Archive for February, 2010

Towell miter – sole assembly

OK – now the plane shell is starting to move along.  It’s time to prep the sole assembly and cut the tongue and groove for the mouth. This part of the plane construction is only applicable for bevel-up planes, and doesn’t generally apply for bevel-down planes — which are the vast majority of infill planes.

For a bevel-up plane, however, we’re looking for an extremely fine, tight mouth – and the only reasonable way to do this with ‘conventional’ tools is by splitting the sole in two, then rejoining it once the ramps have been fit. The first step, then, is to split the sole in two right where the mouth would be, and shape the ramp and rear of the mouth.

Filing the Ramp
There is really no secret to filing this ramp other than to be very very patient and painstaking. Once it’s  close to the final level, draw-filing gives very clear indication of where the high spots in the ramp are, and where it is out of flat. This process is not nearly as difficult as it might appear at first glance, but it definitely requires patience… this ramp is the most critical single point of support for the blade in the final plane, and should to approached accordingly. Did I mention patience?

I’m using a 22.5-degree bedding angle, with a sole thickness that is 5/16″ (.3125″) thick.  Some quick trigonometry calculation give me the following:

          TAN(22.5) = .3125 / x    [where 'x' is the run of the ramp]
          x =.3125″ / TAN(22.5)
          x = .754″

so I mark out the rear of the ramp at 3/4″ from the mouth. In order to make it simpler to cut and file at the angle I need, I position the sole piece laterally in my vise at exactly 22.5-degrees — this way if I saw perfectly horizontally, I know I’m hitting the proper ramp angle.  This makes it simpler to hacksaw the series of comb-cuts to remove the bulk of the material for the ramp.

Here are some closeups of the cuts from front to back – ideally they should get as close to the ‘lines’ as possible, but they mustn’t go over or I’ll be left with a nasty mark in my bed.  And nobody likes that.

At this point, I would now whack out the ‘tines’ and then file to the final ramp surface as precisely as possible.  I wanted to show the method here for the benefit of anyone who wants to try this for themselves — but as I am a weak man who is currently suffering a nasty bout of tendonitis, this is where the siren-song of the new mill became too much to resist. I milled out the bulk of the waste (though the surface still needed some final filing to get it just-so)

The final results are hopefully the same, whether using a mill or a pillar file – a nice, clean, and very flat bed. Here’s what I have after some very careful finish filing:

Tongue and Groove 
Next, I open the groove for the tongue and groove joint that will register the two halves of the sole together. First I mark out the groove – roughly 1/3 the thickness of the sole, and about 3/16″ deep.  I put the sole piece back in my vise, this time perfectly vertical, and with the depth of the groove right at the top of the jaws to give me a guide to saw to:

Then carefully, I saw for the front side of the groove:

Once I’ve gotten to depth, I flip the piece in the vise and saw for the other side, taking my final cuts across both sides of the groove.  Then I refine both sides with files.

Here’s the final groove:

Now it’s on to the front half of the sole. The front piece is filed similarly, but I use a 45-degree angle for the throat piece. Additionally, I leave 1/4″ of extra material for the tongues that will register the front and back portions of the sole together.  I’m cautious to do a lot of test-fitting to ensure the joint fits properly, and that there is as little ‘step’ in the sole from front to back as possible – I’ll have to lap out whatever differential there is at the end.

H
ere’s both halves of the sole, ready for dovetailing:

Dovetailing the sole 
Finally it’s time to cut the dovetails to join the sidewalls to the sole.  In the interest of brevity, I’m not going to detail all the cutting and filing operations here – it’s been well-covered in earlier entries.  One important item to keep in mind, though, is that when joined together, it’s important that the dovetail joints force the tongue-and-groove joint closed, rather than open – so when marking out, I want to ensure the joint is very tight.  I put the sole assembly in a parallel clamp to force it tight, and carefully mark out all the dovetails directly from the sidewalls:

From here, it’s back to – you guessed it: more hacksawing, and much more filing. Finally, though, after a lot of filing, checking fit, filing some more, and checking fit again…

I get to another  milestone… the sole and sidewalls are now fitted, and all of the metal joinery for the plane is now complete.

 The mouth is consistent across the plane, and just a sliver of light – this is good, as it means I’ll be able to lap it open to the size I want once the plane is finished.

Now it’s time for a break – I’ll put the front and rear profiles on the sole next, and then (at last) I get a chance to actually work with some wood. Definitely starting to look more and more plane-like now.

Some new additions to the works…

As one might gather from my blog, I am somewhat fond of good tools as well as quality woodworking.  It is worth noting that the two do not necessarily go hand in hand – one can certainly appreciate the work or the tools without much caring for the other – but a keen affinity for both is a common malady. There are two recent additions to the Daed toolworks shop that I felt were worth noting here. One is large, one small, but in the end, I think both constitute noteworthy contributions.

One
The first is a piece of machinery that I hinted briefly at in my last post – astute viewers may have guessed that the T-slot table I used was part of a larger assembly… in this case a Johansson mill:

My best guess is that the mill dates from sometime in the 1950s; for those who aren’t familiar with Johansson, they were the company who made the venerable 8520 and 8530 knee mills for Clausing. Clausing later bought the company outright, but this mill is essentially identical to the earlier production 8520 mills. What’s notable about these mills is that they’re the smallest ‘serious’ knee mills I am aware of — they’re the basis for most of the small 6×26 knee mills made in Asia for the past couple of decades.

With a footprint of just about 36″ square, including all adjustments and movements, these mills are very popular in space-challenged shops like mine, and generally command a healthy premium for that reason – so I was extremely pleased when this one popped up at a very reasonable asking price near me.  I’d given serious consideration to one of the better mill-drill setups, in particular the Industrial Hobbies version of the RF45 mill platform, but at heart I had always hoped to find an older knee mill I could adopt.  The waiting paid off.
Until I find room for a Bridgeport, I anticipate this will serve all of my needs just fine. While I’ve been quite satisfied with the results that can be had making infills by hand with hacksaw and files, the truth is that it gets very physically demanding.  The mill should help to ameliorate that somewhat, hopefully helping to keep the tendonitis that’s started plaguing my right wrist at bay…

Two
The second addition is a generous gift from my friend Jameel Abraham, one of the most talented artists and craftsmen I know.  Jameel had noticed the Shinwa bevel gauge I often use in some of my photos, and asked me how I liked it.  We talked for a while about the dismal performance of, well, basically all bevel gauges — they’re essentially disposable tools for the most part, and while some are a slight improvement over others, in truth they pretty much all just suck.  They don’t hold their settings worth a darn, no matter what the mechanism, and I’ve yet to use one that didn’t annoy me to some degree or another.

There is one reputed exception to the above lament – which both Jameel and I had seen briefly at the first WIA conference in Berea Kentucky:  the Vesper Tools sliding bevel design. These are serious pieces of craftsmanship, and my brief handling of one at Berea left me feeling pretty impressed. I’d had a purchase of one on my ‘one of these days’ list, but as you might guess Jameel beat me to it.  Within a week of our conversation, this showed up on my doorstep:

Chris Vesper’s new engraved 4″ sliding bevel.  This may seem like a relatively small addition to the toolworks, but it’s not – for a few reasons.  First, because I use a bevel a LOT in laying out planes; second because it’s a phenomenal piece of craftsmanship – something I very well appreciate; and third, because it is a constant reminder that there is a right and wrong way to make everything. This tool stands out because I believe it’s the only ‘rightly-made’ bevel I’ve ever used – every other one is simply wrong-headed in its design, and its execution, by comparison.

In short, Vesper has taken what I considered an entirely unsatisfactory category of tool, and made it a work of art both in terms of function and appearance. You can read some interesting information about the design he uses to make these, but let me just say that a simple twist of the precisely knurled locking knob at the base of the tool locks down the blade, and it stays.  I mean it stays where it’s set no matter what.  I believe I could probably stand on it and get some movement, but there is nothing I can imagine ever happening in my shop that would affect the setting I put in this bevel. There is a lot more that I could say about it – it’s balanced very well, extremely well thought-out, such that there is no way for any part of it to interfere with its efficiency at marking out work – but at its heart, its job is to hold a setting and make it easy to transfer from one place to another.  From that standpoint, and any other I can think of, it is as close to ‘perfect’ as I think one runs across in the tool world.

I am well aware that I’m talking about a ‘simple’ sliding bevel, but a tour-de-force is a tour de force, whether it’s a space shuttle or a shoelace – and this is a tour de force. It really is one of the most compelling examples of pure, functional craftsmanship I have run across. And that is something I think is worth mentioning.

Thanks Jameel.

Towell Miter – bending and peining the sidewall/bridge assembly

When last we left our humble plane-to-be, it was in heaving anticipation of finally being bent into shape, finalizing the shape of the sidewalls and bridge assembly that is so critical to the design. This is, quite frankly, one of the most nerve-wracking parts of planemaking… the bend needs to be carried off with a high degree of precision in order to end up with a proper fit to the bridge, and a square, properly formed shell.

Getting Bent 
Until recently, I did bending operations using a simple wooden or steel form and a vise, in the way Bill Carter demonstrates on his site ( if you’re not familiar with his work, you should take the time to GET familiar with it — he is as close to a ‘hero’ as I have).  This works quite well, with a couple of limitations;  first of all, you are basically limited in the length of the plane, which generally needs to be less than the capacity of the vise you’re using. My modern taiwanese vise has an opening capacity of just under 5″, and my old Columbian 504 opens to a bit over 6″ — neither is nearly enough for this plane.  Also, when you get to stock over 1/8″ thick, the vise method becomes much more physically difficult, requiring a fairly sizable lever to get anywhere.  The biggest problem with a plane the size of the Towell, though, is in tweaking to get the final degree of precision.


So earlier in the fall, I made a modest investment in a ‘compact bender’, which avoids some of those limitations. To be honest, I wouldn’t really recommend it to most people – it’s far from ‘necessary’ – but if you intend to do a lot of this sort of work, it’s worth the $80 or so I plopped down for it.

 As a general ‘rule’, I find that it’s relatively straightforward to get within 1/16 of an inch of the desired symmetry. Unfortunately, however, 1/16″ is nowhere near close enough for proper fit – so I need to be able to correct the bend somewhat to get it ‘just so’.  With smaller stock of 1/8″, this is fairly easy to do gently hammering in a vise as Carter shows at the link above.  With a 3/16″ sidewall of tool steel, however… not so much.  In order to tweak the bend in the location I needed to, a little creative fixturing was in order, as shown below:

It was very fortunate for me that I had come into the t-slotted table pictured here just prior to this point.  It came attached to a number of other bits of metal, but for the purposes of this operation, the table and the clamping setup shown were all I needed.  More on the ‘rest’ of the table a bit later…

If you look closely, though, you’ll see how this works, letting me very precisely adjust the bend at the lower part of the rear curve without risk of losing the straightness of the plane sides — in this case, pushing the ‘top’ side of the plane a smidgen further forward to bring it into exact symmetry with the other side.  The Bessey clamp worked just fine, though it was probably operating at the limits of torque it’s capable of.  At any rate, this got the plane body to just the shape I needed, ready for the bridge and front plate to be inserted, and the whole assembly peined.  First, though, the bridge needs to be shaped and finished.

Shaping the Cupid’s Bow bridge.
Shaping the bridge for a wedged plane is one of my favorite tasks; in general, the ‘sculptural’ elements of the plane are the most fulfilling to me, and I look forward to them. Forming a Cupid’s bow is not difficult, but it helps significantly if you ‘slave’ its dimensions to fit the tools you have — specifically, I try to make sure the concave curvature matches one of my round files, preferably a size I have both roughing and finishing cuts of file in.
I’m not covering the procedure in detail, but the setup I use to make these should give anyone interested all the information they need to make a very solid attempt at one of these.

Notice the angle of the bridge in the vise, which means basically all file work is done with the file perfectly horizontal. Once I’ve roughed out the entire shape, I refine it with contoured sanding blocks (available from any number of sources).

I generally like to finish the cupid’s bow itself to a pretty high polish – leaving the rest of the bridge with a coarser finish of about 320 grit or so.  I like the counterplay between the textures, and find it really draws the eye into the decorative feature.

A few minutes with abrasives on granite surface plate to finish the flat surfaces, and the bridge is done.  I finish the underside as well – if for no other reason than to be SURE I have removed all the burrs from the underside of the bow,  which can wreak havoc on the finish of the wedge otherwise… or so I’ve been told.

Assembly and peining:

In order to slip the bridge in place, I need to open the body slightly.  I use the jaws on my machinist’s vise as a spreader to do this.

Then, using the vise to close the plane up, I can insert the front plate into its dovetailed ‘home’ and the sidewall assembly is ready to be peined together.
 The first section I pein is the bridge tenons – this brings the back half of the plane into alignment, and closes up the sides tight to the bridge. You can see the ‘gap’ at the lower side of the bri
dge here; if I’ve tapered the mortises properly, the peining process will automatically close this up and leave me with a very tight fit.

Once the bridge is peined, I can turn my attention to the front plate.  The front side of these dovetails is a bit of a pain to pein closed, as there is no way to ‘back up’ the hammer blows on an anvil.  The best I’ve managed to come up with is to pein it with the sidewalls pinched in my bench vise, readjusting the plane every minute or so as it slides down the jaws a bit. This is the reason I spend so much time making sure this particular fit is extremely tight when cutting the front plate joints — if I have to close up even a small gap, it can take literally hours to get enough movement to do so.  The goal is to have a joint that is tight enough already that I only need a few minutes to close up the small seams from the front:

Once the front of the plate is piened, it is extremely secure in its ‘seat’, and I can safely pein the sides of the dovetails without concern for pushing the plate outward. 

At this point everything looks good, so I grind off the rest of the excess material. This is one of the nicer points in making a plane like this, giving me my first glimpse of the overall ‘form’ of the final plane.
 
 

So far, so good – now it’s time to start work on the sole.

Towell Miter – front plate and bridge

This entry will cover the remaining steps that need to be accomplished prior to bending the sidewalls.  The front plate needs to be fabricated, and the dovetails cut for joining it to the sidewalls, and the bridge needs to be completed to the point of fitting the tenons to the sidewalls.  After that, its simply a matter of cleaning the interior surfaces, and then I’ll make the bend.

Front plate

This is a pretty self-explanatory process – first I need to cut the plate to size, then cut the dovetail joints to fit it to the sidewalls. The sizing of the plate is fairly straightforward – it’s the height of the sidewall pieces above the dovetails – in this case, the sidewall stock is 2″ wide, minus 5/16″ for the sole, and another 1/16″ for peining material.  So the remainder is 1-5/8″ tall (I measure the piece itself as a check).  In width, the plate needs 2-1/8″ for the interior of the plane, with 2 x 3/16″ for each sidewall, and an additional 1/8″ for peining material on either side – this gives me a width of 2-5/8″.

 
 Once the plate is cut, I need to clean up the cuts from the hacksaw.  I used to do this work with files, but I find I get results that are as good (and much quicker) with a good 12″ disc sander:
 
With the plate square, it’s a simple matter to mark of the ‘tails’ that will join the front plate to the sidewalls.  This joint is responsible for holding the bend in place, and experiences some fairly good lateral stress, so I feel that tails on the front piece are really the only sensible choice here.  At any rate, I’m going to include less and less ‘cut’ and ‘file’ pictures, but I’ll still include them occasionally.  Here are the dovetails being cut:

and then they’re filed.  Flip the plate and repeat.  Next, mark the ‘pins’ from this piece at either end of the sidewall stock… pay close attention to the orientation that will result after the bend – there is nothing more frustrating than cutting a perfect dovetail that is flared in the wrong direction, scrapping a sidewall piece you’ve already got a half-dozen or so hours into… or so I’ve been led to believe.

Here’s the fit of one side.  Ideally, you want this fit tight enough that it requires some light hammer taps to seat or remove the piece – that makes for very easy peining later on.  In reality, you can get away with a good bit more slop than that, but it’s a lot more work later…

 
 The only thing remaining is filing secondary dovetails.  These dovetails are done in 3/16″ stock – not nearly as thick as the sole – so I’ll use a ‘conventional’ full-depth taper for them.  Here they are marked out:
 
and filed:
 
At this point, all of the operations on the sidewall pieces are complete, except cleaning up the interior surface.  It is all but impossible to adequately clean up the inside of an already-assembled plane, so I want to put a final finish on the inside now.  I use PSA sandpaper on a large granite block for this – and for the interior I’ll sand to 220 grit. You can go higher than this if you prefer bright and shiny, but I  I like a little texture to steel. The finer you abrade the steel, the more fingerprints and dirt will stand out and interfere with the surface aesthetics, so bear that in mind if you wish to use a higher grit for your surfaces… I will say that almost without exception, I have tended to back off on my surface treatments over time – I’ve rarely, if ever, felt I needed to shift my regimen to a higher degree of polish.  This is largely a matter of taste, though, so use the finish that suits your sensibilities.
Here is the sidewall, now with all the cutting and filing completed, and ready for the bending operation:
 
 But before I actually do the bend, there’s just one more task I want to finish.

Bridge tenons
It is infinitely easier to fit the bridge tenons before bending the sidewalls, so I cut the bridge piece from 5/16″ stock, and cut/file the tenons in place.  I will finish the actual shaping of the bridge a bit later, but for now I want to get the fit for the bridge correct before I bend the sidewalls up. The tenons themselves are really straightforward – cut, remove some metal, file a bit and – voilá!  The only tricky part of this exercise is filing the shoulder in place on the tenons.  Once I’ve got the tenons themselves cut I have to remove about 1/16″ of material from the top face of each .  This gives a nice shoulder, and a substantially nicer aesthetic fit to the bridge when the plane is completed. Here’s one side of the bridge, ready for the shoulder operation:

What I’ll do is use the vise jaws as a guide to cut to depth with my hacksaw – this will make it much easier to get the filing depth correct in a minute. I’m careful not to cut deeper than my 1/16″ depth, as cuttin
g beyond that point will substantially reduce the strength of the tenons:

Once the cut is made, I use the jaws again as a guide, this time for a pillar file used horizontally to remove the face material from the tenons ( you can’t actually see the tenons behind the file, but trust me – they’re there.)

Once you’ve gotten to depth, flip the bridge and repeat on the other side.  A few test fits and refinements, and you’re left with the bridge piece looking something like this:

Be certain you have a perfect test fit on either side of the sidewall piece (and pay very close attention to orientation).  You want a fit that is easily pressed in place without much resistance, but not too sloppy – you must be able to get the bridge in place after the bending operation without too much trouble.  Here is one side’s test fit:

 
And now I’m finally ready for the nervewracking step of bending the sidewalls.  I find everything prior to the sidewalls is accompanied by a silent prayer that the bend goes off well – it’s not unheard of for a bad bend to mean all of the work to this point is basically scrap metal, so there is a bit of apprehension in building up to it.  
I’ll cover the actual bend in the next installment, as well as inserting the bridge and front plate.
Return top