"Tim Wescott" wrote in message
...
I kinda asked this question once before, and got a bunch of answers that
all seemed off the point. So I'm going to try again, and (hopefully)
make my question specific enough this time.

Everybody tells me "use 12L14, it's really easy to machine". I'm willing
to accept that -- I'm using steel that I get surplus, and there's a wide
range of machinabilities that I'm experiencing; I can certainly see how
using the _right_ steel might make a significant difference. But:

If I take some annealed or normalized 1065 steel (or other 10"something
big" steel), get it to the shape I want, and properly heat treat it, it's
going to be a _lot_ harder than when I started. So assuming that I can
machine it in the first place, then get it to hold its shape (or correct
its shape after the fact) and not crack in the heat treat process, I'm
way ahead.

On the other hand, if I take 12L14, get it to the shape I want, and do
the _very best heat treat in the world_, then -- because of the low
carbon content -- it's not going to be much harder. Right?

What about 1144 or 41L40? Do you get much by heat treating these? How
easy are they to machine compared to 12L14? Why isn't there a 12L50?

I've heard tell that 1144 isn't as strong as other steels of similar
carbon content -- true? What about the mechanical properties of 41L40
after heat treat?

I'm basically looking for a steel that'll be suitable for model airplane
crank shafts. The advice that I get from the hobby builders of model
airplane engines is "use 12L14 and don't crash". I'm not going to
discount this -- but it'd be nice to be able to do something on par with
what I buy at the store. So, basically, something that's easy to
machine, but that'll harden up to something between a grade 5 bolt and a
grade 8 bolt.

--
www.wescottdesign.com


These are really good questions, and I wish I had a specific answer, but I
haven't machined some of those grades and I'm a little confused about the
properties you want.

If "everybody" is getting away with 12L14, at least in normal flight, they
must not need a lot of strength or hardness. They'll get decent toughness
with that grade, at the expense of a low yield point. In other words, it may
not break in a crash, but it will bend, right?

It sounds like you want more strength and toughness. If so, 4140 is a good
choice. Unlike 12L14, adding the "L" in 41L40 doesn't result in a radically
easier-to-machine steel. It gives about 25% better machineability than 4140,
which is mostly of interest in production screw-machine operations.

4340 is better for toughness and strength. These 4000-series alloys get more
hardenability than the 40 points of carbon would suggest, because the
chromium content aids hardening. And those two are tough steels, reasonable
to machine in the normalized condition (which is a way they're commonly
sold), and slow-quenching for safe hardening. 4340 is basically
oil-hardening -- almost air-hardening in small thicknesses.

If it were me, I'd go for 4340. If 12L14 is hard and strong enough in
flight, then 4340 normalized (around 90 ksi yield, IIRC) is a lot stronger
than you need for normal flight, and it's very tough. I'd think it would be
a good choice to tolerate modest crashes. If you leave it normalized, you
wouldn't even have to heat-treat. If you have a fairly rigid lathe, it's a
little slow but not difficult to machine.

This is a semi-educated guess, but I think the ultimate material for
small-engine cranks might be 8620. It's a low-carbon alloy steel that's
really tough. It's made to be case-hardened. If you aren't going to
case-harden, then I'd revert to 4340.

If you can get a piece of 4340 normalized to try, and if you find that you
can machine it Ok, then you might be set. It's a common grade.

Another thing I know little about is the surface finish you can get,
machining small pieces of these various alloys. 12L14 can produce a nice
finish even for hackers. With 4340, it could be more difficult, especially
if you use the annealed grade. I'd consider finishing the crank and bearing
journal(s) by lapping, anyway.

Good luck.

--
Ed Huntress