A planing question.

[D_W]

You ought to set up a camera next time, Charlie. I think people would enjoy seeing you do that entirely by hand in 2 1/2 hours. No need for production video, just off to the side camera catching the process. Would make for an interesting time lapse, too, if it was done all in a row.

 

[D_W]

By the way, a guy who can make something in 2 1/2 hours entirely by hand ought to be able to avoid planing the ends off of a board unintentionally.

 

[Trafalgar]

To be frank, a guy that can make that top in 2 1/2 hours doesn't need lessons from a guy who couldn't. If you can, then do it. I'll be the first one to congratulate you, as I've always done with your plane builds.

I don't plane the ends off unintentionally since I don't run the edge all the way through until I've put a little hollow in. I intentionally use the stop shaving method in order to get the work done, delivered, and billed.

I'm not interested in seeing myself on camera. I'm interested in seeing myself at a teller's window making a deposit.

 

[D_W]

duplicate post.

 

[D_W]

I'm not interested in seeing myself on camera. I'm interested in seeing myself at a teller's window making a deposit.

Oh Jeez, here we go, the boasting starts. There are smarter ways to achieve that (making deposits - I thought you mentioned once a while ago that you were carrying a CPA).

(I misread earlier while watching the kids and thought you said you made the entire thing in 2 1/2 hours and were just giving everyone a yank. )

 

[D_W]

I'm curious as to what steps the "stop shavers", the ones who physically stop the plane short of the edge of the board , what those folks do for squareness.

It'll vary according to board thickness. But for the 1/2" to 1 1/2" thick boards that are the lion's share of cabinet work I'll take a pass or two on the machine planer then clean off the scalloping with one or two through strokes of the bench plane. Then test for smooth, straight and square.

About half the time, maybe a bit more, that's it.

Sometimes though I'll mess up and the plane will skip or skew, generally at the start of the cut. Or I won't find the dead centre of the camber, or very occasionally I'll wander from side to side during the stroke. In any of these cases the board will fail one or more of the tests. I might go back to the machine planer, although in a shared workshop you've often then missed your turn. Alternatively you might have painstakingly matched the paired boards for a certain grain arrangement, so taking off a mill or more of wood for a second time with a machine planer might risk that. Therefore most times I'll go to stopped shavings using the bench plane. And it's during that process that I bring the board back to square.

I'll check every 4 to 6" and pencil in marks that tell me where I have to drift the camber to correct any problems. I'll keep checking, but by the time the stopped shaving sequence is done (a minute or two at most), I've generally got my eye in with the plane settings and the edge is square, but not yet smooth or straight. Then it's one or two through passes with the bench plane and I'll run a finger slowly along the edge feeling for smooth, a final check with a combi square for square/wind, and a long wooden straight edge or the blade of a 600mm combi square for straight or a minute hollow.

Next the two paired boards get tested against each other. One goes in the vice glue edge up, and the other gets placed on top of it glue edge down. I'll pivot the top board, listening for a slight scraping sound, and feeling for some friction out at the ends. If it pivots freely that's a clear fail. And I'll closely examine by eye for any gaps at the joint. If there's a sliver of daylight in the centre that's okay, but I want to know I can block it out just with hand pressure. I place the blade of a 300mm combi square vertically on the two boards at a few different positions and check that they're in line. Any concerns and I'll investigate until I've identified the specific problem and fixed it.

Next the boards get placed on waxed bearers for a dry glue up in the cramps. If the grain pattern across the boards is now out of whack this is the last chance to shuffle the overlong boards back and forth to make amends. If all's well then I generally glue up there and then. I don't like leaving edges for more than 24 hours to gather dust, oxidise, or get dinged.

One last point. I'm fussier with tight grained pale timbers than with open grained dark timbers. Just because glue lines are more obvious against a pale, smooth background. PVA gives the best glue lines, but if the job demands hide, UF, or epoxy glue then I crank up the fussiness a notch to get acceptable results. I'm looking for tight, strong joints, with invisible glue lines and I'll do the least amount of work to achieve that.

Thanks for the summary, Custard. A friend of mine who is hyper about machine accuracy will always glue straight off of his jointer (machine jointer), but he is constantly fiddling with it to try to make it like a machinist's tool (in terms of accuracy). I do find a lunchbox planer very useful (and compact, and inexpensive so abusing it is fair play), but was glad to ditch a lot of the rest of the stuff. I'd still like to have a bandsaw, but not that badly. I literally have never wasted a stick of anything working by hand.

 

[Paddy Roxburgh]

When I read that Charles made that table top in 2 1/2 hours and that DW has never wasted a stick when working by hand I really need to question my woodworking abilities. I'd be happy to have made that top in a day, and I can practically heat a boat with all my mistakes.
As to the discussion about planning an edge, I did a little experiment with a scrap board and found that with very careful hand pressure I could take a board edge from slightly convex to slightly concave with through shavings. This is not my usual technique, I do stopped shavings and then a couple of passes of through shavings. I repeated this a couple of times (anything to avoid my actual work) and can confirm it is possible, however my edge ended up out of square, undoubtedly as I was concentrating on so hard on removing the hump. I doubt this is a consequence of the technique, just the fact I am not used to it.

 

[D_W]

Paddy, the comment about not wasting a stick by hand is true, but not close to true if I use mostly power tools and try to work quickly. I have repaired things while working (dovetail gaps, etc), but never made the kind of measurement errors of haste, like cutting a batch of sticking a quarter narrow or something. Wasting good stock is more annoying than taking twice as long to build something by hand.

 

[essexalan]

Planed up some 12" x 2" cherry this morn ready for an edge grain cutting board. I always use the same part of the bench because it is flat, how do I know it's flat? Because a board planed flat sits on it nicely without rocking with no discernible gap underneath it. How do I know the board is flat? I use a straight edge. How do I know the straight edge is straight? Two boards planed flat and checked with that straight edge fit face to face with no clearance or visible light showing through the join. Why do I use a straight edge? Because I am very short sighted and am unable to tell whether a board is flat or not by looking at it and glasses do not help. Anyway Wearing tells you how to make your own straight edge so they must be OK.
Flattened the concave side first and squared an edge using an engineers square. Convex or the side with a hump I took the hump out using a # 5 1/2 but a # 4 1/2 will do the job. I suppose you would call them incremental stop shavings as in take longer cuts as you flatten the hump out until you get a full through shaving, check with the straight edge and square. Finished it up with a # 5 1/2 set for a fine cut. This time I did continue taking a number of shavings but was unable to recreate a hump unless I kept the pressure on the front of the toe throughout the cut when I rounded one end, a couple of through shavings sorted that. Tried the same procedure with three different planes and the same result. Can't see how a plane with a flat sole taking through shavings of consistent thickness can create a hump unless you make it. Just a raw amateur and this was worked out by trial and error and all made easier using the cap iron sort of correctly.

 

[D_W]

Thanks for reporting back with your results.

 

[Cheshirechappie]

I dare say I'll be sneered at for this, but what the heck - here are an engineer's thoughts on what might be happening.

A few people have been close to mentioning the flexibility of planes and workpieces, but haven't actually done so. I think it may be a factor.

Considering the workpiece first, a piece 4 feet long and an inch thick being edge jointed is usually held in the face vice at about it's mid-length point. Thus, there can be quite a bit hanging free fore and aft of the vice. If the piece is nine inches wide, then applying the force of a plane to the edge won't deflect it much, but if it's only an inch-and-a-half wide, it could well do. Thus, deflection of parts of the workpiece might sometimes be a problem - less likely with a board being face planed flat on the bench.

Then there's the flexibility of planes themselves. An infill plane will be stiffer than a Bailey plane of similar length, because infills are more like a piece of steel channel having nice, wide, upstanding webs pretty much all the way along, whilst Bailey planes have longish parts at the toe and heel with much less stiffening web, so they'll flex more. The amount of flex will not be much in woodworking terms (you won't detect it by squinting down the plane sole), but it'll be there, and depending on how much pressure the craftsman applies and where he applies it, it could affect the plane's sole flatness enough to reflect in the work, especially if multiple shavings are taken at a fine depth of cut setting. Longer planes will tend to have more lengthwise flex than shorter ones, too.

That might explain disparities in results between Bailey planes and infills doing the same thing.

One thing to note is that the deflections in question are probably tiny - perhaps fractions of a thousandth of an inch over the length of a plane sole - but cumulatively over several shavings they could add up to different results from different planes. I don't know if anybody has ever tried to measure plane deflections under working loads, but given that almost everything else in the woodworking tool arsenal has been scrutinised and analysed - almost to death in some cases! - it wouldn't surprise me at all if someone has at some time or other!

The pragmatic craftsman at the bench won't bother with any of the above, of course. He'll do what all his forebears have done - work out what technique works with HIS tools and equipment, and stick with it!

 

[custard]

One thing to note is that the deflections in question are probably tiny - perhaps fractions of a thousandth of an inch over the length of a plane sole - but cumulatively over several shavings they could add up to different results from different planes. I don't know if anybody has ever tried to measure plane deflections under working loads

It's fairly common on many furniture making courses and apprenticeships to rest the toe of a number 5, 6 or 7 Bailey on a thin bit of wood with the heel on the bench, shim the gap at a fixed point, then apply some moderate downwards hand pressure to the mid point and re-shim it. I can't remember the exact results, but it was quite a bit more than fractions of a thou. I've also heard it said that beech bodied jacks and jointers were superior by virtue of their greater stiffness, but I've never seen that measured so I've no idea if it's hard fact or nostalgic propaganda!

 

[D_W]

I dare say I'll be sneered at for this, but what the heck - here are an engineer's thoughts on what might be happening.

A few people have been close to mentioning the flexibility of planes and workpieces, but haven't actually done so. I think it may be a factor.

Considering the workpiece first, a piece 4 feet long and an inch thick being edge jointed is usually held in the face vice at about it's mid-length point. Thus, there can be quite a bit hanging free fore and aft of the vice. If the piece is nine inches wide, then applying the force of a plane to the edge won't deflect it much, but if it's only an inch-and-a-half wide, it could well do. Thus, deflection of parts of the workpiece might sometimes be a problem - less likely with a board being face planed flat on the bench.

Then there's the flexibility of planes themselves. An infill plane will be stiffer than a Bailey plane of similar length, because infills are more like a piece of steel channel having nice, wide, upstanding webs pretty much all the way along, whilst Bailey planes have longish parts at the toe and heel with much less stiffening web, so they'll flex more. The amount of flex will not be much in woodworking terms (you won't detect it by squinting down the plane sole), but it'll be there, and depending on how much pressure the craftsman applies and where he applies it, it could affect the plane's sole flatness enough to reflect in the work, especially if multiple shavings are taken at a fine depth of cut setting. Longer planes will tend to have more lengthwise flex than shorter ones, too.

That might explain disparities in results between Bailey planes and infills doing the same thing.

One thing to note is that the deflections in question are probably tiny - perhaps fractions of a thousandth of an inch over the length of a plane sole - but cumulatively over several shavings they could add up to different results from different planes. I don't know if anybody has ever tried to measure plane deflections under working loads, but given that almost everything else in the woodworking tool arsenal has been scrutinised and analysed - almost to death in some cases! - it wouldn't surprise me at all if someone has at some time or other!

The pragmatic craftsman at the bench won't bother with any of the above, of course. He'll do what all his forebears have done - work out what technique works with HIS tools and equipment, and stick with it!

Bailey plane vs. infill won't make any difference. Wood deflection shouldn't make a difference unless you put a small stick in the vise (and that kind of work is done on the top of the bench). Iron thickness won't make any difference. The only thing that will make a difference is if a user is unfortunate enough to have a plane with a toe and heel lower than the mouth. Other than that, the difference here is user to user.

I don't have a 5 1/2 sized bailey plane, except two with rank set irons in them, and chose only to not use a bailey style plane because I figured if I used a jointer, instead of talking about infill plane vs. bailey, it would just be said that the longer plane was the reason for the flatness. The rank set planes I mentioned were never flattened - they were rough work planes at one point (one mine, and one belonged to a friend's father), so I have no clue if they're flat enough for this, anyway.

I do the same thing with bailey, infill or wooden planes. The only thing that matters (aside from no convex sole) is avoiding tearout (which is not hard) and avoiding the tendency to keep pressure on the front of the plane as it's passing off of the board.

I've never noticed any noticeable deflection on bailey pattern planes. I know you could put one in a vise and flex it if you bent hard enough, and measure that, but when you're planing, if you leaned that hard on one, it would be like planing with the brakes on. I had two LN jointers at one point - a 7 and an 8. The 7 was almost perfectly flat. The 8 was hollow by about the smallest feeler in my set (1 1/2 thousandths). If you were planing a board 42 inches long (remembering the board where I noticed this, part of some casework that I was building), or 42 plus a couple (whatever was to be trimmed off), you couldn't reasonably let the plane run over a jointed edge that was flat without first clipping the ends off. You could probably lean on the plane and get it to cut, but I didn't try it because you wouldn't do it in work. I sold the plane as soon as I noticed that because at that time, if you adulterated a LN plane, it dropped in value significantly. I think people are over that now, and I could easily have fixed it in 20 minutes.

There's no universal rule about what's more rigid in terms of infills vs. other planes, as quite a few of the infills have thin steel or gunmetal sides and a mild steel sole. They could flex at the cheeks. Many of the older ones don't have a universally tight fit of the wooden parts, either. I'm amazed that the old beech planes don't deflect more, but I've never noticed that they do. If they do, it's something you can easily overcome as a user. I can work faster on edge jointing work with them (there is less effort wasted in friction), but not finer. There is also much less tendency to fiddle with adjustment on them and try to be finer than you need to be. If I am matching a panel with one set to cut aggressively, I take a through cut (perhaps two thirds of a hundredth in shaving thickness), match the panels and clip the ends off with a smoother as needed. It's faster. Lateral adjustment on a jointed edge with one at a shaving thickness that thick is super fast. The only real detriment is if wood is really temperamental with tearout, and the fact that if you do a couple of hours of edge jointing (which implies from rough, otherwise you won't use one long enough to make a difference) with a wooden plane, it will have a noticeable area of wear in the center, and if that matters to you, then you will have to flatten the sole again.

 

[Cheshirechappie]

One thing to note is that the deflections in question are probably tiny - perhaps fractions of a thousandth of an inch over the length of a plane sole - but cumulatively over several shavings they could add up to different results from different planes. I don't know if anybody has ever tried to measure plane deflections under working loads

It's fairly common on many furniture making courses and apprenticeships to rest the toe of a number 5, 6 or 7 Bailey on a thin bit of wood with the heel on the bench, shim the gap at a fixed point, then apply some moderate downwards hand pressure to the mid point and re-shim it. I can't remember the exact results, but it was quite a bit more than fractions of a thou. I've also heard it said that beech bodied jacks and jointers were superior by virtue of their greater stiffness, but I've never seen that measured so I've no idea if it's hard fact or nostalgic propaganda!

I've just done the same experiment using a 1980s Record 5 1/2 shimmed up at the toe with a conveniently handy piece of brass sheet. Inserting feeler gauges under the plane sole just behind the mouth, I managed to get 25 thou feelers under, with a sliding feel. Applying a load to the top of the plane (fingertips of one hand, and not a lot of force - a sort of normal planing downforce) I managed 23 thou feelers to give the same feel at the same point along the plane's sole. The 'test bed' is the flat bed of a metalworking lathe, and can be trusted to be flat and rigid enough not to have affected the results.

Must admit, I was quite surprised at how much the plane deflected. I knew it would, because everything deflects under load, but 2 thou deflection for a very moderate downforce was more than I'd expected.

Edit to add - Just tried the same experiment with a 17" wooden jack plane (2 3/8" cutter, 3 1/8" wide body 3" deep, beech, no maker's mark). The result was that there was a deflection under load (same sort of 'working' load as before), but much less than with the Record; perhaps 1/2 thou or less. I could detect difference sliding the feelers along the sole, but it wasn't sufficient to warrant a different pack of feeler shims.

I don't own any infills, so can't try with one of those.

 

[ED65]

For anyone who doesn't own feeler gauges and can't do the test themselves Paul Sellers demonstrated that a plane can deflect in the centre in one of his videos on YouTube a few years ago. I don't recall the context but I'm sure if you hunt around you can find it without too much bother.

I've just done the same experiment using a 1980s Record 5 1/2 shimmed up at the toe with a conveniently handy piece of brass sheet. Inserting feeler gauges under the plane sole just behind the mouth, I managed to get 25 thou feelers under, with a sliding feel. Applying a load to the top of the plane (fingertips of one hand, and not a lot of force - a sort of normal planing downforce) I managed 23 thou feelers to give the same feel at the same point along the plane's sole.

Not to try to out-think an engineer on this but that's not normal planing downforce. The magnitude yes, but where the load is exerted must be important surely [rhetorical] and it's completely different. Hands fore and aft versus pressing in the centre changes things completely, but furthermore there's the surface against which a plane is pressed in normal use.

To my mind in normal edge planing where widish boards are held in a vice, boards less wide possibly resting directly on the bench, if deflection were an issue you'd have to be exerting sufficient pressure to deflect the board, or in the latter case the entire bench. I don't care how many Weetabix you've had that morning that's not happening :mrgreen:

So I expect that in the normal run of things plane deflection simply can't be a significant factor, in fact it may not be a factor at all as your further test strongly suggests.

Because it's just as easy to dub the ends, and just as badly, with a wooden jack or fore plane as with their metal counterparts.

 

[Corneel]

When I wrote about the method of "trying to dig a hole" that was ment as a learning exercise. I am not all the time trying.to dig holes in my timber. Stuff now happens almost automatically with a plane long enough. When I joint an edge I first work hard to square the edge which needs quite some back and forth for me with a trysquare and planning left or right on the stock. When I finally reach square front to back that usually means (9 out of 10 times) I reached straight and flat too.

The stop shavings technique is good to know, sometimes the edge doesn't want to reach straight that easilly so it is good to have another trick.

And really, I don't regard myself as a top qualified planing professional. Amateur for ever.

 

[D_W]

Thanks for that accounting, Kees. I'd also like to be an amateur forever. Getting good at things is fun. Doing them 200 times once you're good at them in a prescribed set of steps - not so much. I can do that at work.

 

[Cheshirechappie]

The experiment could be done t'other way about, as it were. Place a shim under the plane sole somewhere near the mouth, and see how much shim you can insert at toe and heel. Then grasp handle and knob, apply normal working downforce, and ask your glamorous assistant to check toe and heel with shims again. You will have spotted that by doing things this way round, you need at least three hands, which is what makes the original method outlined by Custard somewhat easier - only need two hands at most. The end conclusion will be the same - Bailey planes do flex a bit longways.

The effect in 'working mode' (hands on handle and knob) will be to cause the toe end and heel end to deflect downwards a bit, with the wood holding up the middle. Thus, the plane will act like one with a slightly concave sole. The degree of 'induced concavity' will obviously depend on the amount of downforce applied at the handle and knob, and won't be great using normal amounts of planing force. However, given enough shavings, it will result in a slightly crowned board or edge, as has been noted in Andy Kev's original post, and in David C's posts. Clearly there will some more complexity than that, given that planing downforce shifts as the plane moves from start of cut to end run-off, but with about even force down on heel and toe at mid stroke, the effect noted will happen.

The stop-shaving technique (or 'digging a hole in the middle') will correct for the plane's deflection by leaving up-standing wood at the ends, forcing the plane's toe and heel upwards and correcting for the sole distortion from downforce.

Thus, technique overcomes the natural imperfection inherent in the design of the tool.

 

[D_W]

Working mode off of the end of the board shouldn't have any measurable downforce on the front knob of a plane once it's headed off the end of a board. That's the entire point. It's not imperfection in the plane that causes the problem, it's imperfection in the user. The plane will do whatever you manipulate it to do.

I recall that David is very fond of a "lockdown" position when planing. One gets away from this when they do the kind of work Charlie showed (making a table top in 2 1/2 hours, which is about how long I'd expect it to take me to make something like he showed- not sure if our steps would be the same, but certainly different order than you'd have making that top with power tools). I may not make the first one as nicely as Charlie did, but I'd expect to by the second or third. Anyway, in the "lockdown" position that David C likes, you are not putting yourself in a situation where you can easily adjust pressure, and thus you'll plane the ends off. It's not hard to learn to plane well enough that you can literally let go of the front of the plane - as long as it remains engaged in the cut (something aided by the cap iron keeping the shaving continuous).

Subtle things, but things you can learn to stop doing if you really want to. If Charlie has made stuff like his table top by hand on a regular basis, then he's one of the few on here who has really done a lot of that.

Quite a lot of this is like golfing forums where guys talk about this or that club causing them to hit a draw or a fade when they've got a swing path that hits a draw or a fade regardless of the club. It's not the plane.

 

[Cheshirechappie]

I think the type of plane (and how the different types flex longways) could be a factor. It could explain why David C finds he needs the stop-shaving technique to achieve with a Bailey 5 1/2 what an infill of similar length (but greater stiffness along it's length) achieves without.

Please do try for yourself if you don't believe my experimental findings. I was surprised by how much the Record 5 1/2 deflected under quite moderate forces, and in consequence I'm fairly confident in asserting that the plane could be a factor - a factor which using technique appropriate to that type of plane corrects, but a factor nonetheless.