing of there were problems. I just pointed out that the emissivity would need to be uniform.

Maybe, or it could show they know exactly how it works

...or, not paying attention...

The last Flir class I went to the trainer was of the opinion we should not be reporting temperatures anyway, just comparisons and recommendations.

Mine wouldn't have accepted a project turned in with the emissivity at 1.00.

 

[Shells]

The Service Provider admitted he is only looking for the relative delta T and not the absolute values and error is expected. Point noted though

 

[FionaZuppa]

is the top plane of the fins level? if its tilted to the right then by flow dynamics i would say this is normal. you probably have slightly higher flow on the right vanes, thus more heat will be there.

 

[GoldDigger]

is the top plane of the fins level? if its tilted to the right then by flow dynamics i would say this is normal. you probably have slightly higher flow on the right vanes, thus more heat will be there.

I would call the difference between the right hand fins, the transformer cover and the left hand fins more than slight!
Cover is cooler than right fins and left fins are cooler than the cover.
Or didn't you notice the set of fins on the left?

 

[11bgrunt]

Hey guys

So we had an infrared scan done on our radiator fins. It would appear that one side does not have any flow. The transformer doesn't seem to have any overheating issues but the transformer is also lightly loaded. There is no valve or plug to argue that maybe it is preventing the flow. I am assuming there is a circulation problem but other than low oil level or some other obstruction, what else can cause that? Any other suggestions?

Regards
Shells

Is the original image available for you to send me? I am looking for the image that can be processed in Flir Tools. I would like to experiment with the image. I am working on a transformer now that is running 100% of nameplate kVA and would like to see the similarities in the tank and fins. Are there fans on your transformer? I attached the image below and when saved it could not be processed. I send in emails often but somehow adding to the thread changes the image properties.
Thanks

 

[kwired]

Is the original image available for you to send me? I am looking for the image that can be processed in Flir Tools. I would like to experiment with the image. I am working on a transformer now that is running 100% of nameplate kVA and would like to see the similarities in the tank and fins. Are there fans on your transformer? I attached the image below and when saved it could not be processed. I send in emails often but somehow adding to the thread changes the image properties.
Thanks

Put your images on a host site like Photobucket and then use

 

[GoldDigger]

However the FLIR software may be looking for the FLIR camera metadata rather than just the image. As long as the photo host preserves the metadata from the uploaded file instead of stripping it I expect it will work.
My guess is that the Forum software is deliberately stripping the metadata on attachment uoloads to avoid accidental privacy problems.
Hopefully that is not done when rendering an external image, but if it is you can still follow the link directly from a separate browser tab.

 

[MD84]

How long has this transformer been in service? Is there any oil sample data? Has the oil been processed recently or since the problem was noted? Do you have an oil level indicator?

Oil could be sludging causing an obstruction at the bottom. An air pocket could be present in the top of the left radiator.

Do you have a digital camera image (photo)?

Sorry, a lot of questions.

 

[iwire]

Put your images on a host site like Photobucket and then use code in your message and we will see it as you left it on the host site.
[/QUOTE]

[QUOTE="GoldDigger, post: 1732814, member: 139209"]However the FLIR software may be looking for the FLIR camera metadata rather than just the image. As long as the photo host preserves the metadata from the uploaded file instead of stripping it I expect it will work.[/QUOTE]

I will be surprised if services like photobucket will accept the type of file that contains the raw data.

 

[GoldDigger]

I will be surprised if services like photobucket will accept the type of file that contains the raw data.

An "ordinary" JPEG or other compressed file can still contain metadata. You do not need a RAW format file for that. I do not have any direct information on what the software requires.