thread124-450268 Very Strange Internal Rupture Disc Knife Blade Cracks

[Bambie]

With reference to thread124-450268, LP examination of the 17-7PH or S301 knife blade ends has revealed unusual mind-plane cracking (see pictures on word document attached).

Is it possible that surface hardening heat treatment created layers having different Modulus of Elasticity (E) and thermal coefficient of expansion (Z)?

It seems reasonable that the blade sides and ends would have higher E and lower Z than the core or mid-plane layer.

When the blade cools, the mid-plane layer should contract more than the side layers and be left in a state of tension, especially along the longest dimension – the blade axis.

The blade sides were then fillet welded to the plain carbon steel blade holder.

In-service superheated steam would cause the holder to expand further than the cooler blades, increasing tension in the blades.

Thermal cycling would eventually initiate cracks in the blade welded ends at the mid-plane to outer layer interface.

Does this simplistic model explain the mid-plane crack orientation?

Would the fillet welds have modified the blade metallurgical properties in a way that would produce the unusual mid-plane crack orientation?

 

[EdStainless]

Have you ruled out mid-plane inclusions in the base material?
What about hardness uniformity across the thickness of the material?
I can think of a few simpler reasons for this.

= = = = = = = = = = = = = = = = = = = =
P.E. Metallurgy, consulting work welcomed

 

[gerhardl]

Interesting!

As I read this and the previous thread you seem to have given yourself an impossible task!

The obvious solution is to 'go back to start' and find a better suited type of RD (probably cheapest for this dimension).

The alternative is another type of safety device, at probably higher cost.

I do not think you will find the solution by using your excisting construction with altering materials, fabrication methode or isolating/alternatie mounting methodes.

In my mind the obious reason for failure here is elongation by heating and (perhaps) uneven cooling/shrinking when temperature goes down. Perhaps also increased by corrosive condensation.

One technical solution could be to use thinner, more elastic fastened/weaker knifeblades. In stead of fighting buckling, construct to allow buckling.

And: What is the rason for the knifeblade thickness? It seems very heavy in proportion to needed shearing forces, and seems to have no other construction purpose than in addition possibly protect from 'fall through' of objects.

Good luck!

 

[Compositepro]

Is it possible that the blades are forge-welded from two strips of steel, one on each side of a harder alloy that forms the cutting edge?

 

[Bambie]

EdStainless,

No inclusions were found in the section. Chemical and surface hardness test results are attached.
Unfortunately, further testing is not possible since the blades have been welded back into the holder.
Is there a way to re-direct this post to the attention of Metallurgy?

gerhardl,

The proposed repair is to provide minimum blade preheat (250 degF) and in-service preheat (480 degF) to the holder while the fillets are being welded. This should minimize the tensile loads during operation based on the holder and blade coefficients of expansion, assuming the blades operate at approximately 150 degF.

Compositepro,

The blades are chemically homogeneous with alloying agents similar to 17-7 PH or S301 stainless steels that have been heat treated to increase hardness (see attachment).

 

[EdStainless]

So you don't know if there are inclusions in the centerline since you didn't section a sample, and likewise you have no idea how uniform the internal hardness is.
Do you know what the surface treatment is or how it was done?
Being in plane I am not sure that this crack poses much of a risk.

= = = = = = = = = = = = = = = = = = = =
P.E. Metallurgy, consulting work welcomed

 

[Bambie]

EdStainless,

The manufacturer would like very much to do the repair/replacement, consequently is not forthcoming with heat treatment or other information.
The blades were cut off just inboard of the fillet welds, and LP examination applied to these sections.
I agree this crack orientation may not affect bending strength or blade function.
I am looking for a reasonable explanation for the mid-plane cracking.

 

[don1980]

I think knife blade rupture disk are inherently unsafe, and I wouldn't use them. Honestly i don't understand why they are still being manufactured. I would remove the disk and use another design.