The cap iron video from Kato and Kawai was never intended to provide instructions or a basis for setting a cap iron. How do we know? K&K actually flatly said that. So if you’re telling everyone “it’s just science” that you should use a flat 80 degree bevel angle because you just watched the video and now you know enough to be a doctor of planing, hold your horses and get some actual experience.
So, this whole cap iron thing has made rounds in two different places lately. One of the things that annoys me is it’s brought back up and Nicholson described how you should set up the edge of a cap iron 200+ years ago. It should be rounded. I found out about Nicholson, I don’t know when – probably five years after writing the cap iron article. When I wrote the cap iron article, I was working wood entirely by hand. the only reason I’m doing less of it right now is because of the metalworking.
If you go back and read my article, there are two flaws in it. One is Ellis (the editor) left some tearout in the picture and I am horribly disagreeable – I didn’t like it. I also had become recently aware that being disagreeable and stubborn wasn’t infrequent like I thought maybe it was. Ellis thought the picture was interesting because it showed one of his first few attempts on quartered mahogany, and that removing a lot of tearout from something that couldn’t be planed before was valuable or just as valuable. I didn’t really agree that it was valuable enough compared to showing no tearout, but feeling like I had to get over being disagreeable, I let it go. I’ve heard about the tearout in that picture a lot – but usually from people who pride themselves on opinions without accomplishments.
The second thing I let go was the article at the end saying that it was based on information from the Kato and Kawai video. It wasn’t. I figured out how to use the cap iron on my own as a swan song before quitting woodworking. It worked so I didn’t quit. Bill Tindall sent me a message (email) when I started posting with elation about the world soon changing for hand tool woodworkers. One plane under your bench if all you do is smooth – to replace the myriad of suggestions. Bag the idea of buying infill if you’re an everyman. It’s a waste of money to buy lots of stuff but more importantly (money wasn’t a problem for me), it’s a big waste of time and it also has a little bit of the “i’m a fraud” feel to it if touting that you know 14 ways to slowly get rid of tearout when one works better than all of them 99% of the time. Bill’s message was that they he and Steve Elliot were digging up info from the K and the other K in Kato and Kawai. I think, Bill will remember every detail, maybe they were conversing with only one person. At any rate, there were a bunch of documents and the videos. He sent me a link to the videos before they were public because no permission had been gotten yet in terms of sharing them.
When I described what I did on the public forum, with elation, and as much “best thing ever” excitement as I could, I thought for sure three or four people would try it, see that the Jesus Christ of metal plane parts was right under our nose and we would be saved. Instead, most of the responses were “that doesn’t work, Chris Schwarz says so” or “of all of the people I’ve met or made planes for, only one person other than Warren ever said they use it”. I have no idea who that second person was, by the way.
I either didn’t remember or later forgot the article (on woodcentral) said that it was based on the video. This is an important piece of horseshit for two reasons. The video leads you down the wrong path when setting up a cap iron on a hand plane. You’ll watch it and assume you want to have a plane with an 80 degree flat wall facing the oncoming shaving and have it twice as far away as you would otherwise.
I tried it, it worked like shit compared to a rounded edge. Why I rounded the edges, I don’t recall exactly but it just seemed to make sense. Making sense isn’t worth much, though. It proved to be better in actual use with a shallower flat bevel being an option – pretty sure I wrote 50 degrees if you couldn’t be bothered with curvature. 50 degrees is “fine” like food with decent flavor and a bad texture is “oh well, it’s fine”. That kind of food is a hell of a lot better than an empty stomach, but it’s a missed opportunity. A 50 degree flat bevel will do most things, but it will struggle a little in finer shavings in the worst of woods.
I saw the video as maybe a way to hit people up with some moving pictures and then smack them upside the monitor with a bunch of info about what actually works well in hand tools. But I wanted to avoid the idea that you need to measure things and that you could just take something from the video and have everyone doing it. the “80 degree” setting was subpar and increased the chance that you’d either have the plane stop you in your tracks or leave a fuzzy surface. What’s the point of ramming the shaving so hard back into the wood that it’s compressed and the planed surface isn’t smooth. There isn’t really one. Quickly upon finding a tolerable resistance level planing, you’d then find the thing was too far away and not preventing tearout. The rounded profile suffers the ills of neither the shallow or steep single bevel.
I found it by experimentation, Nicholson documented it long before. It’s too stark to avoid noticing unless you’re only taking tiny thing shavings. If that’s all you’re doing, your advice is no good for anyone doing more. The same is true about general plane design, setup and use if you’re only smoothing and fitting joints. You will never know as much about planing as you would if you did five hundred board feet entirely by hand. Same with sawing and so on. You learn things whether it seems like you wouldn’t or not. What is is. What you imagine is common sense may not be what is.
So, the Video Does What?
The video was apparently a segment to show at a conference in the USA, and according to Bill, the professors stated that it’s not intended to be a guide for setting anything, and Bill mentioned further that the stuff shown in the video didn’t even make the final documentation. The documentation was involved in edge wear profiles and planing kilometers of wood with a planing machine trying out different steels. They found, drum roll, that highly alloyed steels lasted longer, but the wear profiles (going from memory here) also show belly and lack of clearance that the machine can power through that none of us would ever come close to doing. I think Bill was interested in finding a secret to longer edge life and the chipbreaker video was a side show that he provided because I brought it up on a forum. I don’t know if he would’ve worked hard to get permission to show it without the discussion, I think he would have but how it would’ve been presented, I don’t know.
I remember Bill saying there was a separate paper discussing setting up the cap iron with students, or I thought it was maybe students (one of “the K’s” students in Japan). Doing some digging this morning, it was students. The separate paper for hand plane setting is brief, doesn’t explain much and has some comments that are not relevant for western planes. It does not give parameters for set distance or angles, it just says you can’t set the chipbreaker by eye and you need to judge the shaving and perhaps even measure it.
Well, you actually can’t see the end of the iron in a good dai, and if you’re looking in from the bottom, you have a very poor view because you’re relying on seeing depth. So their advice makes sense in Japanese planes. On a western plane with screw together parts that go in, of course you can learn to set it by eye – you can ignore K&Ks suggestion that you can’t. That’s what I was suggesting early on – learn what looks right and then just set it. It took me a week with no instruction to get this. I was planing a fair amount, maybe several hours worth of just planing that week, but it was still just a week.
I think all you need to do is look to see if a shaving is worked if you’re not sure. if it’s starting to straighten out, there is influence and you won’t have a disaster. Noticed in that picture with curly hard maple, the heavy shaving is flattened in areas – there is a feel that planing has when the chipbreaker is engaged.
That’s all the K&K short hand tool article said, though, set it by watching the shaving and Bill was a little unhappy with it. I think like most folks would want, he wanted instructions that were clear? I could ask him -but I’ll go with the guess. I think he doesn’t care. He uses the chipbreaker now quite a bit on furniture work – especially cleaning up junctions and such – and sets it by eye.
The video gets passed around and this part is important – the very guys who put it together said it was not intended to be used as advice for setting or doing anything.
If you’re working by hand, believe what you get for outcomes and what comes from your own hands and nerves and discernment. It’s not that hard.
I didn’t think I found something new when figuring the chipbreaker out, I thought I found something old and at the time, I was definitely pissed that the Chris Schwarz’s of the world wanted to be a source of information but couldn’t figure it out. Using the chipbreaker should be taught at the same time someone is taught about basic sharpening and the adjustment features on a plane.
Oh, and why wasn’t the student paper that big of a deal in Japan? I don’t think anyone does much planing with planes in japan on things like day to day work on building beams and doesn’t set a chipbreaker. Why didn’t we notice it? well, if you don’t know what the chipbreaker is doing, it’s easy to not notice shaving shooting up out of a plane instead of over the iron and back. And a second reason for it is if you get on youtube or the internet and you want to see Japanese woodworkers planing actual work – and not someone teaching students or appealing to tourists, you need to find things that are listed in Japanese. Google’s functionality in youtube still does a terrible job of finding someone planing a beam in what looks like a legitimate workplace.
If you look at this video or this video , notice the shaving come straight out of the plane even though they’re not that thick. Why? I would guess because planing a beam like these without a chipbreaker is a very bad gamble.
This is actually what I thought I’d find out with the chipbreaker – a way to go completely to Japanese tools, especially on smoothing. It just didn’t work out that way.
How about this planing demo? Their chipbreaker is set up at 90 degrees and it’s flat.
LikeLike
Notice the surface behind the setup at 5:00 and then notice the surface left behind with the blunt faced chipbreaker. Watch closely how much stuff is left sticking up (as in the planed surface has more trash sticking up on the surface). it’s a subtle difference but the last setting would prevent tearout, and potentially in some cases better if the wood grain is literally the same angle as the iron’s bedding angle.
But the cost in a lot of cases in so much force from the chipbreaker’s blunt front will be a surface that’s a little rough and a lot more resistance when planing, and it won’t prevent tearout if the shaving thickness is reduced whereas the other setting closer to the chip probably will.
LikeLike
since a couple days im experimenting with the chip breaker, unfortunately your original article is no longer online.
That 90° edge is called “Brechkante” apparently serves the purpose of mimiking the effect of draging a piece of paper over a scissor blade where it curls up. In case of the plane its forcing the shaving to bend slightly backward (as the shaving bending forward infront of the plane is a pain in the butt)
At least thats what i read in prewar literature + limited experience and it even somewhat shows in the video sequences 5:00 where it goes slightly forward and 7:40 where it tries to torque back, even hugging the cap iron instead of detaching
LikeLike
I’ll track down the article. Peter, the administrator of the wood central site, has been busy trying to figure out the best way to get value out of the large historical database of conversations at wood central, and the articles and archives get bumped around to different places as he’s experimenting. It’s likely still there. If it keeps disappearing, I will eventually just post it here.
LikeLike
https://www.woodcentral.com/articles/articles_935.php
There is one issue with this article – somehow in the editing, it was attributed or suggested as a method to be determined from the video, but it is not. I cannot remember why I allowed that to be said after edits, but it isn’t the case. I figured out how to use the cap iron on my own, and also that results in the way the video suggests the chipbreaker or cap iron be oriented for a machine isn’t necessarily good for hand planes. A very small steeper front section – tiny – then rolling off with curvature is far better than the 80 degree setup for a machine as it’s shown. This is easy to figure out on your own as the plane functions better with that curvature and not much of the height meeting the chip being really steep. A machine that will be set for one depth of cut and then used repeatedly is better for this sort of 80 degree wall type setup, but we do not work that way as humans and the adjustment range when set like that is very poor when varying shaving thickness.
I find the search function on the site now to not be very intuitive as it searches nothing by default rather than everything, and one must choose categories they think an article will be in. Otherwise it gives a false result that many things aren’t on the site without notifying you that you’ve searched a database of zero articles from the start.
I believe that some of the reason the editor wanted to tie the video to the article and mistakenly perhaps thought that only watching the video led to figuring things out is people will watch the video and believe it. If I demonstrate how this works or describe it, a lot of people online who know me or know anyone will assume if they know someone, then the article probably isn’t as good as it would be if it was from an imagined unknown person who is an “expert” that they’ve never discussed anything with to disagree. It’s a stupid bias – just as dumb as the assumption that a youtube video maker is friendly and doing things for your benefit just because you don’t know yet that they’re primarily motivated by making money off of you.
LikeLike
Thanks! Wondered if I’d gone too steep at the front of the cap iron, and now I’m sure.
LikeLike
If the front of the chipbreaker is curved and right where it meets the iron it is steep, that’s fine as long as the height of the steep part is very short, like sheet of paper or two thickness short. But that area has to be very small and then get less steep quickly or you will be forced to set it far enough back that it won’t be in place to offer any tearout protection on thinner shavings. And then you’ll have just a short range of working shaving thickness where you’re not having problems. It’s really inconvenient in comparison to a curved profile.
LikeLike
I probably made the flat area too big, though I did blend it into the curve. It’s mostly that I get accordioned shavings when it’s close enough to mitigate tear out and is miserably hard to push. My best guesses were that my cap iron was too steep and that maybe I was taking the “as close as possible to the edge” thing a little overboard. I tried backing off but couldn’t get into a reasonable force range without getting maybe 1/32″ off the edge.
On a tangent, I tried using a Hirsch 0 sweep carving chisel as a regular chisel because I really liked the feel of it in my hand. Very easy to sharpen and dry feeling steel, but I couldn’t get the edge to stop fracturing until I honed a fairly steep edge on it (35-45°?). Buffing a lower angle just wouldn’t work. Any experience with steel like that or their carving tools?
LikeLike