Residual Stress / Relieving in 1045 HR round bar

[tr1ntx]

I searched but didn't find anything real close.

For a ball valve ball (see attached drawing) made from 1045 HR round bar, 12" dia with a 4.5" bore by supplier, there is apparently distortion caused by the relieving of residual stress after the material is milled away on top & bottom to create the trunnion pins.

I would have thought the the HR material would not have had very high residual stresses, but apparently I'm wrong. Any input or other thoughts? Anyone have any input on how to deal with this?

The diameter tolerance is +/- .003 The distortion shows up as reduced diameter at the thinnest portion (at the top & bottom on the bore), .006 undersized.

 

[tr1ntx]

Oops! I forgot to click to insert the drawing.

 

[metman]

I agree with your analysis except that 0.0005"/inch does not seem like high residual stress. Fortunately the bore is undersize and you should be able to rebore it but likely will need to use a 4 jaw chuck to hold concentricity and to be able to clean up the bore.

I suggest that the procedure should be to bore undersize by 0.015" dia, mill the trunnions, and final bore to size.

 

[metman]

The residual stress is obviously introduced from the milling op since it is localized in that orientation. Therefore bore undersize 0.015", mill trunnions, and then final bore to size.

 

[tr1ntx]

I didn't explain it well enough . . . the problem is the spherical diameter of the ball, it's out-of-round, undersized at the top & bottom.

The bore is still within tolerance and is not that critical in the first place.

When making balls out of forged mat'l, we mill in steps. It surprised me to see that much distortion from the HR mat'l, so I thought maybe I have erroneous thinking or there's something going on I don't realize. As you said, the relief/size is not really very large, so I guess I wasn't properly considering scale.

Another point I'm wondering about is the time factor. Apparently the ball passed inspection and then after several weeks/months storage, it was used. I don't know if the inspection was thorough enough to have caught the out-of-roundness (probably not), or if the HR mat'l relieves very slowly. That brings up the question of continuation of the relieving. I was hoping some of the expert metallurgists could comment. I had thoughts about the cooling rate of the large section after hot forming causing residual stress, etc.

 

[metman]

tr1ntx
Yes I believe time could be a factor.

If the housing for the ball can be match-machined then the ball could be saved by machining the spherical OD to clean up and size the housing accordingly because the ball OD is undersize. But first you probably should thermally stress relieve the ball.

What should the stress relief process be you will ask? It depends upon the condition of the ball. Was it normalized, quenched and tempered, annealed, as-rolled? If it was Q&T, I would SR at 75deg F below the tempering temperature for 12 hrs allowing + or - 25deg.

Suggested future process would be to machine OD oversize, bore undersize, mill trunnions (but not necesdarily in that order), thermally stress relieve, finish machine.

If it is not Q&T then I would guess 700-900F stress relief but don't have my metals handbook handy. Someone on this forum more expert than me will answer this question as needed if you will give the condition of the 1045.

 

[tr1ntx]

The 1045HR is as-rolled.

The problem is that the ball is now out-of-round, so no match machining can be done to adapt to it. The surface which seals against it is a conical section and thus it doesn't make contact with the undersized portion of the spherical dia.

Additionally, I wasn't thinking about this before, but the ball is Electroless Nickel Plated, and thus baked at 750F for 1 hr.

So that makes me worry that if we employ thermal stress relief and then are able to machine a nice round ball, are we going to get distortion from the ENP bake because of the much thicker sides of the ball compared to the thin sections at the bore.

Or even if there is thermal distortion, how would its magnitude compare to that of the residual stress relaxation?

 

[metman]

Selective plating, as I know it, is not electryless yet I do not know why it wuld not form a metallurgical bond to the existing ENP.

This could be a vialble repair method because the plating can be locally placed in precise thickness.