"Coating failure"

[TurbineRepair]

Hi All,

I was wondering if anyone could offer an opinion on a recent "coating failure" we experienced. We sprayed 450 and 54 NS via our 3MB onto a fairly large internal diameter engine casing. Both coating looked very good under the microscope (of course? )and both past a bond strength test with using coupons attached to the job itself.
The trouble is after the case halves where joined together and machining underway the coating at the joint lines failed. Numerous areas of delamination appeared along both joint lines but no where else.
I placed the quotation marks around the phase "coating failure" as I don't believe that the failure was down to poor spraying. On the contrary I believe that because the edges of the cases were not properly cleaned up prior to assembly this lead to considerable internal stresses being generated in the coating as the cases were bolted tightly together. These residual internal stresses combined with the additional stress applied via the cutting tip caused the coating to fail at the weakest point, ie the substrate/bond coat layer.

For me this theory would explain why there is no delamination in any other areas at all and the failure is limited to about 2� max from the case joint line. Had the cases not been prepared correctly or the temperature been to great I would expect to see coating flaking off all over the place, not confined to such a relatively small area.

If anyone has any comments on my theory I would be pleased to hear them.

Jim

PS I can't seem to append attachments for some reason

[LEN WOOD]

Hi Jim,

I believe your suspicions are totally correct. I've seen this so many times before with SM 54NS and 601NS. When spraying half casings it is always worth the investment of a little careful dressing before letting the machinist loose on them! Just take the coating height down until there's no way it can be edge stressed and induced to delaminate.

Cheers

[Gordon]

Hi Jim

I placed the quotation marks around the phase "coating failure" as I don't believe that the failure was down to poor spraying. On the contrary I believe that because the edges of the cases were not properly cleaned up prior to assembly this lead to considerable internal stresses being generated in the coating as the cases were bolted tightly together. These residual internal stresses combined with the additional stress applied via the cutting tip caused the coating to fail at the weakest point, ie the substrate/bond coat layer.

Yes, I think this could well be the cause of your problem.

For me this theory would explain why there is no delamination in any other areas at all and the failure is limited to about 2� max from the case joint line.

Yes.

Had the cases not been prepared correctly or the temperature been to great I would expect to see coating flaking off all over the place, not confined to such a relatively small area.

Not in total agreement here. If problem is not a localised one (like say an isolated area of poor preparation/contamination), then I would expect a coating problem to first expose itself at the weakest point, that being a coating edge.

PS I can't seem to append attachments for some reason

Can you describe this problem in more detail - any error messages, file type and size etc. to help me find the cause From my end it appears to work fine. In the mean time you could email attachment to me and I will insert it into your post for you.

[TurbineRepair]

Hi Guys,

Thank you for your responses and reassuances! I've been writting a report on this failure all afternoon and looking in more detail at the problem I'm certain that this is the cause. I see what you are saying about the coatings failing at the weakest point but everywhere else the coating is top notch.

I find it very annoying when people blame the plasma when the error lies else where, especially when the boss is told the plasma as fell off..... again. Still a nice unbiased report will put them right

Thanks again


I've suffered the attachment problem before, I choose browse, select my jpeg and click add attachment but nothing happens. The file is quiet big 2.5mb. Do i need to shrink it (HOW?)? I'll email you the picture Gordon if you wouldn't mind posting them?

Jim

[Gordon]

Hi Jim

I've suffered the attachment problem before, I choose browse, select my jpeg and click add attachment but nothing happens. The file is quiet big 2.5mb. Do i need to shrink it (HOW?)? I'll email you the picture Gordon if you wouldn't mind posting them?

Thanks for the info - In this case it is down to a SEF forum limit of 1MB Jpeg file size, though you should have got an error message saying the file was too big. You will need to resize image in some image editing software, ideally to the smallest size possible without compromising too much view-ability and quality - a few 100 KB's in preference to a few MB's helps the forum from slowing down while reducing bandwidth cost

If resizing the image is not easy for you ie lack of editing software on your works computer - no worries, just email image to me and will sort it

[TurbineRepair]

Gordon,

I've resized the pictures.... fingers crossed....

It seems to have added them but it wasn't happy. It threw up a couple of error messages (Fatal error: Allowed memory size of 262144 bytes exhausted (tried to allocate 38 bytes) in /home/gordonengluk/public_html/sef/inc/plugins/seo.php on line 1248) but the pics seem to have loaded this time.

Jim

[LEN WOOD]

Visuals are great!

Was the substrate clean of contaminants and old coatings before preparation? Can you see evidence of anything? I am assuming the top coat remained attached to the bond coat when it became detached?

Was this blasted and/or sprayed in assembly - if so was it spaced (packed out) on the abutment faces? If the abutment faces were masked (before blast/spray) then it's good practice to make sure that it is a little back from the face.

The lost coating in the left hand shot looks quite severe and I believe more areas will come off. There isn't much of an area to coat between the edge and the vane insert holes. Do the blanks (plugs, masks) for this unmasked area protrude much? Are they made of Silicon and was it the first use? Was the 'demasker' careful? Was the Media fresh and was the blasting adequate. Was the blast pressure correct and was the part blown down with clean shop air (or better) before spraying?

There are many minor factors that can contribute to a major coating failure.

Some edges were definitely crashed together in assembly that would be further stressed in machining. However I believe that these areas may have been "under par".

I would oversee the rerun!

GB

[Gordon]

Hi Jim

I've resized the pictures.... fingers crossed....

It seems to have added them but it wasn't happy. It threw up a couple of error messages (Fatal error: Allowed memory size of 262144 bytes exhausted (tried to allocate 38 bytes) in /home/gordonengluk/public_html/sef/inc/plugins/seo.php on line 1248) but the pics seem to have loaded this time.

Well done I was about to insert them for you, but see you have already managed yourself. Thanks for noting the error message - I think there is a problem with an old forum plugin misbehaving. Unfortunately, not an easy fix, but one I will be attempting soon. See attachment-problems-t-867.html

As to the coating problem I will echo Glens comments, particularly overseeing rerun including masking, dressing and machining etc.

[Alexangel1226]

Jim,

very similar problem we faced where coating flake only at the edges but mine was on the flanges area. Common causes to it:
1. old coating exist ~ 80% of the time, - try grit blast and observe you should notice different color or tone of layers, try locally blend or bench it slighly, grit blast until u totally remove the previous coating.

2. Demaskign induced stress- if u are using a shim to separate between the to end joints, it's probably overhaung coating stuck on at the side of the shim, iniprocess removal of build up coating can reduce this problem. Just use a scrap and remove when u are checking the thickness of visual during in process.

3. Stress build up at edges, as u mentioned- try get the edges shaffered slightly

4. coating thickness exceed the limit (10% of the time)- Do not know what coating you are spraying and the process u use, APS or HVOF. some thickness limit shd be followed, you shd be fine.

5. Just personal thought, how critical is the groove or line separating the joints? I mean if it is not impacting the next process, be it assembly where the groove act as a locater or some datum, I would explore doing away the masking process and spray all surface. This will solve your problem, reduce process cycle time, consumables, and alot more trouble. make so much of business sense, from the lean point of view, do explore this possibility, if customer is ok and engineering justfication is obtained, why not consider this approach.

Just my suggestion.
hope this helps.

Regards,
Alexangel