Thread: Converting spent files into chisels

to use files or O1 and heat to nonmagnetic - and then a little more as quickly as possible - is all you need.

I have no temperature control, either. My three samples to larrin thomas for O1 were 0.1 point apart in hardness. I sent him a flyer that got lost in the forge, or really was a matter of not paying attention - it was too hot for too long on the order of some fraction of a minute - something you'd never allow if counting seconds and looking at color, and it was off of the main group by 0.7 hardness. obviously, the last inch or two of a chisel is the important part, and it's nice to have at least partial conversion further up so the chisel isn't soft and springy.

1084 is just get fully nonmagnetic, try not to get much color change after that, just a little is OK but going, for example, from medium red to medium orange is going to cause problems. going from medium red to a little brighter red in 15 seconds won't. You're doing the same thing for the same time period, just by judgement, how much color change. I excepted 52100 because it's harder to get right. it's often used here because it shows up highly tough on toughness charts, but I don't know that it's common there. I'm fiddling with it because once one goes to round rod here, the market is limited. It's the opposite of 15 years ago, but the retail market has realized specialty steels should be in flat stock now because there are volume buyers who literally do nothing but send patterns off to a waterjet or laser shop, and then have the cutouts sent to a heat treat shop and then maybe even for post heat treat machining elsewhere. This is blue spruce, for example - call yourself a toolmaker, make shiny handles, put parts together. It's more common than that in the boutique tool market - not just blue spruce, and maybe now that things went to woodpeckers, more is done in house.

That excepts round bar and if you want rounds in anything other than the basics (i can't even find it in 1084, by the way) you're out in the weeds.

So, that's the simple process. Snapping a cutoff sample here and there and taking a picture with a $15US handheld digital scope- the PC is needed (or a phone) to run the scope, so when you snap the picture, it's already inventoried. if there's no perceived problem with a tool - like when it works as well as a commercial tool, that's all that's needed - no need to even snap samples. Snapping is the gateway to finding out what you can do or not do. it's worth learning for someone who is woodworking and thinking "I'm going to use a lot of hand tools" - what's available at the end of the mouse pointer or at the flea market is limiting.

No need for anything expensive, no need for temperature control outside of tempering (which is almost always 375-400F for any simple steel for woodworking purposes, so no special oven needed).

how does that lead to files? they're an ideal source - they have tolerance for heat increase between O1 and 1084, so it's like playing the same song but adjusting how long the musical breaks are between melody. when a file is amenable and will transform completely, it's like getting a hold of a japanese steel tool. they grind easily, don't generally air harden if you're grinding them hard and ...well, there's usually something left over that's going in the trash otherwise. they will make little shop knives and marking knives that are to die for without much skill. let alone little knives to cut apart boxes and such.

As far as hardness, some kind of natural finishing stone will tell you all you need to know. If you're in a good range, the tool works well, the wire edge comes off on the finish stone (another thing files are good at - letting go of a burr).

It took me 13 years to break down and get a hardness tester.

Metallurgical analysis beyond snapping is beyond what we need. I'd like to etch samples and look at them out of curiosity, but doubt that will ever happen - there's nothing to diagnose, just "I wonder how big the actual grains are if i can etch the boundaries to see them". that's not easily done and it requires nital, which isn't exactly an over the counter etchant. If it's left unattended and loses some of the bulk methanol in the mix, it can become explosive, too.

I'm not stating this to say there's any requirement to do any of it, but that I think the things useful to us are within reach easier than learning to cut woodworking joints. And making a purely honest statement, for anyone who has gotten good at woodwork, nothing that needs a furnace vs. the method above is actually going to move the needle woodworking. If you have really hard aussie timbers, abrasion resistance isn't your enemy, lack of tearout control combined with a deflecting or nicking edge is. Most plainer steel tools are usually soft - it's created an artificial sense that they underperform because of the alloy rather than the hardness.

I can't think of anything where 1084 comes up short when it's dialed in, either - if I had only 1084 or only O1, it wouldn't make a difference which one until making things thicker than plane irons or bench chisels.