F50TF50 bond coat

[Harvey]

Any body familiar with APS spraying of this GE TBC coating spec? I can achieve the coating standard ok but some add. requirements mean I now have to produce the same coating but with <2% porosity and low levels of unmelts in the bondcoat. I am using the F4 torch. Does any one have any suggestions on powder/parameters/techniques to meet this?

Many thanks

Harvey[/font]

[Gordon]

Hi Harvey

to the Surface Engineering Forum.

I'm sure we have quite a few members with practical experience with GE F50TF50 bond coats, so I hope they can chip in

I take it that you are having problems meeting the additional requirements. What is your present set-up on powder/parameters/techniques? Sorry, got ask, are you confident in your metallography and its interpretation? If this is not right or consistent then you could be chasing your tail

[Harvey]

Hi Gordon,

Thanks for the reply. You're absolutely right about the metallography side of things - we are acutely aware of this potential pitfall but on this occassion I am confident we are ok, besides, one of the issues is concerned with high levels of unmelts which is difficult to change by "bad" metallography - if anything you will improve this aspect rather than make worse.

For those who are familiar we are trying to achieve porosity/voids better than V1 of GE photostandard 8603003. This equates to approaching 1 %, which from an APS coating I think is unrealistic. The unmelts have to be better than 5 per FOV (V3 of 9008007) which is not such an unrealistic target, however, as GE will only allow 5% of FOVs (at 200x mag) to be non conforming it only takes 2 FOVs out of typically 25 & then you have a non conforming sample.

We have tried 4 different recognised powder suppliers (Sulzer Metco, HC Starck,PAC & Praxair) all conforming to the B50TF192 clA spec and all with much the same results. We are using the F4 gun with SM recommended parameters (65 Ar, 14 H2 & 600A) + & trialled about 20 alternative sets, including altering powder injection through gas flows & injector position, primary & seconday gas, current, but with no real impact on the overall result. We have also tried many different spray techniques using both rotation & wall scanning techniques - literally hundreds of combinations. The results aren't disastrous just not meeting what seems to be an overly stringent spec.

So I guess the question is as much have GE got this wrong? Or does anyone have similar experience & know some special trick to resolve the problem.

Any information/suggestions gratefully received!

Many thanks in anticipation,

Harvey

[Gordon]

Hi Harvey

Sound of silence well I suppose understandable in the light that any company spending time, effort and cost on development may not want to give up their secrets.

The specification may be unrealistically tight, but I assume some are managing.

I suppose you are at the point of wondering whether further development is going to be worthwhile and cost effective for this job. Well if you are going to persevere, I can only make a few comments, which probably you already appreciate from your trials anyway:

Powders - I know you are restricted to a few approved powders only, but choosing the one with the tightest particle size distribution and maybe one with lower high end particle size amount may help.

Powder injection - very critical. Injecting powder in through the side of the plasma jet is quite complex. In an ideal world the powder particles would be all exactly the same size and shape, but they are not. The plasma jet has a classifying effect - smaller particles penetrate less than the larger. Larger particles with more momentum may pass right through the plasma jet too quickly, not picking up enough heat and velocity, some ending up as unmelts in the coating. Setting up the injection parameters is really getting the best balance in conditions for the majority of the particles, inevitably that means some particles may be under or over-heated, lack velocity etc. and end up effecting the quality of the coating. Important to remember, when changing plasma conditions, changes to powder injection may be necessary also.

Powder feed rate - It may be worth trying a lower feed rate than recommended, sometimes quality gains can be made at the expense of production through-put.

Plasma parameters - again very critical. Getting right balance of plasma gas flows, plasma gas mix and current. Generally, if unmelts are excessive (assuming powder, injection and feed rate are optimal) then the larger particles need more heat. This can be achieved by lowering the plasma velocity to get more dwell time for particles and/or by increasing heating potential of plasma. This again is complex and not as straight forward as you may first think.

Without going into other parameter considerations, you can see how complex development gets with a mind boggling number of possible parameter permutations.

[Harvey]

Hi Gordon,

Thanks for the input - it is appreciated but unfortunately all tried & tested without success - we're are close however - but I just think the spec is unrealistic to achieve consistently on a production basis.

We will keep persevering!!

Thanks again

Harvey

[dcrawford]

Harvey,
As you conducted your trial sets with various parameters/powders were you using any particle diagnostic equipment (temperature, velocity, or injection imaging)? Did you go through some sort of procedure to optimize your particle injection?

Dan

[derek]

You tried so many different ways to get the perfect coating but have you ever think about the lab result is perfect or not. Some times, the different praperation procedure can get diffenrent result. For bound coat, the smaller load will get more porosity during polishing.