Month: September 2023

Forging chisels by drawing out round rod poses a problem.  I can’t use 26c3, and the steels that I can use are 1% carbon or less. They make a decent chisel,  but I’m not trying to come up with a routinely make able decent chisel. I’m trying to come up with something demonstrably better. 26c3 is better than any production chisel and it sharpens easier than hard 01. Yes, I’m mourning the lack of it as well as the inability to get defectless 115crv3.

I don’t care for 52100. It makes for a very tough chisel at 61. People talk about a tough edge, but mean strong. Toughness leads to a persistent foil at the edge of a chisel, and a deflection with a foil is a bung to push through wood.

However, I’ve had luck pushing up the hardness of 52100 to make gravers, and the tables suggest it should be no tougher than 26c3 assuming 52100 is pushed up into the upper ranges. Double tempered gravers were landing around 64 hardness with no sign of grain bloated and the tips would chip off, which sounds bad but it’s better than seeing them bend, because they would bend long before they break from chipping. Or said differently, there’s more potential to resist edge distortion, a higher point of abuse where things finally fail, if they fail due to chipping and not folding. Why didn’t I do this before? Simple – I wasn’t able to get hardness out of the quench high enough to do it, and most other folks never will, either. 52100’s behavior is a bit of an odd duck and the heat treatment routine needs to be just a little different.

Because I’ve learned this just by experimenting – which is like getting something for free – it seems worth reexamining now that I’m comfortable with brine quenching. 64 hardness after a 400f temper should be attainable. More than that is unpublished more or less, but here we are. I am pushing method experiments to see what is possible, not just what’s possible with a furnace schedule and cryo.

This is the above chisel after a long double temper at 400. You can see the chisel between the anvil and cone at the left. 400 in a .9%+ carbon steel is usually an area of sweetness. In case you can’t see the dial, it reads a strong 65 hardness. Further up the chisel where you may be working with it in 2054, it’s still 63.

I’m pleased with this, the chisel is miles better than the 61 hardness 52100 steel chisel I have on my rack, but even though the alloy is pretty plain, it does start to resist some sharpening stones at this hardness.

I have to test it more. And see if bar 26c3 will also easily reach 65 after a brine quench. I’m fairly sure it will. And I think 26c3 will have a dry razor like bite and still be a little better than 52100 – certainly more agreeable on natural stones at the same hardness.

Who knew that making gravers would lead to improving some of the chisels? That’s the free part – I made gravers, they worked, done, right? No, don’t miss opportunities like this and just write things off without proving they won’t work. it would’ve been easy to say, well, you can do that with gravers because they’re little square stubs, but a chisel will break or crack or warp with the same routine.

It’s very nice to finally have a bolster that’s formed from the same piece of steel and have a tang thicker than a quarter inch. I’ve not yet had a forge welded bolster break loose – I don’t think more than 1 in 100 would ever let go, but all one piece of steel is even better.

Amber varnish on the handle again, of course. No metal driers, so uv light is needed unless a cure time of weeks is allowed. No thanks. Curing I’m the sun is fast, but Steve Voigt motivated me to make a black light box and the cure may be a bit slower, but weather matters not and there are no bugs stuck to the handle.

Oh, and I’ve already tested the edge malleting hard maple. It doesn’t chip, so more evaluation is needed. Even at 65, there is persistence to holding a little bit of a burr when sharpening, which is a surprise.

The heat treat process is not complicated and does not require a furnace, though it has become easier to really manipulate the steel with an induction forge vs. propane. If you want to normalize the steel with a heat all the way to the point that there is scale forming and then let it air cool, it’s 20 seconds with an induction forge, and follow up thermal cycles to shrink grain are a fraction of that.

OK – there’s been some delay in getting this wrapped up. I finished the chisel, but the chisel is a different chisel. The first two parts are unchanged, but charging forward and trying to find something to make a single piece chisel including integral bolster just happened by chance to run into trying a steel that is defective. The 115crv3 bar that I was using is defective. I relayed my experience to the supplier of the steel and they gave me a refund and canceled a back order that I had out there. Phew. So, I went back to W1 bar, and the last post here is a chisel made from W1.

This process changes none for any steel that I use other than the heat treatment, and I think it’s not worth going into specifics about heat treatment.

The W1 replacement looks the same – but I forgot to take a picture of it flat out of heat treatment. I thought this was the chisel, but this is yet another one after, but pretend it’s the W1 chisel. Flat and wedge shaped in profile as far as thickness goes:

I do almost all of the shaping and finishing work on less than precise grinders because none of it is jigged.

Actually, you can see the three things I use. Any of them could do all of the work. For example, on the left is a simple “bucktool” 4×36 sander. That brand is a low cost direct drive brand that you can find on amazon, but the sanders are not all the same. this one has a steel platen bed with some rigidity and a graphite coating that’s infused into woven cloth. it’s the best graphite platen material I’ve seen in terms of durability.

You need a sort of flat area somewhere on something, and an idler.

Which brings up the two grinders. One is back right. the spindle sander is for electric guitars, so you can ignore it.

Both grinders are identical, and they have the cheap multitool attachments. one is 4×48 and the other is 2×48 with a larger contact wheel. The two separate sets of units are some fraction of the cost of a “good” 2×72 grinder. The multitool attachments are not high quality or precise, but I haven’t broken either of them. So there’s a lukewarm endorsement.

The grinders are the Jet IBG-1, which is only important if you consider the power level important. They’re hogs for an 8″ grinder (11.5 amps and strong), which makes them a good candidate. Unfortunately, Jet has decided to almost double the price of them in the last two years at street level (they were $260 each when I got them, and now $470 seems to be the norm). I’m sure there was a “shortage” vs. just money grabbing. Or maybe they’re in the club George Carlin talks about and we’re not. Whatever the case….

I figured I’d show what I use – I’ve made entire chisels just with the 4×36 belt sander and if I did it here, you wouldn’t see an aesthetic difference. It would requirement to hand file the neck/tang area of the chisel, which I did for a long time. To put that in context – forge welding the bolster was later, so you could profile the tang mostly with the idler, weld on the bolster and then hand file to clean up. The bolster is now there out of forging so I do the rough grinding with the wheel. The wheel is 24 grit. Strangely, despite alumina prices being about the same as they were two years ago, these have also increased 72% in price. and when you do heavy grinding, you will actually see consumption of wheels that you’d never have considered before. One because you’re doing lean in grinding and two, the metal up to the point of the picture above is unhardened.

But that’s the past – we heat treated now, and heat treatment should leave the tang below the bolster semi-hardened so that it can be filed. The tang after the bolster inside the handle is unhardened.

Phew, a lot of background here – but it will feed into the discussion forward. I think you do not want to consider trying to jig anything you don’t need to jig. It’s limiting to do that and then the work is also boring.

On to the Finish Grinding and Bevels

The chisel is generally still wedge shaped and not warpy at this point, but if there is a slight lateral warp, observe it so that you don’t forget about it. Visual warps can be a matter of a couple of thousandths and they’re seconds of work to deal with. I leave chisels ever so slightly wide if the width really matters down to a couple of thousandths and then grind the sides to width after heat treat so that they’re straight with a very slight width taper. it makes for a nicer chisel to use than parallel sides.

At this point, I will grind the top curvature into the chisel. that is, the bottom is flat, the top shoulder is thick and the first several inches of the chisel are thinner, so the natural solution to this is some curvature in the taper as you approach the shoulder. It again makes a better chisel, but it also looks more interesting – a nice combination. On a short radius contact wheel it would seem hard to do this because you can create a lumpy surface. Working in slight diagonals both ways will eliminate most of this and if you see a spot that needs more ground off, of course you just do it by eye until it looks about right.

I don’t have a picture of grinding the bevels, but the biggest reason I got a picture of the idler wheel at the end of the belt on the Jet grinder is that is where I grind the bevels on a chisel.

Linearly. As in, I feed the chisel straight up and down with the length of the chisel in the direction the belt travels, not across the idler wheel’s width. That decreases the contact area and slows the process down.

It takes about five minutes to grind both bevels with a coarse belt, and to do it reasonably neatly. If you could do it in 15 seconds per side, it would be hard to do accurately. There is a dip bucket below the idler both to catch swarf and lower the amount of dust in the air. A couple of passes with the contact idler and into the bucket the chisel goes, but it never gets hot enough anywhere that I can’t touch it. This small point of contact on the idler with coarse grit is important – we don’t want to just not exceed 400F, we don’t want to get close to it. Once the bevels are on, heat treat isn’t a reasonable option with a water hardening steel.

Make yourself a routine. Three times up and down the contact wheel, dip in the water, look. I do both sides of the chisel progressively so that I can look down at the bevels, at the sides, and then from the front. There is no jig, so they are even by eye.

Once the bevels are on, I do the same thing with a finer belt, and then into the vise.

The chisel is held by aluminum soft jaws and I work over the ground surfaces with a sanding block with 180 or 220 grit paper until all of the machine marks are gone and any lumpiness. You can adjust lines here a little bit. The perspective makes the bevels look uneven, but they’re not.

This is the actual W1 chisel, of course. I cross file the tang both on the sides and below the bolster and lightly file off anything black on the bolster and then clean it up. The order of all of these things doesn’t matter that much, they just have to get done.

Somewhere on youtube, I have a video of cutting the final bevel on the chisel using a spray bottle and a belt sander. it takes about one or two minutes to fully cut a long shallow cutting bevel after this and do it without heating the edge. you can intermittently grind on a very coarse ceramic belt and dip often and check temperatures with your fingers, but I find a slightly finer belt and a light spray of water just dandy.

You can use a flat belt on glass or a rotary diamond plate or something to finish flattening the back of the chisel – I should mention that I often do that before adding the finished bevels for several reasons, but you’ll figure that out. At the very least, you don’t want to overheat thin edges. I started using 10″ rotary diamond plates to do the final flattening of backs post heat treat because of trouble with a heavy hand on the long sandpaper lap, blistering fingers with heat and at least once, drawing temper by accident. The lap that I use is just a $30 10 inch diamond lap off of aliexpress glued to MDF and fitted to a mandrel for the drill press (low speed, too!! High speed grinding with diamonds is a no no).

After handling the chisel, this is the result:

The handle is oiled at this point, but varnished later. Since this is the first W1 chisel of this type of make, I was hyper to see how it would hold up malleting a couple of cubic inches of maple. It did fine. I think 26c3 is a step better for harder woods, but it’s not available.

Even as I look at this picture, the tang is too long. but that’s a known thing – my grinder wheels are 8×1. I have a narrower wheel coming in the mail to do the early rough grinding and bring the shoulders up closer to the bolster. the tang is plenty heavy, though – it’s just an aesthetic issue for now.

Amber varnish on the handle brings us to a close. Shellac and oil would also be fine, or whatever else you’d prefer. .

I didn’t discuss how I make the handle. It’s separate from the chisel and it’s just a handle with the socket hole step drilled freehand. How do you get drills that like to wander to make a hole in line with the handle? Freehand. you drill a little and adjust the direction of the drill as you’re going so that the handle spins without much runout. You’re looking at the drill that drills the handle in the background. I’ve got no tolerance for trying to jig something like this up – it’s lazy, time consuming and the result is often worse. We are makers, and can trust our hands and eyes.

After the prior post, I had a chisel that needed a tang. Admittedly, I am not quite to the point where I want to hammer that tang out immediately – a little break is nice. However, I’ve done it and can do it. Which brings something to mind for me in terms of hand work. There is a version of tired or fatigued where you can take a break or set up a rhythm that I refer to in definitions on this site as “the count”. Work happens neatly and is pleasant when you can do that.

Then, there is a level of work where you push yourself and perhaps it only accomplishes a little bit more, but it ruins you for the day, or in the case of my arm and elbow, may lead to problems in the long term.

Hand work doesn’t involve that kind of thing. Hand work involves the former, as it allows you to assess what you’re doing and not be distracted by pain which leads to errors of stupidity or contemplative neglect.

Don’t do it.

I will be comfortably hammering one of these chisels out entirely in one stop soon enough – it may be chisel #12 instead of 6. That happens not just from brute strength, but neural development, efficiency and control. That combined with your brain understanding what’s easier is part of what I like to refer to as craftsman’s magic. You get better at things just by doing them, they get easier, you don’t have to become a strong back weak mind version of yourself.

On to the chisel.

Here’s where it stands after guillotine and hammering

Kind of ugly. I would prefer grain direction fan out from back to front but the tip of this chisel sort of has the grain flowing in like the toe of a shoe. I’ll grind that off. I’m sure this steel still has directional favoring as far as toughness or resistance to breaking goes.

there’s a lot left here and in time I may get close to the finished product above the shoulder, but the penalty in this case is about ten minutes of heavy grinding.

After a total of perhaps 15 minutes of grinding, I’ve arrived to this point.

this probably seems quite a leap, but I can’t really give you much advice on grinding other than making sure the general direction or flow in the roughed piece is the same as the ground part. you just have to grind and file things to learn to do it. Unless you make 100 of the same thing, you are not going to establish a mindless routine. Not that getting to that level would be bad if still doing this by hand, but I would say grind just shy of where you want to go and file or then light grind to finish. That’s what i did here.

There’s nothing special about my mark here, by the way. It’s a piece of old file that I heated, then when it was cool (unhardened), imprinted some reverse letters and put on mock serifs, and then used a checkering file to crate a postage stamp border. you can make your own if you can find the reverse stamps, and then just mutilate something to make those little serifs on the ends and bottoms of letters.

My grinder for this is a combination belt and wheel – it’s a strong jet 8″ grinder. They make at least two. This one draws 11.5 amps and it’s nice to have it on tap. You can grind something this size without ever slowing the machine down.

This area is a mess. there’s another one of the same thing with a larger narrower wheel back right, my OSS, which is typically a light use tool for guitars and obviously gets cleaned off if making them, and on the left is a 4×36 direct drive bench sander.

Dust collection or fanning metal dust is an absolute must. Wood dust is annoying. Metal dust is that squared and probably a much greater long term health threat. I have a bucket below the belt both to dip tools and to catch most of the dust so I can throw it away later in a big rusty brick. The vacuum is hooked to the dust port on this grinder and it doesn’t catch everything, but it catches a lot and throws it into a bin.

The fan sitting on the top of the heap sends all of the remaining fines and smoke out of the garage in a gentle breeze. I don’t grind with the door closed. In winter, whatever is being ground is hot enough that just having it in your hands will keep you warm. I thought that was odd until I read that it’s more effective to warm someone by warming hands than full body warming if the input energy is the same. That was something from a doctor discussing hypothermia, not something from bro science.

After some more filing I’m here, starting to file the bolster:

width doesn’t matter to me on tools, and I like a slight taper in width with the bevel edge being widest. This chisel is probably between 5/8th and 3/4ths. It’s a mule, anyway. If accurate width is important, you can leave yourself a hundredth or two of fatness and belt grind it off after heat treatment. If heat treatment distorts more than that, you’ve got bigger problems. Whatever the case, hitting something within a couple of thousandths in finished width is not difficult.

Filing the bolster is done with chisels that have been made safe edge and also had some of the corner transition to the cutting side ground off. You’ll figure out what’s right if you file a few of these.

Files are consumables, and there’s no need for anything expensive here. The round filing at the shoulder is done with cheap files, the heavier filing is done with mill files and a cheap double cut half round file, and taper saw files and chainsaw files do a good job of cleaning up the rougher work.

Just like grinding, filing the bolsters on is better learned by doing it, and not by memorizing a 14 step process. That’s nonsense you will pretty quickly realize you’re filing flat facets on work like this and that dragging the file backwards is a good idea because it prevents pinning. Again, files are consumables. Don’t trade a dollar of file wear for an hour of wasted time or ugly results.

Something important does come up here:

You can overcut things or file into the tang. You don’t want to do that. When filing the bolster, I file five strokes and look, five more and look. This happens in rhythm. It both keeps fatigue away and also allows to see things that are occurring and adjust without having to think about it. And importantly, it builds in rhythm an assessment that prevents overcutting. This is something worthwhile in all hand work that looks good if it’s done just right and terrible of done a little more than just right.

Heat treatment is next, where we could find out the whole thing has been a waste of time of we get serious warping. This will be the first thing I’ve ever heat treated out of 115crv3, but the heat treat schedule and the composition suggest that warping should be my only problem if there is any. Water hardening steels like a fast quench, though, and the potential for warping is already there. Go too gentle on the quench and the chisel won’t be full hardness- that’s even worse.

The most common thing with chisels for me is vertical movement or bow. For example, the flat back of a chisel ends up not close to that and grinding it out leaves the tip thin.

it’s once in a while, not every other, and the skill to do these water hardening skills right – especially with manual heat treatment by judgement as I’m doing – is why you don’t see alloys like this used by the tool making companies who sponsor manbun parties.

I’m going to post the process that I’m using to make a chisel from round bar. Two things about this – I’m sure it will be refined as I make more chisels, but it is already practical. Second, this is my sixth forged chisel from round bar and only one of the other five is fully finished. But, I’ve made a lot of chisels hammer tapering flat stock and then affixing a separate bolster, so those two pieces of information is useful. You may get to this point in six chisels total (i doubt it) but if it takes longer than that, I’ve paid my dues and enjoyed it rather than running from the failures and hoping to avoid them.

The point of this, as mentioned in yesterday’s post, is both to get to more forging, but more importantly, to have the chisel made from a single piece of steel. I’m making this chisel out of 115crv3, a relatively plain steel (less plain than 1084, more plain than O1, and miles more plain than A2, for example). I’m using this rod type even though it’s kind of hard to find because I want a chisel that is going to land in the 63 or so hardness range with a solid double temper of 400F, and 0.9% or 1% steel can suck wind a little bit in trying to achieve that and not be a bit chippy. This steel is 1.15-1.25% depending on the melt. I expect to lose a little bit of that as decarb from forging, but not too much due to the induction heating instead of sitting in a hot gas or coal forge. Induction allows you to focus more on the part you’re hammering and heat adjacent areas enough to avoid cracks, but not have to heat a whole blank end to end to upper forging temperatures.

So, unlike most of my very wordy posts, this is going to be more pictures and fewer words. The goal here is just for me to share what I’m doing. The goal in the future is to really get the amount of chisel made forging vs. grinding and filing increased. I’m not there yet, but have made huge strides just in six.

Here is the round bar – 20mm 115crv3 steel. I’m not fully decided yet and have ordered three more bars of the stuff with some smaller, but I think I’ve learned enough thus far to know that I don’t actually need anything smaller because drawing it out is easier than expected. That’s not to say it’s easy – it’s a little physically demanding by hand.

115crv3 alloy rod – sold in the US only in one place and apparently intended as stock for material conveying systems.

The first step is to start drawing the stock out from the bar. There’s guess work here until I’ve got more experience – as in what will yield a chisel between 5/8 and 3/4″ from the bolster down, and hopefully a little over 5″.

I leave a lot more at the tang shoulder because this chisel will have a classic taper with top curvature (eat that CNC), and then use a guillotone to start to establish the tang.

That gets me to this.

You can already see why I like doing this – the bolster does not have to be upset to greater diameter, the basic rod is there and it is high quality, so I know it won’t be brittle or weak.

This is what the guillotine looks like. It’s apparent from this making and others that I will probably want to make a guillotine anvil/bar set that will taper this more into the shoulder and perhaps leave the tang area a little shorter than the 3/4″ flat default. But those are things to learn. I’ve already ordered more 4140 bar stock to make different anvils – this will become a useful tool for a one-man operation.

The guillotine, resting. You lift the top bit and put the work between the bars and then hammer and turn.

To this point is about 15 minutes of heating and hammering. 1/3rd of 1 and 2/3rds of the other. I’m using a 4 pound hammer and am by no means a physical specimen, but I’ve not handed in my man card yet in exchange for looking in the mirror and always-clean pants and hands.

After establishing the roughed tang, I draw out the steel in about the taper that I want to have less some room for grinding and finishing off the outer layer, as well as the expectation that more will come out of the tang area. Not much needs adjusting near the business end, but the tang area is nice to leave a bit to work with as distortion there is a pain.

The last 1/4 to 1/2 inch of this will come off. it’d be nice to have left a bit more on the end, but we’ll live. The objective here is to get closer to straight top and bottom to avoid distortion in the quench later. I will grind the top curvature in, hoping that the warp in the quench doesn’t make that difficult. The actual tang thickness will be ground down quite a bit, but I will probably mark this one with my stamp so there will be some distortion on the blade below the tang from doing that.

It’s slightly wider at the shoulders than the business end, and grinding will deal with that pretty quickly.