The Plane –
The plane for this project is going to be loosely based on a miter plane made by Robert Towell of London in the early to mid 19th century. The original plane belongs to my friend Joel Moskowitz of Gramercy Tools and Tools for Working Wood, and I had a chance to take some measurements and photography of the plane last fall at his shop. Joel, who is an avid historian of woodworking as well as a woodworker and tool collector, did a brief writeup on Towell and the plane on his blog; if you search his blog for ‘mitre’ (his preferred spelling – in the british tradition) you will also find quite a fair bit of other information regarding these planes, which are somewhat of a curiosity in that there isn’t a firm consensus as to what they were originally used for. I’m not going into that area here – at least not now – but Moskowitz has offered some ideas in his blog.
This is not a project I would recommend for a first attempt at making an infill plane – there are several features of its construction that are somewhat advanced, and add quite a bit to its difficulty. Firstly, the one-piece bent sidewalls are tricky to form and fit infills to; Secondly, this is a bevel-up plane design, which requires a two-piece sole construction — one of the more challenging aspects of planemaking to get right in my experience. Having said that, however, this plane does have a lot to interest people who are considering making a plane, as well as those who are perhaps already somewhat experienced planemakers. It should also provide a fairly comprehensive picture of the basic processes necessary for all dovetailed infill construction. For these reasons, I think it’s a solid candidate for a project writeup – and I’m hoping it will provide a fairly accurate idea of what infill making entails, and some of the choices and problems that can come up in planning and constructing a plane from scratch.
Design – sharpen those pencils and/or clean the lint from your mouse.
The first step in getting the design established is to make some dimensioned drawings, establishing at the very least the major angles and distances necessary to start laying out and cutting the sole and sidewall pieces. I have some limited experience with Sketchup, but I still generally prefer graph paper and pencil for most of my construction drawings. This is largely a matter of comfort and preference, though – you should use whatever format you’re most at home with. Here is the rough construction drawing set I am going to be working from:
The next step for me is making some test bends in my steel bending fixture, establishing the setup I’ll be using and the radius of the curvature of the sidewalls at the rear of the plane. This is necessary to ensure I have the proper layout of the sidewalls, which are all laid out, cut, and filed before the bending happens. While it’s a simple task to calculate the radius of the curve mathematically, there are two problems with using this as the only measurement – one is that the final curves will rarely match up to a perfect semi-circle, such that the final radius will almost always be somewhat shorter than the theoretical one; the second problem is that there is always some degree of stretching and compression of the steel when it’s bent to a curve. The bottom line is that you can get an approximate prediction of the length to allow for the radius, but in my experience it can be off by as much as 3/8″ from the actual results. A test bend using the same thickness of steel is the only reliable means of establishing the necessary length. I use inexpensive cold-rolled steel for these checks, which generally give results within 1/16″ of my final bend – quite close enough for my purposes.
I use a Grizzly compact bending tool, which I mentioned briefly in an earlier post. This isn’t really necessary, however – planemaker Bill Carter, who’s made some of the most beautiful mitres ever produced, uses simple bending forms in a machinist’s vise for all his work. The compact bending tool just adds a degree of simplicity and repeatability that I find helpful if you plan to make more than a couple of planes.
Laying out sidewall dovetails – pins first
Once I have worked out the bend setup, it’s time to generate a dovetail layout template. I do this on graph paper with simple black markings for each of the ‘pins’ on the sidewall. Notice that you only lay out half the plane, in this case from the front edge to the rear centerline – you’ll mark half of the sidewall, then flip the blank and mark the other side from the same template to ensure you have a nice, symmetrical dovetail arrangement. If you enlarge the next photo below, you can see the markings on the graph paper below the steel.
Now it’s time to start preparing the steel flat stock. I’m using 3/16″ O1 steel of 2″ width. Allowing 1/16″ of excess for peening dovetails to the sole, this should give me a final height of 1-15/16″ for the plane. I cover the ‘outside’ face and bottom edge of the steel with layout dye, then mark off the front edges (again, leaving at least 1/16″ excess for peening the front wall on later) and the centerline. Now I use the template to mark the dovetails off at the bottom of the face. You can see the template and the already-marked blank below:
With straight-sided planes, I have some degree of choice as to whether the sidewall dovetails are the ‘pins’ or ‘tails’ — if I use pins here, the sidewalls will drop in to the sole from above; if I use tails,
the sidewalls will slide in to the sole from the sides. For a one-piece sidewall like that in this plane miter, however, the sidewalls can only drop in from above, so the decision is effectively made for me – pins it is.
the sidewalls will slide in to the sole from the sides. For a one-piece sidewall like that in this plane miter, however, the sidewalls can only drop in from above, so the decision is effectively made for me – pins it is.
On the face of the stock, I first scribe a line to mark the top of the dovetails – in this case, I have allowed for a 5/16″ sole and 1/16″ of excess, so this line is scribed at 3/8″ from the bottom.
With the template markings transferred to the plane, I use a square to extend the marks up the face of the stock to the line scribed previously. On the bottom edge of the stock, I then mark off the ‘flare’ for the dovetails as they will be seen from the sole of the plane. My habit is to draw my templates to represent the outside of the sidewall, so each pin expands to the inner face at an angle of about 12-15 degrees, which I find to be a nice balance between aesthetics and ease of peining closed. The photo below shows the completed dovetail layout.
At this point, I cut and file the dovetails as laid out. There will be additional layout for the secondary dovetails later on, as well as the mortises for the bridge – but there is no reason to add to the complexity at this point.
Cut and file… repeat.
A few quick words about cutting and filing these dovetails… In many ways, dovetails in metal are more forgiving than in wood, because you can close up minor gaps during the peining process later on. Having said that, though, it’s very much in your best interest to be as precise as you can at this point – particularly in keeping the pins perpendicular to the sole, and in keeping the baseline intact. Every bit of precision at this point will pay off later on – and minor errors now have a way of magnifying into bigger problems later in the process.
I’ll try to add a post later about the types of files and hacksaw blades I like best for this sort of work, but when starting out you can make do quite well with a set of small Nicholson files if you grind safe and acute edges on a couple of them. I use 18-tpi bimetal hacksaw blades for cutting steel. One of the reasons I prefer O1 to mild steel is that it is much easier to remove waste from harder steels. In the photo below, I have started comb-cutting the waste for the dovetails:
In general, I try to leave somewhere in the range of 3/32 – 1/8″ ‘tines’ between each cut. This thickness of material is very easy to whack out with a cold chisel later on. I actually use an old woodworking chisel with a flat bevel ground on it for this – it’s just the most comfortable and controllable one I’ve found for it:
The simple idea is to tap each ‘tine’ first to one side, then to the other. Generally, two sets of direction changes is all that is required before the material snaps cleanly off. In softer material such as mild steel or bronze,this does not work anyhere near as easily or cleanly – and it’s often much easier to just saw the waste out with a jeweler’s saw if you are working in those metals. O1, however, is remarkably conducive to this.
Once the material has been broken off, it’s time for the file work.
Again – this can be very tedious work, but the more accurate you are at this stage, the more smoothly things will go later on. A couple of hours work later, here’s the sidewall stock with the first-stage pins all cut and filed:
The next step will be to lay out and file the secondary dovetails, and the mortises for the bridge. This entry has already gotten long enough, however, so I’ll come back to that process in the next entry in a day or two.
For those of you reading this, I would really appreciate feedback as to how helpful this is. Questions are very much welcome, as are any comments on what is or is not helpful for understanding and following along with the progress. My habit is to try to err on the side of too much information, rather than not enough – but if the entries are becoming too long and detailed please let me know.
Anonymous says
Amazing. Makes me want to try it. You put enough detail in to make think I could do it. But a couple years of hammering steel into armor let's me appreciate your skill.
Anonymous says
Amazing. Makes me want to try it. You put enough detail in to make think I could do it. But a couple years of hammering steel into armor let's me appreciate your skill.
thewoodshopbug says
How do you grind the safe edges onto your files?
I look forward to watching this project's progress.
Thanks,
Calvin
thewoodshopbug says
How do you grind the safe edges onto your files?
I look forward to watching this project's progress.
Thanks,
Calvin
raney says
Hi Calvin – At this point I have a fairly extensive collection of files, including a lot of pillar files which have safe edges on them. When I started making planes, though, I ground a few of my smaller mill files down – mostly 6 or 8" mill files. One or two of these will serve most of your needs.
I just use my bench grinder; the grey metal wheels that come on most of them are perfectly adequate for this sort of work. Set the tool rest to somewhere around 30-degrees so you can establish an acute angle for getting into the corner of dovetails. Then slowly grind away the long edge on each side of the file and you should end up with a nice large flat that ends in safe, sharp edge. For a lot of work you'll do, this file and a couple of narrow triangular files (such as those for saw sharpening) will serve just fine.
If you are looking to invest in a couple of files, I'd highly recommend a barette needle file for getting into corners, and a narrow swiss pattern pillar file or two – 00 or 0 cut. The one I use most is a grobet 6" long and 3/16" wide. Most jeweler's supply houses carry these files – they are somewhat expensive by mill file standards – probably around $15 each – but will increase the speed and ease of working on metal tremendously. I buy mine from an ebay store called 'tools for jewels', and I can recommend them without hesitation.
raney says
Hi Calvin – At this point I have a fairly extensive collection of files, including a lot of pillar files which have safe edges on them. When I started making planes, though, I ground a few of my smaller mill files down – mostly 6 or 8" mill files. One or two of these will serve most of your needs.
I just use my bench grinder; the grey metal wheels that come on most of them are perfectly adequate for this sort of work. Set the tool rest to somewhere around 30-degrees so you can establish an acute angle for getting into the corner of dovetails. Then slowly grind away the long edge on each side of the file and you should end up with a nice large flat that ends in safe, sharp edge. For a lot of work you'll do, this file and a couple of narrow triangular files (such as those for saw sharpening) will serve just fine.
If you are looking to invest in a couple of files, I'd highly recommend a barette needle file for getting into corners, and a narrow swiss pattern pillar file or two – 00 or 0 cut. The one I use most is a grobet 6" long and 3/16" wide. Most jeweler's supply houses carry these files – they are somewhat expensive by mill file standards – probably around $15 each – but will increase the speed and ease of working on metal tremendously. I buy mine from an ebay store called 'tools for jewels', and I can recommend them without hesitation.
Craig D says
Question about tails on the sides of a one piece side: Is there enough spring in the U shaped side to open it up and slip in the sole? I like tails on sides because I get warm fuzzies thinking that the sole is somehow more securely attached.
Craig D says
Question about tails on the sides of a one piece side: Is there enough spring in the U shaped side to open it up and slip in the sole? I like tails on sides because I get warm fuzzies thinking that the sole is somehow more securely attached.
raney says
Hi Craig,
I was actually thinking of you when you I wrote that part, as I believe I've seen you mention that way of working before. My main concern is that in order to make that work, there would have to be a good deal more 'play' in the fit of the dovetails, particularly at the rear. I am sure it's possible, though – it would probably just mean some larger gaps to pein shut.
To be honest I think that properly peined 'secondary' dovetails are orders of magnitude stronger than the forces trying to separate the sole. I think it was Peter McBride who has posted about trying a 0-taper dovetail (or box joint) peined shell,and he found it more than strong enough for plane work.
Most of the scant evidence I've got – both anecdotal from folks who've dismantled them, and a couple of catalog reprints – would actually seem to indicate that both Mathieson and Spiers used tails-on-sole for their bench planes as well as miters, at least much of the time. I'm not sure I see much advantage one way or the other in two-piece sidewall designs, but I've taken to using that method for basically all my planes, just because of the consistency of it.
That said, however, the most pervasive rule I've found about infills is: anything goes, and if it's possible, it was probably done.
Steve Kirincich says
I have always assumed that placing the tails on the sole made it so that the major hammering would be done to the sides and thus minimize distortion of the sole. Do you have any recommendations for accurately filing to the line along the straight lengths between the pins? Perhaps this will inspire me to use some pieces of 3/16 bronze from a large plate found on Ebay several years ago. Great website!
raney says
Hi Steve
I've had a similar thought, but in actuality since the peining takes place on the edge of the tails themselves (which are part of the sole plate) I tend to think that this would actually cause more stress to the sole than having tails on sides would. It certainly is a possibility, though.
To be honest, my suspicion is that the construction method is most likely just a vestige of miter construction, which was the field infill planes evolved out of. Since so many miters (almost) require tail-on-sole construction, I would imagine most early infill makers were simply accustomed to that practice. It's not hard to imagine that they simply kept the same practice in place when they started making infill smoothers and panel planes…
As for accurately filing the spans between pins, there are a couple of things that help somewhat. First and foremost is getting a good, clean scribe line that is very clear. I prefer to file up to the line, leaving it just intact, but all that is required is consistency. The other thing that I find helpful is having a dedicated pair of inserts for the vise. I have a pair of aluminum sub-jaws that I file in-place occasionally to ensure that they're at exactly the same level, flat across their width, and also perpendicular to their faces. My practice when filing to the baseline is to set the piece in the vise with the baseline about 1/16" above these sub-jaws, and file to within a couple thou of the line. For the longer stretches between dovetails at the mouth, or the curved portion of a miter, I will drawfile right to the line; for the smaller sections where that's not possible, I file right to the line, but for the last thou or two I generally use a fine cut file (I use a #2 cut pillar) rather than a coarse one.
The only other thing I'd suggest is to have a good raking light source. This makes a significant improvement in your ability to judge how far you have to go to hit your scribe line.
Have at that Bronze!