Heat Treating Seven Different Steels with a Forge and Magnet

[Larrin]

I tried my hand at heat treating with a forge. I used a range of different low alloy steels and measured each for hardness and toughness to see if I could successfully pull off good properties with little or no experience. I used the information gathered in my previous article/video about the best prior microstructure for forge heat treating.
Article: https://knifesteelnerds.com/2021/09/23/how-to-heat-treat-knife-steel-in-a-forge/
Video:

 

[Londinium Armoury]

How did you find it? I quite enjoy heat treating by forge.
Good video was enjoyable to watch.

 

[Justin W]

I have zero experience with heat treating. One of the things that always concerned me was the visible appearance of a temperature gradient in blades that have a primary bevel roughed in. It seems to me to be too easy to overheat the edge. In some cases, this can be preferred since the spine doesn't need to be as hard. I guess it is a balance between thickness behind the edge and spine thickness.

 

[N.W. Gean]

Thanks for the article. I could never get Pop's 8670 to harden right in my forge, but I'd bet now that I was overheating it. I could probably go straight into the parks 50 from non magnetic and be perfect.

 

[Mecha]

Good stuff, I was looking forward to this report. What do you think, Larrin? Will you mess around with the forge more in the future? It's fun, and wild to observe decalescence in a blade, although it can be hard to perceive until you've seen it, and then it can't really be un-seen.

A question: does maintaining a temperature of 2100 degrees F continuously for an hour approximate forging, where a billet is repeatedly heated and allowed to cool somewhat, in between being beaten or crushed? I don't know how hot a steel billet typically is when forging it, how high the temperature is held, or what final effect this may have. I know with titanium, grain growth and irreversible gaseous contamination is extremely rapid at those temperatures, forging must be done at much lower temperatures, and that the process of forging does refine the grain size of the ti down a lot. I do not know if the same is true of steel. A lot is analogous, and a lot isn't.

I have a general comment about magnets for everyone. They rapidly lose their magnetism once they approach 200 degrees F. When I was re-heat treating my anvil the magnet I was using was quickly demagnetized by the large mass of glowing steel, but I was able to gain enough input to know the top was non-magnetic, and the bottom was not. A few more minutes of heat input and it was quenched top-down in a cold, flowing stream. The HT was a success (anvil is "crome-moly" steel, aka 4140). So I would caution anyone using a magnet to only briefly touch it to the hot steel and be sure it doesn't get very hot.

 

[Larrin]

Good stuff, I was looking forward to this report. What do you think, Larrin? Will you mess around with the forge more in the future?

We will see. My curiosity is mostly satisfied for now.

A question: does maintaining a temperature of 2100 degrees F continuously for an hour approximate forging, where a billet is repeatedly heated and allowed to cool somewhat, in between being beaten or crushed? I don't know how hot a steel billet typically is when forging it, how high the temperature is held, or what final effect this may have. I know with titanium, grain growth and irreversible gaseous contamination is extremely rapid at those temperatures, forging must be done at much lower temperatures, and that the process of forging does refine the grain size of the ti down a lot. I do not know if the same is true of steel. A lot is analogous, and a lot isn't.

With the high temperature most bladesmiths use to forge, and the very small amount of deformation that the knife sees (and very slowly at that), the steel basically just has grain growth during the forging portion of bladesmithing. If there was a lot more deformation or the temperature was lower then there would be more complex deformation-induced differences. Though even in that case normalizing would return things to a relatively similar state.

 

[tkroenlein]

Crawling out of my hole to comment.

Larrin , this particular article/video/experiment was perhaps my favorite to date, for a few reasons. Actually, many reasons.

First, kudos for simulating heat treatment of forged steel. When you mentioned that you were going to do the forge HT experiments, I was curious if you'd test any or all specimens in the as received condition or "forged." I think your approach here was spot on from a practicality standpoint.

Your presentation of the heat treat response of different micro structures finally clicked for me. I knew that my forged blades got harder with lower temps and practically no soak time, it just never clicked that pearlite would be so much different than spheroid.

This experiment also helps to provide tangible data that will resolve the discussions like "80CrV2 needs a soak/no it doesn't." People, especially first timers, need good advice. This can help them make that "oven or forge" choice more easily.

Many thanks.

 

[Mecha]

We will see. My curiosity is mostly satisfied for now.


With the high temperature most bladesmiths use to forge, and the very small amount of deformation that the knife sees (and very slowly at that), the steel basically just has grain growth during the forging portion of bladesmithing. If there was a lot more deformation or the temperature was lower then there would be more complex deformation-induced differences. Though even in that case normalizing would return things to a relatively similar state.

Ahh thanks for the info. Looking forward to the next installment of knife steel metallurgy info.

 

[Nick-D]

Very interesting.

Given that many beginning knife makers will be heat treating stock removal knives from an annealed state, is there a simple process to take a blank from the annealed state to perlite? Would normalising cycles do it or it it a forge only deal.

I built a heat treat oven, but definately struggled with the "and a bit hotter" part in the good old firebrick forge.

 

[Larrin]

Very interesting.

Given that many beginning knife makers will be heat treating stock removal knives from an annealed state, is there a simple process to take a blank from the annealed state to perlite? Would normalising cycles do it or it it a forge only deal.

I built a heat treat oven, but definately struggled with the "and a bit hotter" part in the good old firebrick forge.

I started from annealed for the study. I didn’t do any forging. You could probably just normalize but there may be some cases where a higher temperature cycle is necessary first.