Hey guys,

I am looking for a way to reduce/eliminate interface separation in plasma sprayed Cu-Ni-In (B50TF72A I believe...) when applied to a stainless steel substrate (or any other in general). Aside from the separation the coating conforms metallographically to the spec I'm evaluating against. The separation is apparent at 200x.

I've heard from my plant manager this coating tends to do this a lot when ground/polished.

Any help would be appreciated. I'm afraid I can't go into too many details due to my companys proprietary nature.

Thanks!

Hi Donald

Is this problem just with metallographic evaluation of test pieces or does it effect the actual job with coating edge separation as well?

If it is a metallogaraphic preparation induced feature, I can only suggest that extreme care is taken on cutting-off, epoxy vacuum impregnation and cold mounting in epoxy resin (as opposed to hot/pressure mounting in bakelite or using high shrink cold resins). With cutting and grinding always try and keep coating in compression against substrate.

Gordon Wrote:Hi Donald

Is this problem just with metallographic evaluation of test pieces or does it effect the actual job with coating edge separation as well?

If it is a metallogaraphic preparation induced feature, I can only suggest that extreme care is taken on cutting-off, epoxy vacuum impregnation and cold mounting in epoxy resin (as opposed to hot/pressure mounting in bakelite or using high shrink cold resins). With cutting and grinding always try and keep coating in compression against substrate.

Just with metallographic evaluation. The actual as sprayed coating looks fantastic on the parts we process (which we can only process after a conforming lab, etc.)

We have a frozen process that takes all the above factors into play and the coating is always ground/polished in compression, so I am not exactly sure why it is this particular coating that tends to do it.

I appreciate your quick response!

Hi Donald

I know some coatings are prone to this effect, even ones with relatively high bond strengths. I take it that the separation is seen running from the ends/edges of the samples. I know some specifications allow to some extent for this effect. I assume no bond coat is involved?
Do you bevel/chamfer the edges of your test pieces? Try to make sure your test piece is seeing the same conditions as the job, particularly temperature wise.

The company I work for was also having this problem, what I started to do was to pre-cut the test pc and also bevel the edges. You may also want to make sure that when you are pre-heating the test pc gets the same amout of heat as the parts to be coated. Pre-heating the substrates is very important with this coating. Also, depending on the OEM you are doing this job for, you may be able to ignore .010 from the edges of your sample.


Erick

Good post Erick.

Quote:what I started to do was to pre-cut the test pc and also bevel the edges

Could you expand on pre-cutting test piece? does this enable you to avoid using the cutting-off part of the metallographic preparation process? I suspect that cutting-off could be a prime contender for initiating this problem.

Pre-cutting a sample is simply just do as much cutting as possible before coating the sample. This reduces the chances of mechanical damage due to cutting. After coating the piece I put the sample in a lab cut-off saw(use wood or other soft material in the jaws of the sample vise to avoid damage to coating) and cut the small remaining material. Make sure you are using some sort of cutting fluid to avoid excess heat and cut towards the coating(keep the coating in compression). Clean and dry, then you are ready for cold mount.

Hi Donald, gordon and everyone,
How did you control cooling air during spraying.
I think cooling air that is very important, because this coating is very special and apt to cause internal stress. It will cause the coating to separate, when you cut the sample.

So control your cooling air and increase surface roughness.

best regards,

William