Month: July 2022

It’s not that important as it’s not a steel I’ll use for much of anything, but solving the grain growth and proving it is something I wanted to follow up on. I may be repeating something from yesterday’s blog post, but you may recall that I made some samples of 1084 and one that I intentionally overheated but only for a short time showed drastic grain growth.

This is a picture of that sample – for scale, 0.1″ thick.

I can tell what made it grow like this, but the loop isn’t closed until I can set up a process that will shrink the grain regardless of where it starts, and I would at least like it to match the best results with a low temperature heat and quick quench. A furnace may do all kinds of things that improve results greater than grain growth, but I’m only concerned with whether or not the steel is good enough for woodworking and if the cycle can be done easily.

The problem was temperature, though – what 26c3 (1.25% carbon with a small amount of chromium) likes and doesn’t suffer from at all is terrible for 1084. it’s just a little too hot, but only a little.

That means modifying the thermal treatment cycles a little and heating a little bit less upon quench, and then the results with the exact same bloated grain tab above moves back to this on the first try:

This is visually similar to what it would look like before overheating at all. I’m pleased with this. The change in what I typically would do is very small.

if I have a use for 1084, at some point, i will have samples tested, but right now, the need doesn’t go past good hardness and toughness “good enough”.

Good enough means not chipping in a plane a hardness above 60.

I’ve learned another lesson – don’t ask any questions about heat treatment on a knife forum unless you’re going to nod and say yes when half of the group says “sell everything you’ve got, you’re not getting good results no matter what, and you won’t until you buy a furnace”. But it’s hard to criticize that, how would someone making knives know what works well in a chisel. One is biased toward toughness first, the other toward appropriate hardness (strength). After all, they’re selling boutique knives that people are going to hit with clubs to split wood. We will probably pick up an axe or a froe.

Just how easy this issue (low toughness in my 1084 samples) was to address, first in preventing growth and then in reversing it without a difficult, expensive or time-intensive process does egg on my desire to solve problems rather than only go for the book solution.

If you want to sell things (including yourself), you’ve got to go with the flow. I have no idea why one-dimensional answers have to apply when someone isn’t doing that, though.

I’ve been making blades out of sharon 50-100 all week – one a day as it’s something I can do in less than an hour, and the ability to use a plane and then examine the edge under a microscope to see what I have is just about the ultimate initial test. If the iron is good, it will sharpen easily and well and wear evenly without chipping and folding.

If it passes that, then it’s time to use it rougher wood. The dilemma in this case is that it’s an antiquated steel by now and the found lot is 0.145″ thick. So, I can make infill irons with it and maybe large moulding irons.

Making a plane iron in a shop without machine tools and then quenching and tempering – especially a water hardening steel that’s just mill finish to begin with – means flattening an iron and getting an initial edge. I consider the entire establishing of the bevel, flattening the back and honing both to be a 10 minute job. I’ve been doing this for a while and have gotten good at all of the steps. That reminds me, I have an improved back flattening jig to share, but I’ll post about that separately.

Forums and Assigning Fault

When you read forums, you’ll hear all kinds of suppositions. Any time someone talks about a commercial iron being chippy or microchippy, or whatever else, I always challenge them to get a hand held microscope and view the edge of the iron before they start planing. A2 is relatively notorious because Lie Nielsen recommends hand grinding it and it’s more resistant to stones than most simpler steels. If someone even manages to properly finish an edge, they’re faced with nicking an iron, perhaps, and having to hand grind out the nicking.

They have practically no chance.

Hand grinding a small nick or small nicks out of an iron means honing off several thousandths of edge length, perhaps 4 or 5 at the most if there isn’t catastrophic damage, and the idea that you’ll do this in the middle of working on something is a no-go. Most of us have calipers – I’d estimate that a brisk sharpening session on a secondary bevel takes about 1 thousandth of length off of an iron.

My point is that what you see occurring is easy to attribute to “microchippiness” of A2. This is often the accusation. However, I fully honed, examined and planed a couple of thousand feet with A2 irons and found no evidence of chipping. The edge can get ever so slightly rough when it’s absolutely dead dull, but what people are generally observing is failure to remove nicks. Not evidence that they’re a victim of a steel that has an underlying monte carlo simulation resident in it to determine when it will mercilessly let out a ball of line-leaving filth.

The failure to get a good surface or good performance of almost any decent tool is either abuse or quite often, blaming damage left in the tool on boogeymen.

Annealed 50-100

I’m looking for super bright and no defects at all on a test edge or test face of a board. This is after dry grinding a full bevel on a new plane iron with a 36 grit belt and then hand honing on an initial microbevel. This is rough treatment – 36 grit ceramic belts grind much cooler than regular belts or a wheel, but they are extremely aggressive. Those two probably go together.

Yesterday while looking at carbides in an annealed iron that’s then quickly quenched – as in, the iron is placed in vermiculite below the temperature where it could be quenched and then it’s allowed to cool slowly in a “sandwich” of pieces of metal. This does nice things to the carbide structure, hopefully making them smaller and more round.

I saw lines on my work. Just two. The annealed iron tempered a point or so softer than another iron I’d done the day before, so I was starting to guess at reasons.

The carbides looked like this.

I scrolled the iron back and forth on the microscope looking for the folded over little area of poor results, and found this.

I guess it’s hard to complain about the quality of the edge when the diagonal scratch points the finger directly back at me. The height of this picture is only about .0095″ (just under a hundredth of an inch). This garish scratch is a couple of thousandths wide, but it looks pretty spectacular here. Interestingly, the edge seems to be closing over it.

The reality is, the steel isn’t at fault here. I’d like the iron to be a little harder, but could hardly claim the edge folded. This isn’t visible with the naked eye and I’m not sure if there’s even enough there to easily catch a thumbnail.

I fit in my own suggestion here – look to the sharpening first when pointing fingers at a blade or steel and thinking that it’s the blade. In fact, I can rarely count any time other than in rough lumber or knots or silica, where edge damage occurs in regular planing.

This idea of finding the right culprit and not being lazy and attributing it to something else is necessary for solving problems. Even though this is a simple one, the trouble is you’re your own feedback loop. If you have an iron that you often see defects without checking the iron, soon your supposition becomes truth with repetition. Except it’s often not true. That becomes even less helpful when you assert that it is when attempting to help someone else having the same problem.

I’ve removed this scratch, of course. But it’s not something the average person will get out of the back of an iron with 20 extra seconds in a fine waterstone.

There’s a rumor that the market included plain steels in tools along with the idea that synthetic sharpening stones are also a new thing. Neither is true, but it is the case that prior attempts at tarting up woodworking hand tools with high speed steel (HSS), or razors with significant amounts of tungsten in them were relatively unpopular.

A lot of synthetic stones were marketed in the late 1800s and early 1900s at high cost and also were hit or miss.

High speed steel goes back at least as far as Mushet steel. Think O-1 steel, which is easier to harden thanks to a big dose of manganese, but with much more manganese to the point that if you overheat the steel, it rehardens just with exposure to air. Hardenability is a term that’s used to describe how slowly a steel can cool and still reharden, and “hot hardness” is a term to describe how well it performs when it’s hot. Air hardening high speed steels are both highly hardenable and with good hot hardness. Others, like A2, are air hardening “high hardenability” steels that don’t fare well once they’re exposed to heat.

Mushet was an early (first?) example, but the steel was brittle and what we will find in woodworking tools is more likely to be earlier tungsten alloys followed by the more common M-series, where M2 took over thanks in large part to being lower cost than tungsten high speed steels.

Mountford’s Revilo HSS Iron

Mountford was apparently a scythe or farm tool manufacturer, and at some point in the late 1800s or probably more likely, early 1900s, they marketed a high speed steel parallel iron for infill planes. You see them from time to time, but they’re not as common as Ward and Payne, for example. If I had to guess about the age of the one that I have by the font and style, I’d guess 1925-1930. An iron like this is something I might buy out of curiosity, but in this case, I bought a plane that already had the iron below as a replacement iron.

What I found interesting before getting this iron is that sometimes I would see listings for planes with a Revilo HSS type iron that was well used. Some even to the slot. When you see this, that usually means the iron was sharpenable and pleasant to use.

We get confused now with boutique offerings that are high hardness and one of the myths of HSS is that it’s always really hard. I looked up an M-2 alloy hardening and tempering schedule and it provided instructions for tempered hardness from 56 to 66 on the rockwell c scale. For someone sharpening on stones, this makes for a huge variation in what you perceive, and most amateurs wouldn’t think two irons at the extreme ends – or even middle and one end – were the same steel.

Where does his hardness myth come from? I guess it must make some kind of sense that a steel that does well cutting other steels would be really hard, but the high speed reference is related to the fact that the alloy can be used for “hot work”. Allowing work to be cut and shaped at higher speed means higher volume, more efficiency. The steel doesn’t have to be hard to do this hot work – it just needs to retain its hardness well past temperatures where cold work steels will become soft.

Moving on to the idea of consumed high speed steel older irons – I suspected that the Mountford/Revilo irons were tempered a bit soft so that they could be sharpened on typical sharpening stones, and that’s the case. Most older tools that are overhard without later correction just go unused, and the listings that I’ve found of these half or mostly consumed tips us off. I can sharpen this particular iron easily with an india stone and washita stone as a finisher.

What is the composition? I have no idea, but an XRF analysis would figure it out pretty quickly. At its early age, the hard steel lamination on the iron could be a T-series high speed steel. It looks like Mountford was in business at least until just prior to WWII, but since plane irons weren’t their main business, there’s no reason to conclude they were making these from introduction to closing business. If I ever have the chance to get XRF analysis done on a group of irons, I’ll try to remember to include this one.

What is it like to use the Iron?

Since it’s tempered fairly soft, it’s hard to tell that it’s high speed steel. If it wasn’t marked, I wouldn’t know either, until grinding it and seeing that it probably wouldn’t spark like a typical older iron. I haven’t ground the bevel on this one in a while, though, and don’t remember if that’s the case. But you can just use it like you would anything else without special grinding or sharpening considerations.

Why didn’t they ever catch on? Unless you want to heat the iron on a grinder, high speed steels in hand tools and things of the like offer no real improvement for professionals. The cost was probably also higher than typical irons, but one would have to find a listing to prove that. I have used this particular iron occasionally and would speculate that the edge life is similar to a good carbon steel iron, but to get a picture of the carbides, I paid a little bit more attention. It seems to lose sharpness and the ability to keep the plane easily at a point and fairly quickly. That is, it planes well for a while, and when it starts to fall over in sharpness, it does so quickly.

This is back to cork sniffing talk again, but I have this same experience with 52100. The edge seems to wear more in a rounded shape and less crisply and if you just keep pushing it, it’ll cut for a while, but it feels less nice to use than O-1. This iron has that, too. I think it would fare fine if it was higher hardness and hold a more “pointy” apex as it wears, but that’s just speculating and at higher hardness, craftsmen would also have liked it little. Could it be that the ability to grind it briskly with a wheel grinder was the selling point? Maybe.

A picture of the carbide pattern and the worn edge is below to illustrate what I found. In short, it does wear a little bit unevenly, and I think the lattice between the carbides lacks hardness a little bit. This is after several hundred feet of planing, though, so it doesn’t just fall on its face. Interestingly, since I work the back with a washita, it retains a haze instead of a polish and often under the microscope, you can find out why. In this case, it looks like the washita hones the lattice but some of the carbides remain in place. They don’t look large. which is good, but the edge still looks kind of ragged.

You can compare the uniformity of the worn edge with yesterday’s darling – the very plain “cold work” 50-100 alloy (1% carbon, 0.6% chromium, and some manganese plus only little bits of anything else).

Sharon Steel Worn Edge / Carbides Picture

I’d rather have a good quality traditional iron if a solid conclusion is desired. I think the market decided the same thing. Around this time or not long after, though, Norris went to R. Sorby irons, which are also soft and disappointing in the planes where they appear, so I don’t know if a crisp new Ward iron was still a possibility.

In the prior blog post, I mentioned that 1095 knives are probably not 1095 steel alloy, and implied that what’s often asked on woodworking forums “Is this old tool O-1 or A2?”. The answer to the latter is in most cases, neither.

After finding 1095 to be unlikely as a plane or chisel steel due to poor toughness at high hardness, I looked around and finally found old stock being sold of one “improved” 1095. It’s called Sharon 50-100. This series of steels is at least three or four different alloys. 1-1.1% carbon steels with some chromium added, and for a B version, a small amount of vanadium.

And for the people lurching in their seats because they would never use chrome vanadium steels, only O1 or V11 – V11 has both chromium and vanadium in it, and O-1 also chromium in it with many of the variants from high quality mills also having additive vanadium.

The 50-100 variant that I was able to find was monstrously inexpensive – think $5.50 of steel to make an infill plane parallel iron with enough left to make at least two or three kitchen knives. It’s only available in one thickness, so no stanley plane irons and no chisels with it, which is kind of a bummer as it may have made a nice change up to the high hardness 26c3 carbon steel that I like to use. Once in a while, I come across someone who doesn’t like a high hardness chisel and there’s no real reason to make a 26c3 chisel, for example, and temper it down to 61 hardness. It excels being 63 on the low side at least, and up from there several points if you desire.

So, back to the irons. 0.145″ is OK for an infill plane iron – or maybe a Lie-Nielsen 8…..a plane I don’t have.

Making the Iron

Making the first iron, the only one I’ve made, I like to see what I can feel. By my estimation, the steel is spheroidized. This means treated in a way so that the steel is very soft and the carbides have been conditioned into little round carbides vs. the elongated types usually found in rolled annealed stock. It cuts like butter, and it won’t air harden while cutting and grinding. That translates to easy working, drilling and sawing.

From bar stock to finished heat treatment is about 45 minutes. I can’t tell anything from the sample other than it is spheroidized-like softness and there’s no feel of alloying like you’d get with highly alloyed steel. By the way, you can find information about spheroidized steel and its workability but sometimes-impediementary (new word!) properties for furnace heat treaters. The way I heat treat in a forge, it makes no difference and given the choice, I like starting from spheroidized stock.

The steel still has scale on it as delivered, but that will disappear just in the making of the iron and finishing of it later.

All in all, a delight to work with. I don’t care about decarb – that is, I don’t care if the outer layer is decarburized from rolling as it’ll be ground off or honed off in short order.

The iron, along with two O-1 tapered irons recently made, shown below. these could be perfectly finished to remove the marks and eliminate evidence they were hand made, but that’s kind of prissy. There’s already too much prissy stuff in amateur woodworking and toolmaking that aims at beginners.

What I am hoping to see in this steel – the 50-100 steel – is a small array of carbides – I’ll show pictures of that in a second, as I have a method to see how they appear, how big, how many, how even. True 1095 itself has excess carbon and I would’ve expected to see iron carbides forming from anything over the eutectoid limit (0.77% carbon, or something like that). For the uninitiated, 0.77% is about the limit of carbon that can reside in a steel lattice before excess amounts start to look for places to reside. At any rate, my method to find the carbide pattern is simple – put the cap iron on the plane, use it and then take a 300x microscopic picture to see what’s not wearing away as fast.

The cap iron holds the shaving against the back of the iron and it neatly wears away a small cup in the top of the iron back. In a sense, it sands away the lattice of the metal leaving anything harder either to be broken and pulled out or standing proud. Whatever happens other than uniformity with the lattice itself, you’ll see the evidence.

Reality in practice, as I’ve found and later read about, isn’t as simple as the eutectoid limit “squeezing” excess carbon out into iron carbides. The reality is excess (beyond 0.77%) carbon can dissolve into solution and remain in the lattice. As temperatures increase in a furnace or forge, more carbon can dissolve into and reside in the lattice. Based on what Larrin Thomas has published in patreon bits, probably public now, 1095 does result in a lot of excess carbon in solution. This results in higher hardness, but lower toughness. Or at least I think it results in higher hardness as I saw an average of about 61.5 rockwell C hardness with O-1 steel and 63.1 with the basic 1095 alloy steel.

So, let’s see some lattice/carbide pictures. What follows is a comparison of 1095 and 50-100 showing that 1095 doesn’t seem to “seed” carbides, but the 50-100 steel has a nice neat even spherically shaped pattern of carbides. Yay! that’s a start. Hopefully, it will lead to a steel that’s a lot like 1095 on the stones and in the cut, but just without nicking. First, “real” plain 1095 steel:

And Sharon 50-100 steel, roughly 1095 plus 0.6% chromium:

Both pictures are taken at the same magnification – light levels are a bit different and the cap iron was set a bit close on the second picture – not recalling the first, so the wear is shorter and steeper. The carbides in the second are the smallest I’ve seen. 26c3 has much more excess carbon, but it seems like fast formation of those carbides leads to less carbon staying in the lattice and despite the formation of those, 26c3 is tough and makes a good plane iron. It doesn’t last long in a plane, though – apparently iron carbides are good for hardness, but they don’t seem to have much of an effect increasing edge life. So I love it in chisels, I like it (26c3) in plane irons, but expect most of the beginner public would have fits with needing to increase sharpening frequency a little bit, no matter how easy it is.

Looking at 26c3, it doesn’t look like there are necessarily more carbides than 50-100, just that they’re larger. They are, however, smaller than something like V11. I no longer have a V11 iron, but I do have XHP, which is probably the same thing. If that’s true, Lee Valley isn’t in danger of anyone copying them. The steel is low availability and it’s expensive. Lee Valley is nearly providing a public service by offering their V11 irons at the price that they ask. I don’t care for the chisels having tested one – I can make a better chisel, but I didn’t go down this rabbit hole to start believing that somehow a chisel that is fine for a plane iron will be the same level of “yay” for a chisel.

What’s the Wear Resistance Like for 50-100? What Else?

I haven’t tested 50-100 against O-1. I suspect it will last less long, or fewer feet. Something that an experienced user won’t care about as it sharpens really easily. I think the edge life of 50-100 is probably about the same as a vintage mathieson or ward laminated iron, and that’s fine with me.

For comparison, 52100, a ball bearing steel, has much more chromium (1.5%) and bigger carbides and lasts about as long as O-1 in a plane iron. If you’re not that famliar with steels 50-100, 52100 – yes, I know these are like calling one guy Mark and another guy Marc and then talking about how different the Mar(c)ks are.

What about sharpenability – not just ease, but how the edge comes about. Sharpenability is as good as anything I’ve seen. Beware, this is about to go full cork sniffer….. though one man’s cork sniffing is another man’s blue collar practicality. 50-100 gets a click or two less hard than 1095 – I’d estimate 60/61 hardness in the test iron, and the grain is fine and uniform with the small carbides. There’s no perceived resistance to the stones – even O1 provides some feel of abrasion resistance compared to older steels. Creating a wire edge on a fine india stone to remove wear and get to finishing an edge is effortless – a matter of several seconds following the india with a worn washita stone.

When resharpening the iron above, I worked through these steps at a leisurely pace, but not dawdling. The total time including walking over to the buffer to buff strop after the washita – 47 seconds. The wire edge after the india stone can be teased off in very few strokes on the washita and the resulting edge would show a microscopic burr but none can be felt by hand. Pure joy in simplicity and ease, especially given it’s not harder than it is. That is, really hard steels often release their wire edge a little bit more easily on a fine stone, and this iron is hard enough, but it’s not icy hardness.

And the beauty of a steel like this becomes apparent to an experienced user if the lack of chipping that I’m hoping for also materializes. That is, it looks like it may be a good candidate to be a steel that maintains a constant undamaged edge. Sharpening probably removes about a thousandth of an inch of the edge and can be done in less than a minute. Add nicking several thousandths deep, and that’s sucky. For what it’s worth, good O-1 is also pretty favorable at this whole idea – sweet to use, but not too easy to nick and not much burden to deal with unexpected nicks.

I have more experience-based work to do. Initial impressions can be misleading and I think there’s a little left in the tank to go a click harder as I worked the quench routine with a bias toward straightness rather than all out hardness chasing. This kind of experience being my change of heart with V11 after being wowed in a standardized planing test planing several 5k’s worth of board length…..and then being unwowed with the same steel as soon as conditions even went to rough lumber planing.

So, confirming that the 50-100 iron will remain defect free just with regular sharpening – something I found V11 unable to do, and the same with house-made XHP irons – is all that’s left. And since it’ll never be commercially available as replacement irons….that’s perhaps the end of this pleasant journey. Hey – do I expect to make waves with 26c3 chisels? No, I’m making them and I think they’re better than anything commercially offered, but the way I’m making them isn’t scaleable.

Pictures of the Results of the First Grind after Making

I gave the iron a quick edge, but a good one, planed a little bit and then refreshed once. I always cut the bevel on a 36 grit ceramic belt, and I had no water available, so I cut the bevel on this iron using only my palm to cool it. This isn’t like your typical sandpaper, so don’t read too much into that. It’s designed for cool metal removal and excels at that. But fetching water would’ve been smarter and faster. 2 minutes instead of 5 minutes, perhaps, to cut the full initial bevel.

The point? I doubt it ever got too warm, but the first grind goes all the way ot the edge, and a 36 grit ceramic belt cuts deep and roughly. Only improvement would be ahead of this if any of the damage due to the rough treatment by the belt goes a little bit past the visible grinding marks.

One Last Speculation

Without doing a whole bunch of research, I would speculate that many of the older irons that are really a treat are that not necessarily due to the complete lack of existence of any other alloying elements, but rather that the ore shown to provide good results was then used. And whether it was known or not, what differentiated one ore from the next was not just lack of undesirable elements, but also lack of traces of desirable elements.

I haven’t had a chance to look much more closely at this because one of my tricks in my small bag now is to wear away some steel by planing and see what shows. I have a lot of older double irons, and expect that in general, they were not high carbon and probably shied away from the 1% carbon level staying more like 0.9% or a little below to avoid the problem mentioned above with 1095 – too much carbon remaining in the lattice. The one thing that could disprove this or may, at least, would be finding familiar patterns of carbides in these older irons – something I’ve really only seen in one laminated stanley 2″ iron.

Another woodworker has mentioned the chance to XRF (nondestructive analysis) some older tools to see what is in them other than carbon. The test does not identify carbon, but does identify most other things we would consider interesting. It is the same test used by two different people (at least) to find out what’s in PM-V11 when LV rolled it out. I had nothing to do with that effort and at the time am not sure I cared that much about it other than minor annoyance of not knowing what’s in the steel. The prevailing notion on woodworking boards, that the steel was a developed proprietary alloy, didn’t make sense to a few people who knew that LV’s cost figure for selecting the steel wasn’t high enough to actually fully develop a new alloy.

But, that’s just another example of overconfidence of the majority aided by lack of exposure or real experience. Sometimes it’s fine to just say “I don’t really know”. Even as much as I’ve gotten my hands dirty, I’m looking for outcomes. As for why they are what they are in each case, “I really don’t know” quite often.