Problem- Shrinkage in heat treat

[vanm]

I have run into something that seems to defy common reasoning when it comes to heat treating.
I am running and ongoing production job using 1050C spring steel. On one batch of 4000 pc, the hole to hole dimension shrank about .006. This is on a
.05 thick part. During the CNC process all dimensions are within spec, but after and outside heat treatment they were all small. Has anyone ever heard of this happening? Conventional wisdom say's they will grow if anything.

[RICHARD ZASTROW]

In most hardening processes on steel, the metal grows slightly. Holes shrink, OD's expand.

That's why they make over sized taps to compensate in advance for hole shrinkage.

We also grind precision gears after hardening the teeth. They distort like winding and unwinding as well.

It's a bit of an art to predict exactly how and how much change will happen.

Dick Z

[juergenwt]

vanm - For low carbon steels (carbon added) allow for an increase in HT. Holes will shrink.
For spring steel and for tool steel allow for shrinkage (no carbon added). Holes will open up.
High alloy tool steels like D-2 will move next to nothing.

[vanm]

I'm having problems with the centerline of the holes shrinking toward each other . There is also a hole to edge dimension that is shrinking. Normally that would grow. I'm having to come up with a plausable explaination before I go to see the VP and ask for money for a scrap and remake. "I don't now why" doesn't fly to well.

[mazaker]

heat treat first then drill or mill

[Jonsmachine]

Originally Posted by mazaker heat treat first then drill or mill

Easier said than done

[ctate2000]

Questions to consider:

1. .050 thick, did the part curl and pull the hole center locations?
2. Are the holes round?
3. Is there something different about the h/t process?
4. Did you use the correct material (request certs from supplier)
5. Did any of the other dimensions change? What direction relative to the problem dimension?
6. Is the part hardness correct?
7. Were they tempered properly?
8. Were they correct before h/t?
9. How many different ways can you measure them? Do you get consistant results?
10. How many people have measured them? Do they get consistant results?
11. Is the edge you are measuring to square?
12. Is there a burr effecting the measurement?

Good luck with the boss.
CT

[roninB4]

It's fairly standard procedure to have a finishing operation AFTER heat treating, material is purposely left on features for that reason. If your tolerances have been ruined by the heat treating then you either have to open the tolerances or you're attempting to save money by skipping a process. Sure, we'd all like to go right to final size and then send it out for HT but it doesn't work that way. Either change the tolerance, the material, or the process. If the VP doesn't understand that then he doesn't know much about machining.

[roninB4]

Originally Posted by Jonsmachine Easier said than done

That's what grinding wheels/CBN are made for.

[vanm]

CTA2000, beleive me when I say that I have gone through that entire self questioning process. The thing that makes me believe it is an incorrect heat treat process, the features on the stamped part that I didn't machine also reacted the same way, ie shrinking toward each other. I have gotten off the hook this time. As for machining hard, I am thinking about that. Problem is if the stamped features are going out of wack. my milled features won't trigger a reject. Then the person you don't want to find an error may, the customer. Not a good thing. Thanks for all the advice.

[ctate2000]

Stamped part? 4000 pieces, any chance the stamping tools are dull? This would create more internal stresses in any material. The h/t process would naturally relieve those stresses possibly allowing shrikage.

[vanm]

That's something that I didn't think of. I'll bring that up with the press room supervisor, see if I can have the parts maintained on a more frequent cycle. We run hundreds of thousands of these things, so they might be getting lax on the sharpening.