Root Cause Analysis - Transformer

There are a few typical issues that pop up in transformers:

• Heat?
• Overheat?
• Insulation degradation?
• Overload?
• poor manufacturing practices?

In my opinion it looks like a heat or overheat issue?

There are other guys that will comment and they are very good transformer technicians!

Thanks for the reply! Tomorrow I will post additional information about the transformer.

Next time you post, consider holding back until you have all the information relevant to the problem, Mildred. You'll get better answers more quickly.

OK!

Has it failed? If so, what is the nature of the failure?

Thanks for the reply! Tomorrow I will post additional information about the transformer.

As the question monitor for the month of September, it is my duty to inform you that you have exceeded your allotment of questions for the month.

You have been added to the list of those who are guilty of overindulgence in interrogation.

You will be on probation for the next 23 minutes.

Similar offenses will be dealt with harshly and may include the deduction of one Atta-boy.

But your Honor, I was just trying to get to the root cause of the problem, namely the OP's reliance that one picture is worth a thousand words!!

I'll take that under advisement and render my decision in 19 minutes.

You are just angry you didn't think to ask those questions.

Thanks for the good answer.

Looking forward to more details.

I love the way this thread has progressed.

Since you work for the insurer of the company that roasted this transformer, you might want to review your company rules on how much the insured must comply with an investigation for full payment.

Thank you! Next time I will put all the information on the main question.

Actually I work in an engineering firm that provides services to many insurers. In theory, the insured should provide all the information, but it hardly occurs (Brazil's problems, you know...).

And if there is delay in research (due to information request, for example), our office ends up being blamed. So in most cases, have to do the analysis with limited information, you know? ...

Excellent reply..

It does warm you heart to get a reply to questions..

My take on your post is this and I conduct in my line of work something called DIFA, Dismantle, Inspection and Failure Analysis.

The questions you have answered either "Not checked", "Not inspected", "Not been told", "Not received information"... etc, are the questions you need to GET answers for.

It might help if you talk to the senior management and explain the following;

1. the omission of information required, and give them a list of info you NEED NOW, will only lead you to make your own conclusions, and you MIGHT not rule in their favour (I guess they are trying to claim?)

2. You need to be able to inspect anything and everything you want to and items so stated in replies to you.

3. You are there to ensure that things are done in a proper manner

4. You are there to help them and NOT work against them. (most important)

For me.. any information requested and then hidden or not provided is an admission of guilt and/or bad practice.

The load history is vitally important, as a transformer does not overload itself!! Also the trip settings, as previously stated by other contributors to your OP

They are trying to claim, you, I guess have to authority to reject the claim. You have to ensure they understand that, as a lack of information from them does not allow you to make an educated decision.

Thank you!

I fully agree with what you said. Usually we try to conduct a good inspection (following the recommendations of IEEE C57.125-1991) and request the documents beforehand, but some insured do not collaborate in any way ... And unfortunately, as I do not work in the insurance (I am only a service provider), I have no authority to reject the claim.

Nevertheless, soon I'll post more information on this case and you will understand what is my real question.
See you later!

You do not directly control the insurance claim money, but you do control your report. It is normal for any report to include appropriate comments on the ease or difficulty to obtain information on an incident.

As for the conditions on the day of the failure, you should be able to at least identify the weather conditions from archived news reports.

Thank you for such a detailed and concise reply.

Very refreshing when a typical poster asks a question and then disappears!

Certainly not OT!

I would add:

Insulation aging/degradation.

Possible arcing sites.

Normal discoloration of the winding insulation is a remote possibility.

The location and type of damage is fairly obvious, but I'm waiting for answers and more pictures before making any assumptions as to the probable cause. Only the OP can provide those previously requested details, plus those which require a close internal inspection, so everything is pure conjecture until then.

Yes I agree. Hopefully they will send it out to a competent repair shop for analysis to determine what is taking place.

Heat.

That is the root cause.

HEAT!!!!!!!!!!

Heat causes the organic coatings to revert back to their original constituents. (Elements/molecules/atoms)

Carbon is one. Carbon is black.

Toast looks black when it is burned.

Your transformer is TOAST.

Find out what burned it!

Hey! You're on to something! Maybe the reason the OP isn't replying is because he/she/it is "toasted"!

Years ago, I worked for the DOD and I've seen this before. Definitely was overheated. To much current, but why?

Definitely need more information.

Heat.

Hello guys!
I'll put some more information and photos about this case:
It is a transformer 7.5 / 9,375 MVA, 13.8 / 34.5 kV. The event ocurred on 06/10/2015. I have NO information about the circumstantial conditions, record of the protections that operated, voltage, current and load.
Following the RAMConsult questions:
1) What did the DGA (Dissolved Gas Analysis) show immediately before and after the fault?

The diagnosis made based on the NBR 7274 (Brazilian standard based on IEC 599) indicates overheating (T> 700 ° C) for the sample of 07/17/14 (the last DGA).
--

4) What type of fault occurred?

An arc of high energy in the X1 terminal region.

--
12) What was the BDV (Break Down Voltage) of the oil, pre- and post-fault?

Break Down Voltage in the last sample: 70,9 (min 30).
--

14) Is this a three winding transformer?

Yes.

--
NEW QUESTION is:Based on the above, I wonder if my assumption is correct that the dark markings are related to the overheated operation (indicated by DGA), unrelated to the occurrence of electric arc on the X1 terminal region.Suppose has been a rapid event not associated with external causes (external short circuits, for example), so that the time involved in the event can not result in thermal effects (subtransient period).
With that, I would intend to aggravate the depreciation of equipment, based on the conservation status before the event.

Discoloration was on bottom part of the thing. Could be a collateral damage both from heat and chemical reaction, perhaps.

If the event happened on 6/10/2015, why is one of the images dated 4/29/2015 ?

Sorry... The event happened on 01/21/2015.

From the information you have provided it is apparent that the transformer suffered overheating over an extended period of time resulting in subsequent insulation damage and eventual failure.

Without a baseline accurate determination of the root-cause(s) is not going to be easy.

The cause of overheating must be identified if future transformer failure is to be avoided.

Here are some suggestions of things to consider:

Is the transformer large enough to handle the applied load without suffering from overheating? (Often over a period of time additional small unplanned and undocumented electrical loads get attached to a transformer which eventually overtax the unit and cause failure.)

Are all transformer cooling apparatus and system flow paths operating properly? (Fans & thermostat(s) operational and set correctly? Cooling media of the correct type, and clean? Are the radiators clear of obstruction and correctly oriented? )

Has the transformer oil been properly and consistently maintained for the life of the unit? (PM interval? Oil filtered regularly on a consistent interval? Oil tested and recorded/documented on a regular consistent interval?)

Is the available power supply to the transformer within OEM specifications? (Voltage stable and within min-max limits? Is the system frequency within specs?)

Is there a high number of VFD or other frequency manipulating devices present in the system? (If so, Has the system been tested to identify any/all harmful harmonics?)

Abnormally high voltage spikes can definitely cause this type of transformer failure.

How old is the unit? (It could well be at the end of life.)

Hope this helps.

Root Cause?

Having to ask an anonymous internet forum for the answer?

I exaggerate when I wrote that I would like to know the root cause ...
Actually what I'd like to know is whether you agree with my assertion that the dark markings are related to the overheated operation (indicated by DGA - see #23), unrelated to the occurrence of electric arc on the X1 terminal region (see #23).
Suppose has been a rapid event, not associated with external causes (external short circuits, for example), so that the time involved in the event can not result in thermal effects (subtransient period).

I could not imagine an electric arc in there. I presumed, when an arc happen, it will consequently burn the surrounding insulation instantly bigger and bigger. It will not reach the point of discoloration, infact, it will result to something more catastrophic than those, I think.

Unless the oil is actively pumped to cause hot oil to exit to out of the bottom of the transformer case, wouldn't the hottest part of the transformer be at the top? I am assuming uniform heating (no shorts) of the transformer. I suspect the dark but not charred regions at the bottom of the images should be there on a cool transformer. The site of the arc agrees with this observation.

I recommend you contact the builder of this transformer and ask for an undamaged image for comparison.

In my opinion the transformer has to be ONAF and was loaded at 9 MVA without ventilation.

The oil was overheated and the lower half of the winding paper was impregnated with oil soot.

Comparing the test results from 2013 with 2014 it seems to me that in the last operation year the oil was degraded by heat.

However, I am not positive this is the cause of short-circuit on phase 3.

"An arc of high energy in the X1 terminal region" said ronaldosalles.

Since 13.8 kV windings are Delta connected there is no grounding connection on low voltage side. That means the exterior equipment or cable was grounded somewhere too.

It could be the grounded part of X1 it is an "old fault" from fabrication or mounting process and was not detected by protection system in time.