Making chisels, the completely empirical, non-metalworker approach....

[IanW]

LOML came into the shed last week & announced she wanted to do some simple carving. So we did some experimenting and it quickly became apparent she needed some smaller tools than my regular carving chisels. Something more like the lino-cut set I'd had for years but gave to my daughter a year or two back would fit the bill.

I have quite a few dead chainsaw files in various sizes so I hauled some out, cut them in two and "annealed" them with a MAPP torch. Heating to orange-red was about as much as the torch can manage outside my "forge", but by heating the whole length progressively & letting them air-cool they were soft enough to file (just). I spun them slowly in the lathe & filed & sanded the teeth smooth.

The gouge channels were formed with smaller round files and then the ends were re-heated and bent to the desired shape. I debated what to use for a quench to harden the business ends, but in the past I'd used water with no apparent problems, so water it was. I'm not sure how hard I'm getting them - plenty hard by the small size of sharpening burrs and the feel on the stones, but I'm pretty sure they are not full-hard - the edges seem tough but not brittle. So I haven't attempted to temper them back at all - they sharpen nicely & seem to hold their edges as well as my good-quality carving tools. I turned some handles suitable for small hands & the budding carver has given them the thumbs-up:

1a Mini carving tools.jpg

While I was at it, I decided to harden a very thin chisel I'd made from an old circular-saw blade, It was made for a job where I had to pare a very narrow slot & has come in handy a few times since, but saw plate is really much too soft for chisels and it needs constant sharpening even on very soft woods. So I warmed 'er up & dunked it in the water:

1b Water quench.jpg

I don't think that went too well......

I cleaned up my now much shorter very thin chisel and tried quenching in very warm oil. No problems - it warped, as shown by an initial rub on a diamond plate:

2 Quench warp.jpg

But far less than I expected & it cleaned up quickly:

3 Quench warp smoothed.jpg

Again, it came hard, but not brittle-hard, sharpened well, & pares end-grain without losing its edge after a couple of strokes:

4 Ultra thin paring end grain.jpg

Emboldened by that minor success, I made a slightly more robust ultra-thin small paring chisel. Again the file was annealed with the torch, ground clean (I tapered it slightly from the ~3mm thickness at the tang to about 2.4mm at the tip. The lands were ground using a very simple wooden holder & eyeballed for symmetry (they are slightly different, just like on many old hand-ground chisels):

5 File conversion.jpg

Again oil-quenched, it has come out at a workable hardness. The edge retention seems a bit better than the saw-plate version.

6 File chisel paring end grain.jpg

This one might be a bit over-hard, but I'll need to use it a bit before I decide to try tempering it.

So there you go, don't be intimidated by the mysteries of metal working, very little science went into these, and I've got usable tools. They may not be world-beaters, but for very little effort I have some custom tools that will do the jobs required of them quite adequately....

Cheers,

[Skew ChiDAMN!!]

They may not be world-beaters, but for very little effort I have some custom tools that will do the jobs required of them quite adequately....

Really, that's all that matters. (Aside from the bonus of pleasing t'other half.)

Nicely done!

[Bushmiller]

Emboldened by that minor success, I made a slightly more robust ultra-thin small paring chisel. Again the file was annealed with the torch, ground clean (I tapered it slightly from the ~3mm thickness at the tang to about 2.4mm at the tip. The lands were ground using a very simple wooden holder & eyeballed for symmetry (they are slightly different, just like on many old hand-ground chisels):

5 File conversion.jpg

Again oil-quenched, it has come out at a workable hardness. The edge retention seems a bit better than the saw-plate version.

6 File chisel paring end grain.jpg

This one might be a bit over-hard, but I'll need to use it a bit before I decide to try tempering it.

Ian

Very satisfying to make something out of nothing, but I couldn't see what the metal source was for the "ultra thin small paring chisel." I may have confused the circular saw blade and saw plate (handsaw?).

Regards
Paul

[IanW]

..... but I couldn't see what the metal source was for the "ultra thin small paring chisel." I may have confused the circular saw blade and saw plate (handsaw?)....

Paul, apologies for the confusion, so to set the record straight:

The "ultra thin" chisel is made from some circular saw plate and is ~1.6mm thick. It was a very old saw from pre-TCT days or at least before they became the only game, so I assume it was nominally 1/16" to start with.

The slightly thicker one is the bit of re-purposed file.

I was using the file chisel today paring river oak, which ispretty tough stuff, but my new chisel kept on cutting sweetly all afternoon, so I reckon it's a keeper....

Cheers,

[Bushmiller]

Paul, apologies for the confusion, so to set the record straight:

The "ultra thin" chisel is made from some circular saw plate and is ~1.6mm thick. It was a very old saw from pre-TCT days or at least before they became the only game, so I assume it was nominally 1/16" to start with.

Thanks Ian.

I know people have used old circular saw blades for knife making so I see no reason why they shouldn't work for slim chisels and your first indications are that it is good. Nicely made too.

Regards
Paul

[MartinCH]

Hi Ian

Thanks for the ideas of where onto source suitable steel cheaply - and the experimental evidence that simple process can make an adequate result.

Need at some stage make some narrow thin 6 to 12mm , blunt and double bevel "woodworking knives" - that are stronger than the modified scrapers in use at the the moment - imagine mine will be infinitely crude, compared yours.


Regards

[D.W.]

I've got nothing to add...currently in the throes of experimenting with full forging, but there is a big difference between my antics of trying to get it just right and making a chisel that will get you through 100 jobs. The latter can be simple from almost any source and it's a gateway drug for toolmaking.

At some point, if you're doing a lot of work by hand, it starts to get difficult to find the hand tools you need, and not that difficult to build something suitable.

re: sawblades - they can be all kinds of things. L6 is a popular steel in the US, which is somewhat marginal, and from what I've read, there are various chrome vanadium plate steels and then some only partially hardenable stuff on coated blades that have carbide teeth because the plates are thin, don't really tension and the teeth have nothing to do with the base.

I'm sure there were plain blades made of something like 1084 in the long past.

At first, you can sort of use whatever you want and see what your results are, but at some point, you'll find a bar stock supplier with inexpensive 1084, 1095, etc (O1 will usually cost a little more) and you'll like the predictability.

Variation on blacksmiths who discuss using scrap steel and others who won't. Usually, the pros won't. Not because you can't make a decent knife or punch out of a coil spring, but because you get tired of finding cracks that you didn't cause in finished work.

[auscab]

Nice chisels Ian. What are the timbers you used for the turned handles?
The set for LOYL look interesting that they are different from each other . Or a couple are the same maybe ?



A bit of a ramble about a chisel handle I'm playing with.
I turned a small handle for a HSS 10 x 10 chisel yesterday . A tool for shaping brass on the metal lathe. Holding HSS blanks without a handle is an accident in the waiting if it grabs. I'm trying to see if I can make a handle that the HSS can be withdrawn from and flipped to use the opposite end .
Two chisels in one if it works. I don't want other fittings to hold it and just a 10 x 10 long square hole in the wood made on my chisel mortiser seems to work. I'm trying Blue Tack down the end of the hole today to see if it works at sticking the HSS in position and stop it easily sliding out. LOML bought it home from the supermarket last night for me . Cost was $3.50!
If it doesn't work then Ill do a handle for each of the 4 or 6 HSS shapes I think I need and stick them in.
I used some English Oak that was grown here for the handle. A small off cut from a huge table top we made. Amazing wood that Oak compared to the other Oaks Ive got. Its a slightly figured bit and almost glass like in comparison to the others.

That little knob handle was my THRILL of the day yesterday after bending over planing US Oak all day long.

[IanW]

..... What are the timbers you used for the turned handles? ...

Rob, the 'ultra-thin' handle is"western rosewood' aka Acacia rhodoxylon:

A rhodoxylon handle.jpg

This is a wood I didn't know existed until about 8 or 9 years ago, despite growing up not far from where it grows. It has that deep red-brown colour, with small golden highlights (you have to move it in good light to see them clearly). It works a lot like true ebony & has a similar almost invisible grain, & polishes to a glass-like surface like ebony does. It turns like ebony & takes a similar fine finish. Sometimes when planing, I get little patches of fine tear-out, even with a carefully-set plane, but a scraper eliminates that in a few swipes. It makes the best tool habdles but the tree is small & given to forming deeply-fluted trunks, so I've never had a piece big enough to infill a handled plane, unfortunately, but have used it for some smaller planes.

Ther other handle is just some "black wattle" from my backyard. It was a scrap cut of a larger bit when I was squaring it up for the lathe - I was about to chuck it in the kindling wood box when I noticed it had some figure. It was an odd, tapered bit but I managed to get a good handle from it (almost, the small end didn't quite make the full circimference for the thumb-rest):

A sp handle.jpg

..... The set for LOYL look interesting that they are different from each other . Or a couple are the same maybe ? ...

The carving tool 'set' includes a straight in-canel gouge, two bent gouges (one tiny, the other tinier ) a little spoon gouge and a bent chisel. They are all made from 7/32" files, to give you an idea of scale. The shapes are what I reckoned she'd need to do what she wants to do, which is add some fine details on some turnings I've done for her. She managed amazingly well on her first attempt, but she's a very deft little worker, so I was not all that surprised. Gotta watch it or she'll be turning out stuff that puts me to shame in no time....

Cheers,

[truckjohn]

That's some very nice work. Your chisel grinding skill far exceeds mine. I wouldn't hesitate if I had a decent surface grinder, but alas, I don't. So far, I've made one chisel out of a file, and it is very good. All I did was grind some teeth off the "Bottom," flatten it a bit, and put the bevel on the end. Far too much work by hand.

The only thing I've run into is that many files are stitched out of very low carbon stuff and then case hardened or nitrided. This way, they can have screaming hard teeth without worrying about it snapping in half in use. Unfortunately, while the skin is super hard, the core is mushy.

I appreciate the tutorial as well as the pictures.

Thanks

[D.W.]

That's some very nice work. Your chisel grinding skill far exceeds mine. I wouldn't hesitate if I had a decent surface grinder, but alas, I don't. So far, I've made one chisel out of a file, and it is very good. All I did was grind some teeth off the "Bottom," flatten it a bit, and put the bevel on the end. Far too much work by hand.

The only thing I've run into is that many files are stitched out of very low carbon stuff and then case hardened or nitrided. This way, they can have screaming hard teeth without worrying about it snapping in half in use. Unfortunately, while the skin is super hard, the core is mushy.

I appreciate the tutorial as well as the pictures.

Thanks

I haven't found many files that have a central core that isn't hardenable, but what a lot of files are is thick cross section of a file with a lot of excess carbon and they don't fully harden through and through. If they are not made very thin by grinding or forging down, they will not harden in oil, either, and if they're ground and not rehardened, they'll be soft under the fully hardened outside.

https://i.imgur.com/CgnLghi.jpg

There's the center of a brand name file (which, i can't remember -but it's probably an NOS nicholson mill file). The little black dots are carbon that stabilized and the little shiny dots are carbides. if a file isn't hardened through and through, it'll have steel that looks like this around the edges, but the center will seem coarse and torn.

Parks50 or faster is needed to get a lot of them to harden, and some of them will not fully through harden in parks (which was probably on purpose so they wouldn't be brittle). I forged a huge 7/8" square file into a chisel shape last week and quenched it in parks and thought maybe I'd finally found just what you're describing - it was only partially hard. But after faffing around with brine due to the defective 115crv3, I gave that thing a cool lunch in brine and ground off some of the surface - 68 hardness out of the quench. I believe most of those old files were made out of a steel called C125U or C130. it's similar composition to 1095 or 1084, but with more carbon and lower manganese - I think the lower manganese was to prevent it from through hardening as easily.

A very good skill builder with hardening things that have excess carbon is to get a file or something the picture shows above and see if you can refine the grain. if you heat treat something like 1095 and learn thermal cycling (which is sinfully easy, I'm baffled why it isn't talked about more with hand heat treatment - it's the link to getting top performance and being able to reharden the same piece of steel over and over with little or no degradation....at any rate, then you should see something more like this

https://i.imgur.com/fAEI7Xk.jpg

these are, of course, highly magnified. I think an hour of making little test pieces and snapping them will pay back in spades if one is going to do more than make one or two things, but old files are definitely an excellent source if:
* one is willing to heat them three or four times to transition and no hotter and let them cool
* one has the ability to get them to transition and then ramp up the heat in the key area to a full shade brighter, evenly, in about 15 seconds and quench (this will blow up the grain of 1084, but 1084 doesn't benefit from it, anyway - files and O1 benefit enormously from this vs. just going nonmagnetic)

All of this can be avoided with O1, though, if a good version of annealed or not coarse spheroidized is found in bar, it can just be heated a step past nonmagnetic and quenched.

if you decide in the longer term to explore driving the files up, they are a cut above as far as steel goes when it comes to chisels. I got hooked on 26c3 (almost identical to C130, but cleaner spec - C130 is not stocked anywhere that I could find) because it was the first steel that I found that would match or slightly better vintage ward chisels. And I took my lumps on older files not yet being aware at the time when I started making chisels out of them that a bench chisel would really struggle to get hardening depth with canola oil or soy oil.

if you really had precise control of hardening depth on a chisel, it could be an asset, but I think it's never going to be that easy - it's either ramp up the pre-quench heat to increase the transition demand and then a harsh fast quench or not. Parks 50 is a game changer, but it's stinky.

it's down now to about $45 shipped on amazon, which is nice. The first I bought it, I had to get it from a steel supplier - there was none on amazon - and it was $80.

[truckjohn]

In addition to being made of fairly simple water hardening steel, most of the "good" chisels have some sort of ultra-hard surface treatment. This improves the life of the teeth substantially, but the massively hard skin makes for one heck of a job if you want to make them into anything. I think that and the weird cores drives much of the advice to anneal them first.

Interesting the problem with 115CRV3. I wonder if it truly is silver steep drawn rod, or just some miscellaneous carbon-bearing Chinesium. My Woodcraft green handle cheapies advertise being made of that stuff, but they are by no means soft. Ironically, Pfeil allegedly uses something similar. I had endless trouble with their bench chisels rolling edges, but I never had that problem with their carving tools.

Last time I tried to buy Parks50 was about 15 years ago. The price was crazy and you couldn't get less than a 5-gallon pail of the stuff. Sounds like they have changed their sales model, so probably time to buy some. Water/brine quenching works, but it's also really prone to cracking.

[D.W.]

In addition to being made of fairly simple water hardening steel, most of the "good" chisels have some sort of ultra-hard surface treatment. This improves the life of the teeth substantially, but the massively hard skin makes for one heck of a job if you want to make them into anything. I think that and the weird cores drives much of the advice to anneal them first.

Interesting the problem with 115CRV3. I wonder if it truly is silver steep drawn rod, or just some miscellaneous carbon-bearing Chinesium. My Woodcraft green handle cheapies advertise being made of that stuff, but they are by no means soft. Ironically, Pfeil allegedly uses something similar. I had endless trouble with their bench chisels rolling edges, but I never had that problem with their carving tools.

Last time I tried to buy Parks50 was about 15 years ago. The price was crazy and you couldn't get less than a 5-gallon pail of the stuff. Sounds like they have changed their sales model, so probably time to buy some. Water/brine quenching works, but it's also really prone to cracking.

Things have fortunately gotten better for the garage chisel and tool maker in the US at least. I don't know who is rebottling parks 50, but it's the real thing. It's mr fireburner man or some weird site (that's not it, but it's not like it's jantz now, which is where I had to go). Jantz is OK, but the shipping costs are made for someone making big orders and you get killed ordering something like a bottle of parks. It's an antiquated business model now that you can just send a pallet of stuff to amazon, pay the fees and still sell at a lower price and make the same amount.

I do think there's something going on at the surface of files, even if it's just a matter of low tempering and the surface hardness is close to what the files are capable of. I would imagine if I really pushed the limits, I could get close to 70 out of a quench on a file.

I guess my contribution now of making this thread more obscure is that if folks are willing to push to be able to through harden a ground down file, it is worth the trouble. The hardness post temper will end up around 64 and your tools are going to have the feel of a japanese tool because C130 is literally the same thing as White 1. P and S (contaminants) are barely more than hitachi white allows, and even the manganese and silicon levels are almost identical.

Which makes my spinning in circles below of trying hitachi 1 just because it's there perhaps a waste.

I did have cracking issues with plane irons and water. Brine is faster, but without agitation, i haven't cracked anything in it yet. the fact that 115crv3 took no distortion from it in any measurable amount is sort of an indication that something is really wrong there. Your mention of the woodcraft chisels elsewhere is an interesting thought, though - to just grind the wings off of the wider chisels, reprofile one to thinner taper and then harden it and see what I get. I could sacrifice the thin ones to snap them to make sure the grain structure looks grainy enough to ensure that it's got carbides in it. Something really smooth like the picture of 1095 or this (80crv2) show a whole lot less to the grain and very little in terms of carbides. For 1095 that means there's a lot of carbon in solution - 1095 if pushed can be the worst as far as "plate martensite" or discs that don't have a fingerlike grip on each other and easy breakage. Pushing carbon levels higher helps the carbon stay sequestered in the carbides and you can fiddle faff with quenches to decide if you want to release some of it or something gentle and get lower hardness. The iron carbides (like in a file) add to the feeling of bite at the edge, but they also make it easier to get a really thin shaving for reasons I don't really understand.

I believe 80crv2 is probably what pfeil is using. it's a little harder to heat treat than O1 even though it's often described as not that -but the difference is that it needs to be pushed to get to a hardness you'd like in a chisel, and like 52100 if it's not pushed up, it has too much toughness, and edges roll and won't let go. No 19th century chisels that i've used will do that - they seemed to have a good idea of what's nice in a chisel and I seriously doubt that was by chance. That continued through the hand ground era in English chisels probably into the 50s or 1960 in some places.

80crv2 if it's really pushed will be 61/62 with a 400F temper, and I haven't hardness tested any pfiel carving tools but they have the benefit of not relying on a set angle for us - and they may be a touch harder. I've always liked their chisels a little, but they do have that buttery smooth feeling on the stones that you get with 80crv2.

Which leads back to the point of this thread, I think if people made simple chisels out of a good file and managed to tinker a little with snapping some samples, they would be shocked at the quality of their crude shop tools. the feel of a file or 26c3 fully tempered at 64 is that of white steel, and different than run of the mill chisels and when malleting or paring gets a little rough, they tolerate more rungs on the ladder.

if 52100 is really pushed hard until it's too hard to retain most of its toughness, it probably will make a decent chisel.

Comparative terms here for like tools, I made big chisels out of 80crv2 and really had to push it to get it where I wanted hardness to be. I then made the same chisel out of W2 and the W2 is just sort of the first attempt in a big propane forge - and it's better than my best effort of 80crv2. it's got enough carbon to have more bite.

By the way, if you want me to send a single chisel to fiddle with, let me know - it may not be what you like - warren doesn't care for 26c3 compared to a touch softer cast steel, and so I gave him a 26c3 chisel that I think was ground a little harshly and is probably closer to 61/62 and it starts to exhibit toughness there. they are the ones on my front page. First thin chisels, always the ones where you figure out what you have to do better and since then no issue. But warren said "I couldn't tell when I got them sharp on the stones, but they worked well in production work". he's picky!!! I know what he meant, though - the really good English chisels are pushed past the point where toughness is a feature and shy of being brittle. Every one of these (files, 80crv2 - though it's not a gimmie because it's lower carbon and has some surplus chromium to bind carbon in carbides more than 1084) is a good garage man's steel with

[D.W.]

61.5 for a pfeil carving tool. I would guess it's 80crv2 heat treated in a salt bath at an upper normalization and austenitizing range and then tempered Somewhere 375-400F.

they would never tell anyone what they're doing but they're not using expensive steel and their ad copy is sort of from the old days where everyone used "special secret steel". It's a boring truth that inexpensive stuff makes great chisels.

I've ordered 52100 round from admiral or jantz today just to give it one more go on chisels by pushing it like I did with gravers (68/69 out of the quench - not how it's supposed to be treated). I don't like it in plane irons, but maybe it can be pushed hard to get carbon in solution (as I did to make stitching tools for rasps) and end up with toughness about like we are used to in chisels. I have a huge distaste for it around 60 hardness, as you're aware. It would make a terrible carving tool at that, or bench chisel.

Forging it will put all of the carbon in solution to the extent possible, too, which may help things. I recall Carlton (?) cutlery putting up a video saying they were making 66 hardness thin paring knives with it and they just didn't care for it in bar form. I don't know what they do or don't know, but it's often delivered coarse spheroidized which makes it even harder to push the hardness up in a regular heat treatment cycle. forging and uncontrolled air cool and then air normalization or vermiculite anneal will prevent that spheroidizing from occuring. Larrin did a good article on dealing with heat treating different brands of 52100 - same alloy - enormous disparity in initial hardness with the same heat treatment schedule. Something like 59 at the low end (where do you go tempering when you're starting at 59? might as well just assign that result to machetes) and 67 at the top. this game of really pushing carbon in solution is won by hand heat treatment, though - the soak in a furnace creates too many problems because it's a soak and not short term. I think in another 10 years, what I'm doing won't be such a forbidden process and will be scientifically recognized. Carlton cutlery was bettering peters heat treatment in mules, but he definitely won't tell people what he's doing...I'm pretty sure i'm doing the same thing).

[D.W.]

Tested one other straight pfeil carving tool. When I looked at it, it has some rolling at the edge.

the hardness tester says it's struggling to get to 60. almost humorous that it's showing rolling and denting. that 1.5 means a lot and is outside of the range that an accurate hand hardener would do, but it could be a different batch. one of my other gouges has a dent, so I must've tested it and put it on the hardness test list on my page. Just found my google sheet - that gouge is also 61.5. no visible ugliness at the edge.

62 is my personal lower limit as a maker, but no such condition for quick one use tools or purchased tools. i'd say 59 and below that I start to get annoyed in hardwood.