PDA

View Full Version : Home oil hardening 4140 and discoloration



SirLesPatterson
05-24-2015, 12:09 PM
Today I have been experimenting with hardening at home by means of an oxy/acetylene torch and oil quenching. I had two parts that I messed up when making, they are about 0.25"x0.25"x1.25" and are machined from annealed 4140.

The first one I heated until it was yellow (I think, colorblind guy here) and then immediately quenched it in nasty old motor oil that probably has other goodness in it such as transmission fluid, dead bugs and mice, etc. The part did harden but could still be filed with medium effort - it was significantly harder than the annealed state. The part is black now.

So then I got online and did some homework. For the second piece I made a few changes...
1. New clean oil, I used non detergent motor oil that I use in the shop.
2. Pre-heated the oil to make it thinner because it was pretty thick.
3. Heated it to "bright orange" with my teenage daughter informing me of color.
4. I held it at that color with a soak time of about 2-3 minutes.
5. Quenched in the oil, swirling, etc. then let it set in there about 5 minutes.

The outcome seems pleasing. The second part is significantly harder than the first. I believe I overheated the first one and didn't give it any soak time before quenching. I did another high tech test involving scratching the parts with a drywall screw. I am very pleased with the outcome compared to annealed, I am unable to scratch the hardened part - merely making what looks like pencil marks on the blackened finish. I still need to try tempering them.

Question 1. Are these parts blackened for good? I can live with that but am curious if there is a way to avoid or clean the discoloration when oil quenching at home.

Question 2. Are there any suggested modifications to the technique I used?

Thanks.

MTNGUN
05-24-2015, 12:42 PM
I'm told that you can buy an anti-scale powder (http://www.use-enco.com/CGI/INSRIT?PARTPG=INLMKD&PMPXNO=25226997&PMAKA=328-5920) but I haven't tried it myself. I've also heard that borax works. Apparently you warm up the part, dip it in the powder, then proceed with heat treating.

My 4140 parts also turn out with a black scale (I too use old motor oil). The scale can be removed by grinding (flat parts usually warp during heat treatment and need to be ground flat, anyway). You can also remove it by beadblasting but it takes an aggressive blast because the scale is pretty tough. If it is a round part, put it in the lathe and polish it with emory.

Your technique sounds reasonable. Obviously a furnace would be better but one makes do with what one has.

flylo
05-24-2015, 01:07 PM
Find a dental oven. I'm always buying crazy junk. These are 12v & go to 2000 deg F in short order, just set the dial & watch the gauge. I use mineral oil used in aircraft. The ovens are small, cheap & work great for small parts.

mars-red
05-24-2015, 01:24 PM
If you heat treat the parts wrapped up inside stainless steel foil, it will prevent that heavy black scale.

HWooldridge
05-24-2015, 05:59 PM
Better to use a magnet to get around the problem of color judgment. Simple steels like 4140 will lose magnetism when they reach austenitizing temps; quench in warm oil and draw to about 400-600 degrees, depending on intended use.

The black color was carbonized oil - just like seasoning cast iron. As suggested above, you can bead blast, sand, wire wheel, etc. to remove - it is simply a thin surface layer.

Another simple test for hardness is to acquire a set of Rockwell files. They come in groups sorted for hardness and you use them by sliding each across the surface of the heat treated part. File slides = part is harder than file; file bites = file is harder than part. Works well to get you into an approximate range.

Lee Cordochorea
05-25-2015, 11:51 AM
As H said, use a magnet. The Curie temperature for 4140 is just below the transformation temperature. Translation into English: The magnet will stop sticking just a smidge before the steel is at the right temperature. This trick won't work for all steels, but it does work for the one under discussion.

Once one knows exactly what color that is in a given lighting one may judge by color in the same given ambient light.

The black coating of burnt oil may be sanded off.

garyhlucas
05-25-2015, 08:52 PM
I use the magnet trick when welding bandsaw blades to hit the right annealing temperature. A lot better results than using color.

metalmagpie
05-26-2015, 07:04 AM
Bead blast your parts. Or rough your parts leaving maybe .003" then harden then grind to size.

Doozer
05-26-2015, 08:38 AM
Try liquid laundry detergent to quench your parts in.
It is said to make lots of rainbow temper colors.

-Doozer

Evan
05-26-2015, 11:21 AM
Just for reference:

http://ixian.ca/pics7/colortemp.jpg

SirLesPatterson
05-26-2015, 08:27 PM
Thanks for all the feedback everyone. I will explore the magnet and other techniques. Very helpful info!

mattthemuppet
05-27-2015, 02:29 PM
Jumping onto someone else's thread I know, but would that black colour work as a rust preventative? I have some things I'd like to make out of steel that I don't want to rust, but hardness and most of the dimensions aren't especially critical. I only have a hand torch and some oil (or transmission fluid), although I have heard that new/ clean oil cuts down on the crud. thanks!

thaiguzzi
05-28-2015, 12:40 AM
Jumping onto someone else's thread I know, but would that black colour work as a rust preventative? I have some things I'd like to make out of steel that I don't want to rust, but hardness and most of the dimensions aren't especially critical. I only have a hand torch and some oil (or transmission fluid), although I have heard that new/ clean oil cuts down on the crud. thanks!

Exactly why i finish all my steel tooling projects by hot oil blackening. Used diesel oil is the best, far better than new oil which will give a brown finish, and not as rust resistant. It is a problem here where i live due to the humidity. Leave the item straight off your torch in used diesel oil overnight, you will be impressed.