What's new
What's new

Need help with machining/heat treat process

Flash Gordon

Aluminum
Joined
Jan 29, 2004
Location
Chicago, IL
I’m machining some complex parts from a forging that gets heat treated after machining. I’m having problems with the parts shrinking after heat treatment. The finished part is roughly 1” x 2” cross section by 14” long. They are shrinking about .017 over the 14” length. I’m sure they are shrinking all over, but it is most apparent over the 14” length. Most dimensions are +/- .020 and are not a problem. However, there are several features that are +/- .002. The parts check good before heat treat, but are under by .017 after heat treat.
I’ve had the forgings annealed and normalized before machining with the hope that it would eliminate any shrinkage during the heat treat process.


Material spec:
Steel forging, ingot or continuous cast, graded D, Spec MIL-S-46172, make from steel, composition 8620, killed, ASTM A304.

Heat treat spec:
Heat treatment per SAE-AMS-2759/7:
A. Austenite in carburizing atmosphere. Case depth .014 to .018 oil or salt quench.
B. Temper 1 hour min to harness RH A80 to 85.


Here are some ideas that were already considered:

1. Leave extra stock on the end of the part and grind to finished length after H.T.
Won’t work since I would be grinding through the case depth and the required
.014 - .018 case depth will not be there anymore. (see H.T. spec)

2. Have the heat treater drive the case depth deeper (i.e. .040 deep) and then grind or machine to finished size after H.T.
Won’t work since I would then need to machine the extra .025 from every
feature of the part. (50 tools and 1 hour cycle time with many slots, holes)

3. Adjust the machining process to account for the shrinkage, leaving some of
the features about .017 longer than the print dimension to account for the shrinkage.
This might work, but it’s not my first choice. I would need to rely on
a consistent amount of shrinkage in every lot of parts from the
heat treater. It would also be a big project to edit the program
to adjust all the dimensions.


Should the parts be shrinking that much during heat treat? Ideally, I'd like do something to the raw forgings that would eliminate any shrinking during H.T. Is there some process like normalizing that would eliminate the shrinkage?
All ideas are appreciated!
 
I'd assume the shrinkage is coming from a transformation from body centered to face centered cubic structure, or vice versa, since one of those structures actually does take up less space than the other one. Shrinkage upon hardening is not unusual on parts made from 4340, and since your 8620 has the same primary Cr-Ni-Mo alloying elements, the shrinkage seems logical. You've got something barely more than .001" per inch shrinkage, which is also well within the range of what I've seen on small alloy parts I've hardened in the past. You just have a lot of inches over which the .001"+ can work its magic :D

I'm wondering if you could have the part taken to the austenitizing temp and quenched as if hardening to change the internal structure of the material prior to machining. Since it's low carbon, the quench isn't going to give any high hardness, but it should give you the more tough characteristics common with the core properties of heat treated 8620, and it should pre-shrink your forgings. You'd still have to have the surface hardening done after machining of course.

I'm certainly no expert, and just know enough to be dangerous, but AFAIK there is no way to prevent the parts from shrinking. All you can do is try to find some way to make them shrink before machining such that they'd still remain stable thru a surface hardening process.

Since your spec mentions salt quench, I'm not sure what processes that might cover, but if it covers something like salt bath nitrocarburizing then that might be a viable alternative for post machining surface hardening since its a relatively low temperature process as compared to the normal gas carburizing temps.
 
Get someone who uses Mastercam to write the program for you. The part can be scaled in X or Y (whichever axis shrinks). He (or She) can write and post-process the program for your cnc. If you run one and find out the program needs to change the scale, it is a simple tweak and then re-post-process to get a new corrected program. Manual program editing is not needed.
 








 
Back
Top