Heat-treated steel with minimal distortion- recommendations?

ChipSplitter · Jun 15, 2020

I am working on prototyping a part that needs to be case-hardened to 45-60 Rc and .02-.05 deep (not critical).
The most important thing is minimizing post-HT distortion.
It also needs to be readily machinable in the annealed state.
We use 12L14 on some applications, but it distorts about .002 for every 1/4" of thickness.
The first alloys that came to my mind were 4140 and 8620.
Any suggestions on material to use?

Edit: I also thought about A-2, O-1, and S-1. 8620 and 4140 are definitely the cheapest......the part is made from 4" round.

Dan from Oakland · Jun 15, 2020

8620 is pretty stable in heat treat- even more so if you stress relieve / anneal a semi finished part. 4140 is a through hardening steel and not quite as stable as 8620, but again, a pre finish stress relief will minimize final heat treat distortion. If you are only making a few, consider S7 tool steel- its air hardening and very stable.

Terry Keeley · Jun 15, 2020

Just made a few parts from S7. Specs say it grows about 0.001" per 1" of material, I found that pretty close. They stayed dead straight.

I hate to admit it here (cough), but I did it at home with my ceramics kiln and tempered in the kitchen oven. Wrapped the parts in ss foil, air quench, temper, I went for 55 Rc.

Easy peasy. Except for the bag that opened up, I triple fold them now.

One that the foil opened up:

Garwood · Jun 15, 2020

I have a fair amount of 1018 parts case hardened and they mostly come out as flat as they go in. About .120" thick and 3.5" long. I go to the low end of depth, around .02" depth of case. Heat treaters told me less soak time usually translates into less distortion.

sfriedberg · Jun 15, 2020

Garwood said:
Heat treaters told me less soak time usually translates into less distortion.

Down to a point, yes. The soak has to be long enough for internal temperature to equalize if you want to avoid banana-shaped parts on quench.

ChipSplitter · Jun 15, 2020

So how much distortion would you expect from a .75" thick by 4" dia. disc from 8620?

We need to press in 3/16" dowels after HT so distortion must be <.001.

greggv · Jun 15, 2020

normally, material has thru holes are drilled, material is heat treated, ground, and then wire cut.

Scott314 · Jun 15, 2020

ChipSplitter said:
So how much distortion would you expect from a .75" thick by 4" dia. disc from 8620?

We need to press in 3/16" dowels after HT so distortion must be <.001.

Is that <.001 distortion in flatness of the disc, or distortion in the dowel pin holes that you're worried about? If it's the dowel pin holes, I would be surprised if the holes grew or shrank more than a couple tenths. Not sure about your initial distortion numbers, .002" per 1/4" thick? Do you mean distortion as in the part is turning into a potato chip, or do you mean growth/shrinkage factor?

How cost sensitive is it? And how much does machinability matter? There's a fairly big difference going from 12L14 to 4140 to A2. What about 1045 for hardenability, machinability, and low cost?

What about heat treating something on the low end of your hardness range and doing some post-HT machining? What about some other surface hardening process, like ion nitriding?

Just to throw a few things out there.

ChipSplitter · Jun 16, 2020

Scott314 said:
Is that <.001 distortion in flatness of the disc, or distortion in the dowel pin holes that you're worried about? If it's the dowel pin holes, I would be surprised if the holes grew or shrank more than a couple tenths. Not sure about your initial distortion numbers, .002" per 1/4" thick? Do you mean distortion as in the part is turning into a potato chip, or do you mean growth/shrinkage factor?

Yes, I meant the dowel holes. Flatness is an issue too, just not to .001......if I can hold .01 I'm happy.

How cost sensitive is it? And how much does machinability matter? There's a fairly big difference going from 12L14 to 4140 to A2. What about 1045 for hardenability, machinability, and low cost? What about heat treating something on the low end of your hardness range and doing some post-HT machining? What about some other surface hardening process, like ion nitriding?

We do a bunch of 12L14 already. And yes, it is easy street.

Machinability isn't as much of a concern to me as heat-treat stability. I'm a machinist, not a metallurgist......

Cost is a real concern. Not a deal breaker if there is a post-HT op but it would sure be nice not to.......

How is 1045 for stability? We machine some here and it's not too bad.

Mike1974 · Jun 16, 2020

I would not consider 4140. We used that a few times (larger peices) and found it to move and warp all over. A2 and D2 are fairly stable, but D2 machines/grinds like crap at 58+Rc IME. S7 is pretty good in stability, and price wise (of course been a while so...) should not break the bank. Don't know about 8620...

Mike1974 · Jun 16, 2020

ChipSplitter said:
Yes, I meant the dowel holes. Flatness is an issue too, just not to .001......if I can hold .01 I'm happy.

We do a bunch of 12L14 already. And yes, it is easy street. Machinability isn't as much of a concern to me as heat-treat stability. I'm a machinist, not a metallurgist......
Cost is a real concern. Not a deal breaker if there is a post-HT op but it would sure be nice not to.......

How is 1045 for stability? We machine some here and it's not too bad.

I would wire the dowels after heat treat like someone else said. I remember the "good old days" when we ground our pieces into the dowels, then we were enlightened

with wire edm and wow what a relief!! :drool5:

Dan from Oakland · Jun 16, 2020

The 8620 will stay reasonably flat- certainly better than probably .005". The 3/16 holes will most likely shrink slightly. Centerbores of any size will almost always shrink to some extent. If the dowel pin holes go all the way through and you can live with about 35Rc in and around the holes, put SS screws/nuts with washers on both sides and it will mask the carburizing in those areas and you can ream the dowel pin holes after HT. We put SS bolts in threaded holes in 8620 parts prior to HT to keep the threads from getting brittle and it works pretty well.

thermite · Jun 16, 2020

Mike1974 said:
I would not consider 4140. We used that a few times (larger peices) and found it to move and warp all over. A2 and D2 are fairly stable, but D2 machines/grinds like crap at 58+Rc IME. S7 is pretty good in stability, and price wise (of course been a while so...) should not break the bank. Don't know about 8620...

8620 is beloved here because it give yah lots of CHOICES. Most of what 4XXX has, plus even "fair" weldability if need be.

I ain't on "commission", but it isn't hard to look all of these goods up online.

Prolly a six-inch stack of "quality paper" John Wiley, McGraw-Hill hardbounds up in the attic on all that but..

.. online - the already-published pages, not just "PM" - is a faster and more "current" source for metallurgy, machinablity, and the various heat and surfacing treatments.

AISI 8620 Steel, single quenched and tempered (230degC (450degF)), carburized

Note that it does not need a Collitch degree to get use out of:

AISI 8620 Steel | 1.6523 | 21NiCrMo2 | SNCM220 - Otai Special Steel

..because...a(ny) Machinist - and MOST "Engineers" - expect to UTILIZE a given alloy.

Not invent a new one from scratch.

2CW... of "ordnance steel". Talk about "well proven technology"?

ChipSplitter · Jun 16, 2020

thermite said:
8620 is beloved here because it give yah lots of CHOICES. even "fair" weldability if need be.

I ain't on "commission", but it isn't hard to look all of these goods up online.

Prolly a six-inch stack of "quality paper" John Wiley, McGraw-Hill hardbounds up in the attic on all that but..

.. online - the already-published pages, not just "PM" - is a faster and more "current" source for metallurgy, machinablity, and the various heat and surfacing treatments.

2CW... of "ordnance steel". Talk about "well proven technology"?

I have a 1946 Machinery's Handbook. Interesting to read, but not very, *ahem*....up to date.....

I admire any metallurgist that knows his stuff. They're like God almost. :willy_nilly:

thermite · Jun 16, 2020

ChipSplitter said:
I have a 1946 Machinery's Handbook. Interesting to read, but not very, *ahem*....up to date.....

I admire any metallurgist that knows his stuff. They're like God almost.

Some stuff has simply not CHANGED. By that date, for example, 8620 had been proven in millions of weapons - and-not-only - at MORE THAN just the First and Second World War ordnancing.

My oldest dead-tree books on all that are from the 1890's through about 1926 - Dad's collich books. Then my own - 1960's and a bit.

Other books? Well ... some spent their pocket money in bars and brothels, others in second-hand bookstores!

Yah.. I remember hanging on every word of our Company's "on retainer" PhD RPE Metallurgist when he'd visit. He'd started out makin' black Iron stove bolts. War Two had switched his company to making Oldsmobile auto-loading cannon barrels. After the war ended, he went to Collitch to learn more about what mysteries had bustid his chops. Got addicted to learning.

I figured by the time I met him, he knew several ways to heat-treat even sunlight or moonbeams.

No youngster, he. Grandest dam' "war stories" - usually about folks f**king-up their metals choices.. then gittin' big surprises over stuff they COULD HAVE known in advance "if only".. etc.

Sure helped keep me outta trouble... At least learnt I hadda go RESEARCH stuff.

Not "ass sue-me".

Modelman · Jun 17, 2020

Dowel hole size isn't going to change much, after all these are to be press fit, right? But dowel hole spacing will... if there are two holes on 4" centers they will likely end up 4.002 - 4.005. One way, If wire EDM is too expensive is to do a run of samples with holes varying in .001 increments, mark them, see which is usable after heat treat, then make the run of parts to that dimension.
Dennis

thermite · Jun 17, 2020

Modelman said:
Dowel hole size isn't going to change much, after all these are to be press fit, right? But dowel hole spacing will... if there are two holes on 4" centers they will likely end up 4.002 - 4.005. One way, If wire EDM is too expensive is to do a run of samples with holes varying in .001 increments, mark them, see which is usable after heat treat, then make the run of parts to that dimension.
Dennis

^^^ THIS!!! ^^^

Or two thou. Most of the time. Even when you think you already know the answer.
Just cheat. And keep notes.

Small ration of steel and time is cheap. So are notebooks. Or notes kept with drawings or files.

Bigger ration of steel plus time wasted re-doing f**ked-up production runs?
Not so cheap.

Search

Largest Manufacturing Technology Community on the Web

Search

Heat-treated steel with minimal distortion- recommendations?

ChipSplitter

Titanium

Dan from Oakland

Titanium

Terry Keeley

Titanium

Garwood

Diamond

sfriedberg

Diamond

ChipSplitter

Titanium

greggv

Hot Rolled

Scott314

Plastic

ChipSplitter

Titanium

Mike1974

Diamond

Mike1974

Diamond

Dan from Oakland

Titanium

thermite

Diamond

ChipSplitter

Titanium

thermite

Diamond

Modelman

Titanium

thermite

Diamond

Similar threads