Making chisels out of files

Help Support

This site may earn a commission from merchant affiliate links, including eBay, Amazon, and others.


Established Member
24 Aug 2015
Reaction score
I have a long (short?) history of trying to figure out how to make ideal tools - like ones that you'd choose to use over tools that are purchased.

Before the whole unicorn thing last year, I got a fascination with making a chisel that had both excellent edge holding and terrible wire edge holding properties (as in, one that you could sharpen on a washita and the wire edge would sort of fall off without ever being organized).

I also remembered that files are generally the plainest steel one is likely to find cheaply (in a used file), but I've snapped a file or two, so I got the idea that maybe they should be hammered mostly to shape and then I could finish them by grinding and by hand. I've also read in some literature that files are more durable if the carbides are left large, and this is something we don't so much want for woodworking (all though they're probably not *that* large*

So, I made one, and then gave it away expecting to make more, and finally made a second this past weekend. I don't want these chisels to look like they're amateur made, so I'm experimenting with ways that a guy with limited tools can get a bolster on the chisel, and that turns out so far to be a mild steel bolster drilled and filed just undersized and then heated and hammered onto the chisel tang so that it shrinks the to tang (high carbon steel cracks if you do this...i found that out the hard way).

Having made two now, I have a better idea how to get the aesthetics right and will safe edge a file for the bolsters.

The cutting properties of these are superb - they remind me more of japanese chisels than western chisels in terms of how the edge properties are, but they're long and light like an older English chisels. It takes about two or three hours to make one this way, steps as follows.
* belt sand the teeth off of a file (teeth could have anything in them)
* cut the end of the file off with a grinder and rough further in to the file for the tang
* heat and hammer the chisel to rough shape (now it's annealed)
* belt sand the scale off
* shape by hand most of the rest of the way with a vixen body file and a mill file
* find suitable mild steel plate and cut a hole in the bolster blank, then file it square until it fits the tapered tang
* cut the bolster loose, and then heat it to bright and hammer it onto the tang
* file the bolster
* heat treat (oil is fine), temper and then install a handle (I like the burn in method) - you can heat the tang but not threaten the chisel by holding the chisel in a wet rag when heating the tang

They seem to come out a little harder than O1 or 1095, and they can warp a little bit which requires more additional work, but they grind (file steel really sparks intensely) and hone really nicely and have extremely good edge holding.


both of the chisels that I've made have better properties (sharpening and letting go of the wire edge and then holding an edge longer without being brittle) than any commercial chisel that I've gotten aside from japanese, and it makes me wonder why that is. I think it's likely that the steel that's in files is so plain that manufacturers have zero interest in dealing with its warping in heat treat.

The two files that I made were with a chisel marked "heller" and "india". I've since bought a lot of used old chisels off of ebay (they're only a dollar or two that way, and the file and gas used to make each one is probably less than $10). I've ground the teeth off of two more, another heller and an old nicholson. Even on a slow belt sander, the sparking lets you know that they're very plain and high carbon - it comes off in a shower and almost makes a flame at the point where the sparking is coming off of the tool - and when it's kind of flame because of the intensity, it will actually make a small cloud of smoke.

I see a lot of videos of turners grinding skews out of files and having broken some and looked at the steel (it isn't fine like razor steel), I can only guess that some of those break - it's worth going the extra mile and hammering the steel to try to refine it a little bit.

(used saw files make superb marking knives, too - same cutting properties and baiscally same method - grind the teeth of, smash them to rough shape on an anvil and then refine, harden, temper and sharpen.
(I'll clean up some of the cosmetics on this chisel, but the first one or two of something are always to learn where the mistakes will be rather than expect perfection and get frustrated). a hard block with sandpaper will be enough to correct some of the wonky lines on the top of this chisel and make it look less homemade. Something other than the fastest possible cowpat shaped handle would also be an improvement, but I don't want to invest that time right away. A finely made handle above a bolster that has gaps between it and the tang looks pretty stupid. My second attempt, no gaps on the bolster.
I have way too many good chisels already, but I do like this.

Being in what was once the city of file making, it's easy to find good old files and rasps for next to nothing (if I look, it's usually for hand stitched cabinetmakers rasps with reasonable points left on the stitches (just run a finger over) - now have several for peanuts). Worn out straight files should cost even less although the sellers will think I'm even madder.

Old files were of course the country blacksmiths first source for all tool making, but it's very good to see your simplified routine and hear the results.
I haven't ground a rasp yet. When you do something unusual (in this case, forging a file instead of just grinding it), you get all kinds of bad advice. When I made my first marking knives out of files, one person was insistent that they were case hardened all the way down to discouraging even trying to make anything out of saw files. The way they break made this seem unlikely.

Since then, I've been told a couple of other times that the files can't make good chisels because they're case hardened. I haven't encountered a case hardened file yet, but I'm sure there were some cheapies made at some point with case hardened teeth on them just to pass off cheap goods that look like something better.

That said, I don't know if the core of a rasp (where the teeth have to be a lot tougher) is the same high carbon steel that files are.

But if you can find big double cut files (half rounds are ideal for hammering unless you can find really thick bast-age files) that have a lot of core below the bottom of the teeth, then it's good to go (just need an anvil and enough heat to hammer them).

I don't have the heat (or don't set up enough) to make them bright orange or yellow and hammer them at mid orange just to get basic shape and try to refine them a little bit.

You're right in the implication of need - there are so many good files, there's no need to do this. I wanted to experiment with proportion (less thick like a lot of newer chisels and more like vintage chisels) and see if the edge properties that I found in saw file marking knives would transfer to chisels. It does.
Thanks for pointing that out. I think I mentioned it above -I said something years ago about not making chisels because there's no point. There still may not be, but after making a couple of marking knives out of saw files and comparing them to a couple that I made out of O1, they are undeniably tougher and let go of their wire edge much more easily. Strange when something makes O1 seem "too highly alloyed", and before the whole buffer sharpening thing, I was interested in taking those properties to a chisel that would sharpen ideally (finely) on natural stones without retaining much wire edge to even need to bother with stropping.

Slowly getting to my point. When I first started making tools and repurposing things, I did it just to do it and it was fun. The novelty wore off and then you were back to using whatever you had before because it was better. I don't do much kind of quick modification of something or acid etch, etc, because while it may be a way to avoid buying something new, it doesn't pass the preference test. The surprise has been that once you kind of narrow down desirable properties, there are still things like this where the result is preferable to commercially made tools. But some time has to be spent experimenting with use of them. I threw away too many old planes that I'd made before I learned that they were kind of a waste of time to make if not investing a little more time to make them as good as or better than anything that can be bought.
Thaaat's not a kniofe! 🤠

Edit... I actually would like to see your marking knife David.
Two very very cheap knifes I have bought from axi, not telling the brand, are made of butter, an am getting fed up with them.
I do hope they remedy this, wouldn't be that much to fix.
On a plus, its taught me how to sharpen them so I can get that point back,
extremely low honing angle, and not skewing the knife.

I like single bevel ones, and am shying away from making some from a kitchen knife as they might be a bit flimsy for me, with the holes I dig for myself.
Last edited:
I also have some old woodies that are too far gone,
Not saying I can bring myself to cutting another plane iron up for making knives from, but I might as well ask if anyone has made a single bevel knife from old laminated steel plane irons.

I'll get pictures of some. They aren't really refined at all - just sort of a wharncliffe or sheepsfoot profile and then the sharp corners ground off of the rest of the file so that they can be gripped as is or wrapped with cord/tape for a more comfortable grip.

I grind them thin - this may start sounding like unicorn talk. I make them reasonably thin in bevel profile and then buff the edges so that the tip is strong and they become.

The virtue, though, is that you can make a bunch for almost nothing in any profile you want and then you don't feel the need to be precious with them.
I also have some old woodies that are too far gone,
Not saying I can bring myself to cutting another plane iron up for making knives from, but I might as well ask if anyone has made a single bevel knife from old laminated steel plane irons.


I haven't, but there's no reason you couldn't. Soy oil is fairly common here in the states and has a high smoke point and good thermal transfer, so I haven't had to reharden anything using water (most laminated irons are probably water hardening, but thin cross sections usually do well in oil).

The real issue with a laminated plane iron won't be in the business end, but whether or not the mild/iron end is strong enough - slap a handle on it that goes up to the hardened steel and it should be fine. This stuff is all much easier to do if you resign yourself to annealing, shaping, hardening and tempering rather than trying to work with hardened metal.
Thanks David
I should clarify that I only want to make a pair of single bevel marking knives with an acute point.
Good to know the business end will be suitable.
Plenty of metalwork to be done in a few months time, so don't mind spending lots of time angle grinding a wee bit of tough stuff, rather than learning how to work hard metal cleanly by tempering and using saws instead.
Just don't have the funds nor space either.


I learned from these (the top two are filthy with shellac - I can't remember what I was doing with them, but that'll come off with alcohol) - the bottom knife is the only one that I really spent time thinking about blade aesthetics because I was trying out filing a pocket knife blade profile to shape. But that taught me something -the chisel was rubbish and I thought maybe I could improve it by hammering. It's still rubbish. Fine for marking, but it never really gets a really fine edge.

Lesson being if a tool is rubbish, it probably doesn't make sense to spend time screwing around with it to see if you can solve its issues (it's lakeside brand, a second rate chisel at best here in the states).

The rest, you could make two an hour of the type by belting off the corners, heating them quickly and hammering them and then belting or filing a rough profile, then sharpening, hardening and tempering. They need to be reasonably flat when done so that they will sharpen well on stones before the buffer - that's, to me, most easily done with files and stones rather than getting really lazy and trying to grind or belt some wavy profile into a knife (if even that level of care wasn't needed, you could make 6 of them in an hour with a decent belt grinder).
DW, I'd be surprised if any 19th/20th century files are case hardened in the sense of being a piece of low carbon steel with added surface carbon. It is poss that when quenching and tempering, the surface accidently (or on purpose?) got a different treatment so was differently hardened.

But there are some versions of the milled files (in the UK FBT or Aven) that are flexible, so I think they will have a soft core and be surface hardened (they are harder than usual files/rasps) - these may well be a different alloy from good old plain carbon steel, and maybe nitrided or similar -- anyone know for sure, I'm just surmising?

Regarding using old rasps for tools -- old country tools, especially French, were often made from rasps as is easily seen from the remains of teeth.

Re marking knives - I quite like them a bit flexible - if this is your wish, look no further than your broken or worn hacksaw blades, they even have a ready hole at the end for pinning a handle (or just wrap).

300mm HSS hacksaw blades of the basic type are fine, and can be ground with little worry about overheating, some are flex, extra hard or bimetal - make your own decisions if using these. I actually have some old carbon steel blades which I like. (I'd avoid some very cheap new HSS blades which are thin, but if you need extra flex--?) -- wear eye protection.

For a bigger knife, keep an eye open for an old worn thicker machine hack blade.

These have worked for me -- any problems for you?
Last edited:
yes on the files - lots of people said case hardening, and I doubt at least in the last 125 years there was really any precedent for files other than early third world copies where a dollar saved on tool steel stock would've made a difference. The fellow over here who had been a professional toolmaker never actually warned me of such things when forging decent files but mentioned that at some point in the past (which could be 50 years ago now), he'd seen really cheap overseas files that were case hardened. (or long story short, only more or less early fraudulent tools were case hardened).

I was thinking rasps being softer at the bottom of the tooth only to prevent breakage (I actually can't think of a rasp that I've used that's broken), but any expensive rasp would only have minimal tool steel cost in it, especially if hand stitched.

The really thin knives in the picture above are about 60-62 hardness, but have decent flexibility (they're tempered medium straw) I don't know what their breaking point would be, but they're too hard to have the kind of memory that a hacksaw might (bend and stay bent).

Hacksaw blades and bimetal recipro saw blades are probably a good starting point for "grind it and forget it" marking knives. I've seen thin parting tools made of HSS demolition blades and they worked very well.
Looking at a tool store catalogue of about 70 years ago I see that all the Sheffield made files and rasps are advertised as made from 'best cast steel' - ie plain carbon tool steel. (The (more expensive) milled files (some flexible) are described as made from a 'specially hard alloy steel')
your post prompted me to do a second sweep, and it looks like nicholson's recent offerings are 1095, or at least that's the consensus. Their black diamond files (much harder) are 1.3% carbon but undescribed what else beyond that allowing their hardness spec of 72.

I did see some specification that the outside of some files is case hardened (I haven't had one that I can think of that had the kind of flexibility mild steel would have, though, and the ones I've broken were definitely not case hardened).

That leads to a follow up question, assuming some tool steel files were actually case hardened, why would someone advise that means I couldn't hammer and reharden the whole thing.

I have heard that early 1900s files (I have some austrian needle files from the early 1900s, there's nothing now remotely similar in quality to them) were generally higher carbon steel, a little harder and longer lasting than more recent plain files. Aside from knowing the quality of the austrian needle files that I have, I can't really comment on that.

I do have a nicholson half round in the next batch of blanks with teeth stripped off, so I'll see if it seems any less capable than the heller and "india" marked files that I have. I'm guessing not. 1% plain carbon steel should make a dandy chisel (1095 being generally better as a chisel steel than O1 - same with razors, 1095 would make a straight razor miles better than O1 - I've used expensive boutique O1 razors that people sent me and O1 has just enough wire edge holding nonsense that it's not very good as a razor).
There also some foreign files listed as being made of t12 alloy. This is a chance for confusion as it appears to be a chinese spec or foreign spec 1.2% carbon, more in line with some of the comments about historical files that I've seen.

still hanging on the idea that older files being more like 1.1-1.2 carbon steel and able to hold an edge at high hardness better being an explanation as to why old timers call older files more crisp working and longer lasting (1 vs 1.2% plain steel is meaningless in difference for a garage toolmaker - both will have good working properties at high washita hardness - or something like 62, or light/mid straw oven temper).
Thaaat's not a kniofe! 🤠

Edit... I actually would like to see your marking knife David.
Two very very cheap knifes I have bought from axi, not telling the brand, are made of butter, an am getting fed up with them.
I do hope they remedy this, wouldn't be that much to fix.
On a plus, its taught me how to sharpen them so I can get that point back,
extremely low honing angle, and not skewing the knife.

I like single bevel ones, and am shying away from making some from a kitchen knife as they might be a bit flimsy for me, with the holes I dig for myself.

I made one from a jigsaw blade, single bevel but spear point - works very well and lasts for ages before needing a touch up, I did a WIP here which you might be able to find, but there are others as well from forumites.
There are two parallel threads on using worn files - I posted on the other about some new rasps being sharp but flexible - ie probably case-hardened in one way or another.

Yesterday I was in a junk-shop and came across a fine chisel made from a file (at a guess from a Sheffield engineering works - looks like the application of a serious grinding machine (or more patience than I'd have)) - blade length 8ins, width 1in, thickness 3/16in. It is ground on all faces but shows the remnants of the teeth on the top, also the Sheffield makers name stamped, it has a very small bevel to top edges and the kind of small tip bevel sometimes added to a heavy-duty chisel (occasionally sold new with this). A quick file test shows it's suitably hard, - it cuts nicely - whether it's too hard and chips, I'll find in practice. It has a pretty flat back (v slight convexity one side) and has a small back hone at the tip). No added bolster, but the handle is a registered type and the steel ferrule backs up the tang nicely.

Next to it is the new import cabinet-makers-type rasp, very sharp fine teeth that cut very quickly but do leave tramlines. It is very flexible - I can bend it in my hand easily (could be useful for a sculptor). I added the handle. This one could never be repurposed as knife or chisel.
Last edited:
I've read a bit further on files and hardening, and (maybe I said it in this thread already), it looks like there are some files that are case hardened even though they're high carbon steel through and through. I guess that could be for flexibility.

I've finished a couple of more chisels and think I'll make these for a while instead of just quick experimenting - all of the chisels that I've made (annealing, then hammering and reshaping, vs. just tempering) have incredible edge holding, like a middle of the road japanese chisel. Guessing (probably already said this, too) that moving toward oil hardening steel away from water hardening for chisels had more to do with stability than anything else (in heat treatment) as a forged file is a markedly better chisel for wood than O1 flat stock hardened and tempered.


I ordered a larger two-burner knife maker's forge this morning so that I can make some full length paring chisels, also.

Talked to George Wilson last night (we talk about nothing of significance on a fairly regular basis), but mentioned that I was making chisels out of files, and I guess at some point in the past, he tempered a large flat file back (but didn't forget it or fully anneal it and reharden) and made a lathe skew out of the file. It ended up breaking and flying. I haven't seen the same brittleness, but also don't use anything other than cheap HSS turning tools on a lathe - there's no real risk in hand or hammering use, anyway, vs. a lathe.

The one that you found and pictured looks pretty neatly refinished.