Material Growth During Heat Treat

[PinMan]

Hi everyone. I have a bit of a stumper here, at least it is for me lol. I have thumbed through my Machinist's Handbook and found a few things that will help me out, but not exactly what I'm looking for.

Here's the situation.
I am putting varies size holes in M2 Tool Steel that will be heat treated to 65Rc. My tolerance for the holes are +/- 0.0005". I cannot grind the ID of the hole like I want to, so I need to drill/ream the holes a size that will shrink to the correct size after heat treat.

Im looking for a mat. growth formula or some advice that will help me out.
Thanks in advance

[NC Cams]

We calc shrink all the time and have not found a way to hold shrink to withing 0.0005:.

Did some tool steel cams and ran into a similar issue with a dowel pin. WOrse yet, not only did the hole shrink but it also did so in an "hour galss" fashion as you went deeper into the hole.

SLow careful reaming with carbide solved our fit issue.

Finally, from what we've seen, you can't hold shrink that close from batch to batch with any tool steel we've worked with - batch to batch shrink consistancy just ain't that good.

[PinMan]

Hmm, I figured that I might get an answer similair to yours lol.
I suggested Carbide reaming post heat treat, but that got shot down due to cost. This is a custom, one time job and the suits wont approve an order of 22 separate sized carbide reamers for this job. Not to mention, some of my holes are not standard drill sized holes

On the larger holes, I am using Carbide endmills and doing a pocketmill through the part. I just cant get away with that on the smaller ones.

Guess Im back to square one.

[Geof]

Is the material annealed/normalized before you start?

The 'hour glass' distortion NC Cams refers to could arise from manufacturing stresses varying across the thickness of the material being relaxed during the heating for heat treatment. If these were relaxed before machining by normalizing the distortion could be less and more-or-less even.

But 0.0005" is probably too optimistic. Get 'the suits' to tell you how to do it.

I have lapped hardened drill jig bushings to a larger size by using custom sized brass laps and grinding paste, very slowly I might add. Doing this on 22 holes could be tedious.

[PinMan]

Originally Posted by Geof Get 'the suits' to tell you how to do it.

LOL, thats exactly what I told them.

I am not sure how the heat treating process will go as we send all our stuff out to heat treat.

[JWB_Machining]

While I have no experience with this I am a Mechanical Engineer. The equations you are looking for are the Equations of Thermal Expansion. These equations can be found in any decent Engineering Material Science or Strength of Materials text. You might try calling your local university for help from an engineering professor on the topic if you wish to pursue this but in my opinion NC Cams is right and the equations you're looking for take too many assumptions. It's going to be nearly impossible to hold within +/-0.0005" especially between batches of material.

(Alpha v) is the Volume Coefficient of thermal expansion.
"In many materials the value of (Alpha v) is anisotropic; that is, it depends on the crystallographic direction along which it is measured"

"From an atomic perspective, thermal expansion is reflected by an increase in the average distance between the atoms."

Source for Quotes: Materials Science and Engineering: An Introduction by William D. Callister, Jr. Page 661

Oh I forgot to Add the hour-glass shape is most likely due to a non-uniform rate of cooling, obviously the interior of the hole will cool slower then the rest of the plate giving the atoms nowhere to go on the interior of the hole so they get pushed out to form the hour-glass shape.

[Geof]

Originally Posted by JWB_Machining .....Oh I forgot to Add the hour-glass shape is most likely due to a non-uniform rate of cooling, obviously the interior of the hole will cool slower then the rest of the plate giving the atoms nowhere to go on the interior of the hole so they get pushed out to form the hour-glass shape.

Phrased as humbly as I know how, I think you have it backwards.

It is not clear from previous posts whether these are through holes or blind holes but consider through holes:

Imagine a cross section of the hole and surrounding material heated to the heat treating temperature and soaked to allow the entire section to equilibrate to the same temperature; the entire volume will have expanded the same amount and provided there where no pre-existing unbalanced stresses the hole will be as parallel after expansion as it was before.

Now quench the material: This has to be done from the outside and while the quenching fluid can theoretically enter the hole the Leidenfrost effect will prevent the quenching being very effective here so cooling only travels in from the outer surface of the material. The result is that the outer surface contracts first while the core is still hot and somewhat plastic. The hole remains parallel because the plastic core is pulled in by the contraction of the outer surfaces; what also happens is that the material grows in thickness because the contraction of the surfaces has to be compensated by expansion to keep the volume constant. Now when cooling continues through the bulk of the material the effect is that the core continues to contract but is constrained between already contracted surfaces which are no longer plastic; this results in an unbalanced stress distribution across the thickness with the surfaces subject to compression and the core tension. The tensile stress in the core will tend to pull the sides of the hole out to a barrel shape. Because this tensile stress in the core is mirrored by the compressive stress in the surfaces the ends of the holes will be pulled in slightly making them smaller than prior to heat treatment and actuating the barrel shape.

I think the hour glass shape arise from an opposite effect where pre-existing stresses in the material are relaxed unevenly during the intial heating process.

[PinMan]

Originally Posted by Geof It is not clear from previous posts whether these are through holes or blind holes but consider through holes:

They are through holes. the parts are disks.
OD - 2.2500"
ID - ranging from .1905" to 1.0035"
Thickness - .5000"

[Geof]

Originally Posted by PinMan They are through holes. the parts are disks. OD - 2.2500" ID - ranging from .1905" to 1.0035" Thickness - .5000"

Because not much material is involved my approach would be to experiment a little bit.

Rough machine the parts all over and do the hole leaving maybe .01" on for finishing.

Get them normalized.

Finsh machine everything and get really good measurements for all the dimensions; you can measure to a much higher precision than you can machine so you can find out very precisely what size they are as opposed to what size they should be. For these samples it does not matter even if they are out of tolerance so long as you do know the dimensions.

Now get them hardened and remeasure; this, with luck, gives you the correction factor(s) that you need to use so that they finish up within tolerance after heat treating.

If you make several samples and they all show the need for the same correction factors you can be fairly confident in doing everything at once provided they are all from the same batch of material.

[MrWild]

"Now get them hardened and remeasure; this, with luck, gives you the correction factor(s) that you need to use so that they finish up within tolerance after heat treating."

Provided you don't run out of material and need to order more from a different batch.

[Geof]

Originally Posted by MrWild "Now get them hardened and remeasure; this, with luck, gives you the correction factor(s) that you need to use so that they finish up within tolerance after heat treating." Provided you don't run out of material and need to order more from a different batch.

A good observation but, again with luck, if the parts are normalized before finishing and heat treating there should not be any difference between different batches.

It certainly is an interesting challenge, and you can be pretty sure if things go wrong you can guess who 'the suits' will blame but if things go well you know who will take the credit.

[RICHARD ZASTROW]

Just a thought, would it help to plug the holes prior to heat treat then remove the plugs after?

Dick Z