Sole flattening - Paul Sellers

[David C]

I believe Richard Jones bends planes, but I prefer the cambered edge approach, as it is easy to teach.

The flatness of Granite tiles bothers me too.

David

 

[Harbo]

And it was not supported very well?

Rod

 

[JohnCee]

And it was not supported very well?

Rod

Yes, but at least the cap iron was tightened down.

 

[Jeff Gorman]

Custard wrote:

Karl Holtey says most woodworkers would be astonished if they realised how much the sole of the average plane can deflect in use.

The proof of the pudding is in the testing as at http://tinyurl.com/7oeqdez

I wouldn't be surprised, for similar body shapes, to find that mild steel deflected more than cast iron, but this is only an instinctive response.

Jeff
http://www.amgron.clara.net

 

[Corneel]

How about a wooden plane? I think it is a lot harder to bend a 3" square block of beech.

 

[Sgian Dubh]

I believe Richard Jones bends planes, but I prefer the cambered edge approach, as it is easy to teach.
David

David, like you, my plane blade cutting edges are generally curved slightly, but I still bend a plane's sole when doing so is useful. This is particularly the case with my no. 7 try plane, rarely so with no. 5 jack planes, and never intentionally with something like no 4 smoothing planes. I don't use a no. 6 plane: I don't even own one, so I guess this is a size for which I've never found a purpose. Slainte.

 

[Cheshirechappie]

Custard wrote:

Karl Holtey says most woodworkers would be astonished if they realised how much the sole of the average plane can deflect in use.

The proof of the pudding is in the testing as at http://tinyurl.com/7oeqdez

I wouldn't be surprised, for similar body shapes, to find that mild steel deflected more than cast iron, but this is only an instinctive response.

Jeff
http://www.amgron.clara.net

Interesting - thanks for posting that, Jeff.

According to classic beam theory, the deflection of a beam is proportional to the cube of the length (that's why 3' of 6"x1" PAR redwood seems quite stiff, but 10' of it flaps about). So a longer plane will deflect much more than a shorter on under a given load.

Another thing that beam theory tells us is that stiffness is proportional to the cube of beam depth, so thicker castings will be much stiffer. (That's also why joists are always edge-upwards - much stiffer that way.) A very 'quick and dirty' check on my LN 5 gives a sole thickness of 3/16" full, and the Record 07 gives about 1/8" - so the resulting increased sole stiffness of the premium planes may be a factor in their better performance. (I don't have a Clifton, Veritas or Quangsheng to measure, but I suspect their soles are quite meaty like the LN's.)

It's also probably true that a wooden plane with a 3" square section will be less subject to deflection than a 1/8" thick (or thinner) cast-iron soled plane. However, the woody will be more likely to distort with atmospheric changes, and will wear more quickly, so there are advantags and disadvantages to either. I suppose by trying to get the best of both worlds, it might be possible to make a much thicker metal soled plane, which would be much stiffer, more wear-resistant and less subject to atmospheric-induced distortions than the woodies, but it might be a very heavy plane to use!

On balance, given the very fine work that has been done and is being done with both wooden and metal planes, I think that in plane design, we have found the happy balance between theoretical perfection and practical pragmatism that works best. It would be possible to design a plane that was so stiff as to almost foolproof in use, but it may be so heavy as to be unusable. It is possible to manufacture to insane degrees of flatness, but as we have seen, normal use imposes deflections that negate that. The secret of success is the combination of pragmatic design, pragmatic manufacture and good technique in use.

Which basically pretty much confirms what we all 'knew' anyway.

 

[Harbo]

Are you talking about bending moments here?
They are dependent on span but your argument seems to be based on the idea that the plane is only supported at the front and back edges.
The actual span will vary according to irregularities in the sole and the profile of the wood being planed. No hollows - no BMs as it is being fully supported?
It will however be subjected to forces from the front and rear totes and any twisting that the woodworker might induce ?


Rod

 

[bugbear]

Which basically pretty much confirms what we all 'knew' anyway.

Absolutely! Any theory which shows that something which is demonstrably working can't work is clearly a theory with some "issues".

Conversely, the laws of maths and physics apply wether you care about them or not.

Personally I find it of great interest to understand as fully as possible why and how something works.

BugBear

 

[Cheshirechappie]

Are you talking about bending moments here?
They are dependent on span but your argument seems to be based on the idea that the plane is only supported at the front and back edges.
The actual span will vary according to irregularities in the sole and the profile of the wood being planed. No hollows - no BMs as it is being fully supported?
It will however be subjected to forces from the front and rear totes and any twisting that the woodworker might induce ?


Rod

I'm not trying to make any precise or analytical point, here. Just that longer planes will deflect more than short ones under a given load, and that quite small increases in plane sole thickness (and perhaps more significantly, deeper side webs) will stiffen them up quite a bit.

Jeff's experiments showed that applying loads of the sort that you might generate in normal use will deflect the sole of a Record 05 by detectable amounts, certainly enough to nullify, or certainly to influence the effects of, 'last thou' plane sole flattening.

It's perfectly true that under normal working conditions, there are a lot of variables - elasticity of the workpiece, plane design, the exact position and magnitude of loads imposed by the operative. A plane sole is a fairly complex beam, but it's still a beam in effect. Doesn't really matter what sort - the point is that it'll deflect, both under it's own weight, and under normal loads in use. The exact design of the side webs will have a significant effect on overall stiffness, too.

It may be possible to analyse exactly what happens to a plane in use, using such analytical tools as finite element analysis. However, I don't think there's much point. We know how to set up planes and use them to give acceptable results. A reasonably flat sole helps to that end. In the end, that's all that matters.

If Mr Sellers gets acceptable results with his planes, then don't knock his method. Other people get equally acceptable results with other variations of sole-flattening methods. Plane soles concave in length or width don't work well, soles very slightly convex in length can do - and Sellers seems to indicate that soles slightly convex in width work OK as well. Maybe, as long as sole flatness is in the right parish, the last thou or so doesn't matter as much as sound planing technique.

 

[bugbear]

If Mr Sellers gets acceptable results with his planes, then don't knock his method.

Some people get better after homeopathy

BugBear

 

[Racers]

So if I dilute my grit down to where there is none left, it will flatten a plane in next to no time :wink:

Pete

 

[János]

Hello,

The behaviour of a cast or dovetailed bench plane resembles more closely the behaviour of an "u" channel. By the way, this talk about 0,01 mm and smaller deflections and tolerances in a hand woodworking context is not a really valid one. Wood is a very resilient material, and actual deflection of a wooden workpiece might be much larger than any deflection/bend in the hand tool itself. In fact, the only really serious flaw/issue of a bench plane, the one which renders it useless, is a lengthwise warp of its body. In fact the cast iron Stanley planes I have slowly but permanently change their shapes, under the static load of the installed frog/blade assembly and knobs. The area behind the blade bulges out again and again, a little twist develops in the sole etc...

Have a nice day,

János

 

[Mr T]

Hi

Watching the video it is interesting that Paul doesn't actually test whether the sole is in fact flat after his treatment, he just tells us it is.

On this flexing of the sole. I sharpen my blades straight across but find I can get a laterally tapered shaving by changing the position of the plane on the wood (I have a video on Facetube about this) . Is this because I am flexing the plane as Sibn Dubh suggests? I don't know, I just know it works!

Chris

 

[bugbear]

Hi

Watching the video it is interesting that Paul doesn't actually test whether the sole is in fact flat after his treatment, he just tells us it is.

His justification/rationlisation is that the scratch marks (and the removal of his magic marker lines) show that the plane's sole has made a contact (presumed to be uniform) with a lapping surface (presumed to be flat).

BugBear

 

[David C]

Chris,

I wonder if your plane sole is convex in its width?

With a straight edge blade, this would give heavier shavings towards the sides.

Do you move your plane towards or away from the high side of a 3/4 to 7/8 inch wooden edge?

(In my workshop, if it hasn't been tested it's not finished).

best wishes,
David

 

[bugbear]

I sharpen my blades straight across but find I can get a laterally tapered shaving by changing the position of the plane on the wood (I have a video on Facetube about this)

Can I have a simpler clue (or link) please?

BugBear

 

[Mr T]

Hi

Bugbear, the link is:

http://www.youtube.com/watch?v=TJfD...DvjVQa1PpcFNQggsCSQRqkfYKlHz1cg5Y6Hqrg-o2y8Q=

David C., the plane sole is flat across the width. I know you suggest that tapered shavings cannot be taken with a non cambered blade, all I can say is it seems to work. Is it to do with weight distribution or the flexing of the sole suggested earlier?

Chris

 

[Sgian Dubh]

... I know you suggest tapered shavings cannot be taken with a non cambered blade, all I can say is it seems to work. Is it to do with weight distribution or the flexing of the sole suggested earlier?
Chris

Neither of those I suspect Chris because I bet your blades aren't sharpened straight across. More than likely your sharpening technique includes the normal practice of bevelling or slightly rounding over the corners so that they don't dig in which would leave a sharp ridge on each pass of the plane. This bevelling will be enough to do what you describe and demonstrate in your video.

I used to be a straight cutting edge type myself, with the aforementioned corner bevelling, and did pretty much as you demonstrate. Nowadays my blades are generally sharpened with a very slightly curved edge, but I still bevel the corners a little more, so I do pretty much exactly what you do.

When it comes to plane sole flexing I mostly use this to bend the sole lengthwise so that I can deal with hollows and humps on either the face or edge of a board. Twisting or torsioning a sole deliberately to get a specific effect is less common on my part because it's a technique primarily for straightening an edge that's high on one side, and as you demonstrate in your video there are alternative ways of achieving that end result. Slainte.

 

[woodbloke]

David C., the plane sole is flat across the width. I know you suggest that tapered shavings cannot be taken with a non cambered blade, all I can say is it seems to work.

Chris

Same here...my blades are always honed flat (difficult to do anything else with a Kell guide) I've never had any sort of problem in planing an edge and do it in the same way as it Chris's clip...'cept I don't put the plane sole down on the bench :lol: - Rob