Amateur Chisel Making

[Frosty]

I sincerely apologize John, I did not intend to single your post out to see how much Buster actually knows. Seeing it seems he's into non-functional aesthetics I'm bowing out of this thread entirely.

Frosty The Lucky.

 

[JHCC]

Two things:

1. No offense taken.
2. I don't think all of what Buster is talking about is "non-functional aesthetics". Regarding bevels on chisels, they do lessen the weight and allow a chisel to reach into, for example, the acute angles of a dovetail.

[Buster Bolster]

You are correct, John - there are aesthetic things, but the discussion of these points is to tie functional into them and try to make them as aesthetically nice at the same time. 

The seaton chest types predate factory bevels but pictures of them show that a cabinetmaker would add them to a few chisels. They're clearance for dovetails, and importantly, things like sliding dovetails on cases or even dado joints where it's nice of part of the side has some clearance to be able to work all of the way into a corner. 

The top having some curvature is a functional thing - it gives a length of a paring chisel that will not fatten immediately, and the spring of a paring chisel is important.  If the chisel is too thin, it will not stay in a cut, and if it's too thick, it will dive in one. These little nuances are why I'm always amazed when someone makes tools or violins or something without being a user of what they're making. In the old days, there would've been professionals giving feedback. 

I like to keep the side bevels fine, but not existent, and leave them a little heavier toward the back of a chisel for shoulder strength. some older chisels are like that and some aren't. A lot of modern boutique chisels have the sides ground down evenly along their length or come out to a sharp edge, which leads to weak corners at the front of a chisel. 

How much bevel there needs to be is a "i'll know it when I see it" kind of thing. If the corners are durable, then things are good. if not, then they aren't. And if the side bevel is too fat as is sometimes the case in machine ground chisels, the user is left searching for a chisel with finer sides since most people are afraid to grind the side bevels down on their chisels. 

My comment about the tang being thicker is a combination of functional and aesthetic. I'd like the thickness taper to continue to add stiffness in the tang, but it's not been an issue thus far. The business end of the wider chisels is about 0.11 or 0.12. these chisels can be malleted (I test them before calling them good by malleting rosewood or something stiffly) - the flat stock type like this (blue spruce is an example as a maker) leads to a flimsy design toward the tang and thus the paring chisels are shorter and the user of the chisel is told not to ever strike them. As an amateur user comparing old tools to new, I think that's a shame. A ward paring chisel from the 1800s looks delicate but suffers nothing from striking. their tangs on a 7/8" parer tend to continue to taper - the balance is nicer. 

the older chisels were balanced so you could keep one hand on the handle and one on the mallet and easily place the chisel with some experience - no time wasting or arthritis inducing stopping and starting mid work to move hands around to place a chisel. It's hard to get much done like that and it's also not very physically pleasing. good hand work is productive and physically pleasing at the same time. 

[Buster Bolster]

The aesthetic discussion one is interesting, by the way. I think when the chisels are first viewed, it looks like they have a bias for aesthetics, but not one that's over the top. 

I think these are the same set - there aren't too many people I would make parers for, but this set was for George Wilson. The top two are rejects, so I keep those for me, they're handled and not finished as well. 

George was the toolmaker at Williamsburg and is a supremely fine maker. He doesn't always try to make everything perfect. Eons ago, i was wallowing in bouncing from making one thing to another and it was fine, but George saw a saw that I'd posted and aesthetically, it was OK. George called and said I think you can improve this and here's how. I'd never talked to him before and didn't like being saddled with expectations that I was pretty sure I could prove him wrong...that I would try what he suggested and fail. 

I'm not a gifted maker, so I tend to have to experiment and refine rather than second one is as good as the 100th type of thing. A couple of years ago, George asked what I was doing and I said I had been making chisels. He wanted to see them. I hid that I was making them from him because I want a mentor who tells me how to improve on my own, and not one who will necessarily tell me how to do things. George is that person, so there's no reason for me to bounce things off of him unless I get stuck. He asked if I would make him a set of paring chisels, which he could easily make better than I can, but it was a nice gesture, so I made him these chisels. 

There are a couple of stray marks in them and the bolsters are filed OK but they could have bene turned more toward jewelry status if I wanted to spend twice as much time on each chisel. I don't ever really want to do that and the marks are not much different from what you'll find on old tools - a mark here or there that didn't get totally ground out. He called when he got the box, and he immediately pointed out that the mark or two and I pointed out that I did not jewel the bolsters by spending a lot of time trying to make them look like they were practically polished but still crisp. I really thought about removing the marks just because of who I was sending them to, but I decided I both didn't want to spend the time and George of all people wouldn't expect them to be "fakely perfect". 

Immediately after he said he saw the remarks and I said "yeah, I left them". he said "I wouldn't have removed them either, and a 19th century maker wouldn't make them any more neatly."

I gain experience and don't seek much advice, and haven't for a while - though trading chisels to professional users does have a benefit - if they don't like something, I solicit them to find out what it is. That's the kind of thing that's good advice. 

The aesthetics comment from George was important, but i'm a big boy enough now that if he disagreed about a few stray tiny marks in places that don't matter, I'd still be fine. I owe changing my mindset about trying to make things a little better each time and making a lot of a few different things rather than just making a few each of a lot of different things. Some people may be talented enough to do that well, but i'm not. I hope that my 400th chisel will be better and faster to make than the 100th was. 

I don't know if people know who Peter Ross is on this board, but the only reason I have a clue who he is is because he was at Williamsburg at the same time George was there and he's one of the "real" blacksmiths who can make good chisels but doesn't seem to be interested - at least according to a few folks who are willing to take tools from a garage maker like me.  

 

John - I forgot you also mentioned violins. George was the instrument maker at Williamsburg before making tools. Not repeating anything he hasn't said - after some fraction of two decades having to make violins and other instruments in front of a question-asking public, I guess the toolmaker was a trade more or less to make things for the other trades being demonstrated there and didn't require being patron facing at the museum. 

You probably won't recognize his name, but if you did, that answers that question. 

[Frosty]

57 minutes ago, JHCC said:

Two things:

1. No offense taken.
2. I don't think all of what Buster is talking about is "non-functional aesthetics". Regarding bevels on chisels, they do lessen the weight and allow a chisel to reach into, for example, the acute angles of a dovetail.

True John, the in depth metallurgical and heat treating is valuable info, I'm sure Patrick hasn't covered it all. Agreed again, there are good reasons to grind relief into any cutting tool, whether it's near the edge or way back at the handle. Weight doesn't require pages of detailed discussion, fluting or drilling holes like aircraft metal work lessens weight with no loss of strength or rigidity in either case. 

Perhaps "non-functional aesthetics" was a little harsh but with so much emphasis on aesthetics as a goal I couldn't not point out the superfluous nature of that as a goal for functional tools. 

Frosty The Lucky.

[Buster Bolster]

I think there may be some cases where aesthetics can't be handled at the same time as the functional parts. On chisels, I have some aesthetic wants, but none of them are tuxedo t-shirt in nature where they don't serve a real purpose. Put differently, I'll chase aesthetics when they can be done along with the functional, and sometimes they can't and I don't care to chase them then. 

the heat treat stuff has an off shot. I've got methods to figure out how to do heat treat hand and eye with simpler steels, how to control grain size, and dial in hardness. Someone told me they snapped samples in the old days to grade batches, so I started doing that and taking pictures - once you are consistent, everything looks relatively small.

If that shows up, it's W1 magnified at 50x, or to scale it, the sample is .14" thick. Since I'm just starting to draw out and forge chisels, there's some hesitance on my part as to whether or not I can get the grain back and get decent toughness. Since some of the chisels draw out longer than I need them, it gives an opportunity to use the cut off end length to test things. to get a handle on this takes a $15 handheld scope and about a 10 minute process to go from forging to normalizing to thermal cycles to heat and quench. If I had three samples, I'd intentionally overshoot them to see how much grain growth there is, what it looks like to get it and then that's sort of the end of it. 

Patterns of behavior emerge and then you don't have to "remember how to do 10 different steels 10 different ways" but rather you find some to lump together. 1095 and W1 and W2 (no surprise) seem to like about the same thing. 26c3 is slightly more tolerant of pushing the heat briefly before quench and 1084 is less. 

I think this is very very easy to learn to do, but it's limiting to some extent. 

At any rate, a little diligence with these quick trials answers a question for now - I *really* like drawing the steel out and forging these chisels instead of just tapering flat stock and forging on a bolster. It'd be a real shame if in trials against my other chisels and known standards if they don't perform as well. I can't live with that if it happens - as in life with it and ignore it.

don't worry - I'm running out of new topical material to post long posts about. I'll lay around in the long grass in case anyone wants to get really good at heat treating hand and eye specifically because it either fits their workflow like it does for me, or if they really enjoy it and want the challenge. And that's before saying there could be three other people on here 10 times better at it than I am and I just don't know it yet because I'm new. 

this sample tested 68 hardness out of a brine quench, by the way. Another little learned thing after starting to forge. No round 26c3 forces me to look at other steels and W1 really seems to like brine better (I thought brine was kind of a mortal sin, but haven't cracked anything chisel shaped yet - I'd be much less confident about a thin tall knife), so I'm learning things that I didn't expect to learn.  

[Goods]

From the blacksmith standpoint. I’m not sure I would move to round stock for the chisels. As you’re drawing the tang, with solid blows and dressing, you should be able to gain most of the thickness you need. Obviously, it’s something you have to work out, but in the end it would be a lot less work. Of course, with some precise hammer work, from round stock you would be able to isolate mass for your bolster, and avoid the forge welding. You do seem to have a good handle on the forge welding though!

By the way I’ve enjoyed your posts. I enjoy tool making and you making some really nice tools!

Currently, all of my heat treatment is done by eye and testing, but I’m not getting results I could with some steels. For example, I’m just not real comfortable with the soaking time need with 52100 (which I have readily available). I’m always worried about overheating and grain growth. Seem like a petty fine line…

Keep it fun,

David

[Buster Bolster]

Hi, David - I haven't been accurate enough in the past to get the tang drawn to greater thickness by hammering, but it's been quite some time and it's possible I could do it now. the other thing that's popped up is 125cr1 showed up at New Jersey Steel Baron, and it should be similar to 26c3, but I can buy it in 3/8 bar stock, and that's fine for anything I'd make tang wise. I think in the past, I didn't want to spread out the rest of the chisel vs. just tapering lower half or a little more of the chisel toward the bevel, but I'd be fine hammering the whole length to thickness from thicker flat stock now - which just leaves the bolster still being forge welded. obviously, it's one of two things. On a chisel with a ferrule, it's stopping the handle but if it were to fail and stretch down the tang a little but, the brass tube would stop the tang from going in further at some point. 

But on the earlier styles, it's the only thing that keeps the tang from eventually splitting the handle. Point with that being I leave the bolster a bit big out of caution, I guess, but I haven't ever proven that's needed and if a bolster has failed, it fails to forge weld. I've never had one break after the weld is good, but if it's formed on and the weld doesn't take, they're no good - and it's apparent immediately. 

Forging the chisel out of round stock alleviates that I guillotine the bar in front of and behind the bolster section and then grind and file and the bolster can be thinner, which looks nicer. 

52100 is a bit of a trick. I can go into this on another post sometime or perhaps make a video of doing it, but I think a temperature overshot for a short duration is a better bet than trying to soak it. I am about to forge 52100 bar. I'll snap samples and let you know what I can get, but it was the one steel that vexed me when I first started - it tolerates more short duration high heat than you'd expect and needs it to hit high hardness. it's this process of testing how high and for how long to get grain to bloat that gets you where you're comfortable not bloating the grain but getting high hardness. Little short bits of it are nice for this, of course, because you can manipulate heat level with them easier and they're quick to cut off of a bar or whatever else, hammer out - whatever you want, and then easy to break in a vise. If 52100 is hard to break after quench, really at all, it's got more room to go before things turn sideways. 

80crv2 is a little like this - it really needs a decent dose whereas 1084 tolerates almost nothing and bloats quickly - like in 10-15 seconds of overheating. doing the exact same thing with 26c3 results in a superb chisel at high hardness, and with thermal cycles preceding the quench, snapped grain is half the size of a commercial vintage file cross section. I think 26c3 is ideal for this method - it responds well but the carbides block grain growth a little bit. 52100 is harder because you really have to push it more to get the carbon out into solution to get high hardness. it's just torture to use in woodworking tools if it retains a lot of toughness and holds on to a deflecting edge. And when it's got subpar hardness or issues with retained austenite, the edge just deflects easily. 

For gravers this past year (to stitch rasp teeth), I saw 69 hardness out of the quench with 52100 and 64 after a solid double temper at 390F. it behaves like a less tough steel when it's successfully done that way, the fine edge taking is better and it won't roll. Rolling is toxic in woodworking tools. Even if it's failure by rolling. 

I think it would be very hard to nail with a longer duration soak, even using a muffle and thermocouple. 

this method I'm talking about was probably done regularly 175 years ago, but it's frowned upon now. I just fail to see the reason as once you get the process down for finding the bloat point and doing some samples inside of it, the whole process takes a total of about an hour and you need to do it once in a lifetime for each steel.

[Buster Bolster]

This is just a short (but long due to pictures) follow up after getting 52100 round bar. I get it that 52100 varies a lot in bar form depending on the condition it arrives in. If it's coarse spheroidized, then a fair bit has to be done with it to switch the microstructure to something that will hit high hardness on a regular furnace cycle. I don't have a furnace, and screw around with hand/eye type adjustments to get good hardness without bloating grain. 

The only round bar I found was from Jantz, which isn't so consumer friendly on small orders ($60 of steel, $40 of shipping), but I can't find much in round bar. It seems like some time ago, it was easier to find round bar 52100 and the flat bar was being rolled by suppliers to cater to the knife folks, and it didn't lend itself to just a quick normalizing and then a sort of regular feel treatment (or furnace schedule). 

Long story short, I forged a chisel out of this bar, and it came out an inch long, which gave me a piece to cycle. Quench hardness after forging, normalizing, and thermal cycles for a .15" thick sample is 68. The hardness has to be there before temper to make a good chisel. The rest of the nits here can't really be addressed in terms of are they going to affect performance without just making a chisel and testing performance. Testing chisel performance is pretty easy, though - a legitimate woodworker uses a bench chisel pretty much straight in either pushing or malleting. I pick wood harder than would typically used and a mallet that's 50% heavier than I'd use - if the result is good (tired by the time it's time to touch up sharpening, and touch up a brief 30 second to 1 minute process), then the chisel should be fine in use. 

that's a cross section about 0.15" thick or so. The bright dots are unusual - large grain is either uniformly all that, and I haven't seen something like this in a snapped sample. 

A closer picture just to confirm that the grain isn't big:

I'd guess those spots have something to do with chromium, but they still are strange to see - why wouldn't it be more uniform? I'll have to cut a sliver off of the round bar and heat treat that and break it to confirm that's in the steel, and not something occurring at my hand. 

the hand held scope is pretty easily fooled by bright things that are grainy into making them look like one shiny bit, so I set the sample in a better (but less flexible) metallurgical scope.

 

The picture is bigger just due to the resolution of the scope. This is steel at the break. I don't have etching solution and don't want to go that far - the round will either make a good chisel relatively easily or it won't, and if it doesn't, I'll move on. 

The size of the large splotch in the middle is about 10 microns. fortunately, it appears to be an agglomeration of things and not a big solid defect. 

To compare this 52100, here is a picture of steel that was bar and that I'd normalized, cycled and hardened using an induction forge to make gravers to stitch rasp teeth. These came out of the quench 68/69. 

Compare this to picture 2. The last sample is buderus 52100. I have no idea where the jantz stuff comes from. There's nothing involved forge wise for the last picture - it was just quarter flat stock cut into gravers, so the treatment of it was all thermal. 

Since I don't have a furnace, I can't answer this question, but maybe someone else can - if that's chromium in those splotches that's poorly dispersed, is a regular furnace normalization actually going to disperse that evenly and make it disappear? I think this is a hard question to answer unless you've seen it and done it. 

Reading this makes it look tedious, but I'm already heat treating by hand and eye. if there is a problem with the steel and I am heat treating by hand and eye, I'm sure everyone will immediately assign it to the heat treating. It's usually the steel in my case if there's a problem. chisels don't benefit much from alloying because they don't wear due to abrasion, they wear due to distortion of the edge in use. So far, the best chisels I've made in use are 64 hardness 26c3. They seem to hold together practically forever. here's what 26c3 looks like snapped, same quench hardness, and holds maybe some little bit more after tempering and though 52100 is "more hardenable" getting 26c3 right is a lot easier. If it sold in round bar, I wouldn't be talking about other steels for chisels. 

 

[Buster Bolster]

A sliver sawn off of the rod and hardened without any forging looks much the same as far as the shiny bits.

So far, I've had 50 or 75 flats with only one showing irregularities.  Round bars show some issue with 2 of 4, but it's no issue if the chisel works well.  68 quench hardness is pretty solid given no cryo. Quench tail was at zero F, and quick after the initial quench based on larrin's discussion of cold finishes.  That is, freezer makes a measurable difference,  but it has to be immediate whereas a delay before nitrogen doesn't reduce the hardness increase much. 

[Goods]

If those spots are from concentrations of chromium, I don’t think you could really do anything about it. Chromium atoms are very large compared to the iron and carbon atoms. They are not going to migrate much if at all out side of a molten state.

Maybe one of the metallurgists one here will give better insight?

Keep it fun,

David

[Buster Bolster]

I think reasonably, they're not something for me to solve, either. I gather in the old days, the steel would've started with excess carbon or been carbonized to deal with heavy forging, but I'm also working by hand so that heavy heavy forging isn't going to happen, anyway. Drawing out and hammering to width and thickness is enough. 

I'd be curious if anyone knows enough about how the rod vs. the flats are processed given my sample is fairly small - four or five rod purchases and a bunch of flats, I've only had one purchase of flat stock that was really equivalent to this. 

So far, O1 and W1 have looked fine in rod, and 115crv3 (tons of graphite or something and wouldn't harden) and this 52100 haven't looked as good. The 115crv3 was no good and the supplier gave me a refund. the 52100 could be inconsequential once I get a few chisels made and handled and tested. if it's just cosmetic, I don't really care. it's problematic in a plane blade only because allying differences that aren't fully dissolved will actually leave different marks on wood - like little lines of a different sheen, probably different wear rates, too. That's not a supposition, but actual experience from the 1095 sample I think I showed earlier - I thought the edge was chipped but it was just the bands of what looked again in that piece, like chromium, leaving little faint stripes on planed wood. In a chisel, it may be no matter. 

I wonder if the rod is drawn or rolled in a way that the refinement isn't as good as flat bar.  

[Goods]

I’ve not heard of steal bar or round stock made any other way that rolling. Some thin bar stock I s sheared from larger rolls sheets, but drawing dies for high carbon/alloy steal wouldn’t last long.

There are a couple guys here who have worked in a steal mill. There input would be interesting!

Keep it fun,

David

[George N. M.]

Back in the '60s I worked at the Alloy Bar Mill at US Steel's South Works in South Chicago.  At that time we rolled round, square, and rectangular cross section bars.  Since that time the processes may have changed, e.g. continous casting which is how most rebar is made these days.  I would think that dimension steel would have to meet more stringent standards for variation along the length of a bar than rebar which only needs to meet certain tension and shear standards.  That said, I have come across some mild steel bars which seem to vary along their length.

"By hammer and hand all arts do stand."

[Buster Bolster]

Thanks, guys. that's what I thought. The only thing I can recall seeing on video is square ingots heated and run through rollers, and maybe the comments mentioning descaling somewhere in the apparatus that I couldn't see, but I'd imagine on high quality bar, descaling is part of the process and then some kind of surface treatment like pickling. 

maybe the quality difference so far is just by chance. 

[Buster Bolster]

Same procedure for a chisel in process with 52100 bar and I think I've mastered getting the retained austenite out of it so it acts like a more plain steel in a chisel (less tough, but more strong as far as edge stability goes). 68.5 hardness out of a brine quench and quickly into the freezer, and that being after thermal cycles to shrink grain. For clarity, the hardness test (diamond cone 150kg type, not superficial) is done after resting in the freezer. I'm not looking for industrial warp prevention type measures that sacrifice a little hardness. 

Hopefully all of it was successful, but the side by side test in wood vs. another similar chisel that's known very good will answer that. 

I realize there are probably a lot more people making knives out of 52100 who are having their eyes cross with the idea of brine quenching, but chisels are a narrower and thicker cross section and it's a better match than knives. Water without even the salt addition will crack the thinner modern plane irons, especially near the keyhole in a stanley plane iron. BTDT.

I do better with 52100 if I give it an actual quench before the final quench. I can't think of anything else like that. It has come behaviors that W1 and 26c3 and others don't have. 

With some luck, a double temper at 400F (long duration) will yield 63 hardness, which is cryo territory on larrin's chart. Cryo steels are generally well thought of in woodworking because there is an improvement in edge stability. they would break bending more easily, but there are specific designs for chisles (mortise chisels and registered chisel types with a fat cross section and short fat tang) that were intended for that.  

[George N. M.]

IIRC descaling was done with water jets and we also used a BIG (about 10'x30') vat of hydrochloric acid as a pickle.  We were VERY careful when lowering a bundle of steel into the tank with an overhead crane.  I don't recall a neutralization process after pickling but these memories are 57 years old and imperfect.

"By hammer and hand all arts do stand."

[Buster Bolster]

Thanks - i did see a lot of spray and steam as the material went between rollers. I wondered if it was for cooling the scale and blasting it off or if it was for temperature control of the rod. 

I also wonder if the speed and force of the rolling could maybe keep more heat in the steel than you'd want - or in crease it. 

As a side note, after tempering the chisel above, I'd forgotten about the ability of 52100 to buffer a bit of the initial effect of tempering. As in, you temper something with iron and carbon it it and just a little manganese and trace of chromium (like 26c3 or 1095), it starts to drop hardness. 

If you compare something that converts some martensite at low tempering temperatures - like AEB-L, the hardness sort of holds steady and then drops. I'm guessing that the fall off of 52100 is less. 

why?

the chisel after a 400F double temper remains at 64.5 hardness. Brine is a treat - even as the cross section thickens well up the chisel, the hardness is 63.