Sole flattening - Paul Sellers

[Aled Dafis]

I wasn't to sure what to think of Paul Sellers the first time I came across him, but the more I read and watch, the more I recognise that he's a very skilful craftsman with very efficient working practices.

His sole flattening video on Youtube is quite interesting, I'm not sure if I'd go this far, but his arguments for rounding the edges slightly do make sense. It may be why so many old planes we find at car boot sales aren't totally flat, the craftsmen of yesteryear might have known something we've seemed to forget. A bit like the Romans using cement, and then it's use disappearing for a few hundred years.

http://www.youtube.com/watch?v=zQyjLV92 ... ature=plcp

What do you guys think?

Cheers
Aled

 

[Paul Chapman]

Rounded soles; rounded bevels - next they'll be telling us the Earth's not flat...... :lol:

Cheers :wink:

Paul

 

[Richard T]

That's just how I flatten a sole - with a longer flat surface maybe.

As for taking the edges and corners off, I can't remember ever catching with a plane - I think the way I have avoided doing that is to just not do it.
Maybe they will get left very sharp from sole flattening, but I would barely touch them personally - maybe the merest wipe with some emery.

 

[Sgian Dubh]

I recall encountering Paul a couple of times eighteen or so years ago when I lived in Texas. He came across as an uncomplicated type when it came to tools in the sense of just setting them up good enough to work and get on with the job. That philosophy fits in very neatly with the way I like to work.

So, in general what I looked at on that video makes sense to me and follows pretty much what I do with a couple of wrinkles. I stick 60 or 80 grit paper down on to the bed of a surface planer or saw to get the sole flat quickly. You can get long rolls of coarse paper from places like B&Q and elsewhere so you can set up an abrasive surface with a long stroke if you like for particularly out of flat plane soles. I differ again from Paul where I reckon anything with a sole flattened on 80 grit paper is more than good enough to work-- you can go finer if you like, and I generally do just for aesthetic reasons, up to about 180 or 220 grit generally. Anything beyond that isn't unnecessary with regards to ability to do the job.

I've never eased off the outer edges as he does, but what he demonstrates is really no more than an extension of the need to round off the sharp cornered edges all around the sole anyway, which he did at the back end in his demo. The easing off using a rule seems logical and reasonable and I'd say almost certainly won't hurt performance, and could well help too.

I found it interesting the point he made about not bending the sole of the plane during the flattening process. He's right to note that plane soles can be bent of course because I've been purposely bending and twisting plane soles in use to achieve specific effects for decades.

Absolutely perfect sole flatness is not necessary for a woodowrker to achieve excellent results. Concave soles are often a problem and can cause great difficulties in working wood, but slightly convex plane soles are much less challenging, and can even help a bit from time to time. I suppose the reality is that I've always just needed my tools to work effectively, and I've never had the time, need or inclination to worry too much about perfection, including the use of planes with perfectly flat soles, because, as noted earlier I sometimes purposely bend a plane sole whilst I'm using it. Slainte.

 

[richarnold]

All this talk on the forum about how to flatten planes made me think about my own planes I use every day at work. My main workhorse is an early panel plane marked Buck, but is actual a Spiers, so i dropped a straight edge on it yesterday, and lo and behold it's not flat!!!! Its actual convex in it's length. Shock horror!! what should i do? break out the 60 grit and spend hours truing it up. I think not. having used it for years like this, and not noticing any problem with it, i think i will leave well alone. Having said all that I have to say that years ago I made a violin, and when it came to shooting the center joints on the back and belly, i opted to use a shooting board and the longest plane i had which was a no7 record. I seem to remeber spending hours trying to get those damned center joints right, but to no avail. It eventually twiged that the no7 was far from straight, so i tried my trusty 15 1/2 inch early norris panel plane instead. within seconds the joints were perfect!!!!. Now that plane was flat, and still is!!!!

 

[JohnCee]

I never understand this business about having to have the blade fitted when flattening the sole on the assumption that the pressure on the blade changes the shape of the sole in some way. If this were true then how come plane makers surface grind soles without frog/blade fitted? I'm sure this is just an old wive's tale.

 

[matthewwh]

Hi John,

The castings are held in jigs that replicate the pressure applied by the frog and handle screws. The planes are then checked after assembly with all the goodies fitted to ensure that they are correct.

You would be surprised how much difference the pressure from a lever cap makes.

 

[ac445ab]

Ciao,
I stop lapping the sole onto 180 grit abrasive paper. I think so is enough.
I use to round sole corners of smoothers keeping the plane with an angle to the abrasive surface and moving it back and forth several times.

Giuliano

 

[JohnCee]

Hi John,


You would be surprised how much difference the pressure from a lever cap makes.

My planes and straight edges tell me it makes no difference whatsoever.

 

[James C]

Hi John,


You would be surprised how much difference the pressure from a lever cap makes.

My planes and straight edges tell me it makes no difference whatsoever.

That's quite interesting, I suppose that not all planes are the same. It is possible that the slackening off of the lever cap causes the plane to even twist slightly but I guess that could only be due to imperfections in the casting on one side compared to the other.

 

[Jeff Gorman]

Alded Dafis wrote:

> but his arguments for rounding the edges slightly do make sense.<

It is possible to think of a bench plane as running on two skids ie, the 'lateral margins', ie the areas each side of the projecting part of a cambered cutting edge, and running the length of the plane.

A highly critical perfectionist might wonder whether it is a good idea to modify these areas, even if the modification is quite small.

Folk who want to look into the pro and cons of plane flattening might like to look at http://tinyurl.com/3xjpbm2

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

 

[matthewwh]

My planes and straight edges tell me it makes no difference whatsoever.

Best you pop round to the Clifton factory and put them straight then.

Just kidding, have you tried loosening off the frog retaining screws? Normally there is a good couple or three thou of flex between loaded and unloaded.

 

[JohnCee]

the frog retaining screws making a difference is something I'll buy, but not the lever cap.
The chap in the video is not flattening to anything like 2-3 thou tolerances anyway, with his granite tile and his magic marker.
Not that any of this really matters in the grand scheme of things.

 

[Dangermouse]

Well I hope a flat sole gets you into heaven or at least helps you walk straight or even keeps the fish flat on the pan.

 

[David C]

Matthew,

I warned for many years, that excessive torque on the frog fixing screws might distort a Bailey style body casting, just behind the throat. The machining was very poor in the 80s and 90s and the cast metal was very soft. I was sure this happened.

Recently I was restoring some old Stanleys, no5 & 5 1/2 from about WW2, and decided to see if I could detect any deflection, with a test dial indicator and magnetic stand. (After "seating" the frog). I could not.

I then tried varying lever cap tension, as I had seen it suggested that this might have some effect, but could detect no change, as I expected.

Accurately machined Bedrock frog fixing has no effect on the sole. I am quite certain of that.

So was it only new Bailey planes, in the second half of the last century, which exhibited deflection of the sole, and who might have done work to demonstrate this?

Best wishes,
David

 

[bugbear]

I then tried varying lever cap tension, as I had seen it suggested that this might have some effect, but could detect no change, as I expected.

That would be extraordinary - the lever cap would have to flex the sole indirectly, by flexing the frog, which is a VERY thick, blocky casting.

As to knocking off the sharp arris after flattening a metal plane's sole - yes, this is no more than common sense.

But actually creating a substantial "round" at the perimeter seems unwise. The Japanese tradition is to keep plane soles very "square"; the stated reason is so that shavings on the work surface are pushed away, and do not introduce themselves under the sole.

BugBear

 

[Cheshirechappie]

Matthew,

I warned for many years, that excessive torque on the frog fixing screws might distort a Bailey style body casting, just behind the throat. The machining was very poor in the 80s and 90s and the cast metal was very soft. I was sure this happened.

Recently I was restoring some old Stanleys, no5 & 5 1/2 from about WW2, and decided to see if I could detect any deflection, with a test dial indicator and magnetic stand. (After "seating" the frog). I could not.

I then tried varying lever cap tension, as I had seen it suggested that this might have some effect, but could detect no change, as I expected.

Accurately machined Bedrock frog fixing has no effect on the sole. I am quite certain of that.

So was it only new Bailey planes, in the second half of the last century, which exhibited deflection of the sole, and who might have done work to demonstrate this?

Best wishes,
David

David,

To get any conclusive data on sole deflection, you may have to resort to an inspection grade or calibration grade surface plate and a tenths clock. You may have to devise a jig to hold the plane under test, as well. I doubt that a DTI with 0.001" resolution would be sensitive enough to detect deflections of even a try plane, and a workshop grade surface plate (especially the budget ones) may be more 'out of flat' than the plane.

All castings as thin as plane bodies (and in casting terms, plane bodies are very thin castings) will distort under their own weight, and the added loads of frog screws, lever caps and even handle fixings will produce distortions. How much? Enough for some to be significant, others less so; but if the plane sole is finished with those loads applied, it will be true enough, and certainly within the British Standard allowance for out-of-flatness. About a century and a half of experience also suggests that once the plane has been 'set up', whatever deflections there may be are not noticable enough to affect adequate performance.

In practical terms, to avoid any deflection problems, lap the plane sole with the frog in place and the frog screws at working tightness, the blade in place (but retracted) and the lever cap at normal tension. Also pay close attention to the flatness of the lapping surface - the plane flatness can only reflect the flatness of what it was lapped on. Most of us find that a piece of float glass on a clean, flat bench is plenty good enough to give good working planes. If you need anything flatter than that, you may have to resort to surface grinding or similar techniques. (Personally, I don't think it's worth the bother.) Checking HOW flat it is afterwards will need some very fancy metrology kit.

In general, if a plane will take shavings thin enough to be transparent, it's flat. Another test may be to make three straight edges. When all three will mate against each other without showing any light, they are straight, so your plane sole is plenty straight enough.

 

[custard]

Matthew,

I warned for many years, that excessive torque on the frog fixing screws might distort a Bailey style body casting, just behind the throat. The machining was very poor in the 80s and 90s and the cast metal was very soft. I was sure this happened.

Recently I was restoring some old Stanleys, no5 & 5 1/2 from about WW2, and decided to see if I could detect any deflection, with a test dial indicator and magnetic stand. (After "seating" the frog). I could not.

I then tried varying lever cap tension, as I had seen it suggested that this might have some effect, but could detect no change, as I expected.

Accurately machined Bedrock frog fixing has no effect on the sole. I am quite certain of that.

So was it only new Bailey planes, in the second half of the last century, which exhibited deflection of the sole, and who might have done work to demonstrate this?

Best wishes,
David

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

I believe him.

My father (a cabinet maker) used to work with another cabinet maker who used a straight honed iron to true an edge (my father was of the cambered edge school), and I remember seeing this chap physically twist his Stanley 06 or 07 while truing an edge. I remember my father saying he couldn't fault the results, so it just showed there's often different ways to the same conclusion.

 

[bugbear]

Having now watched the video a coupla' times, I have to say that his plane must have been very-near-to-flat to start with, because he got it "dead-true" (he claimed) in around 90 seconds on 120 grit.

More seriously, I question how flat (during use) his lapping surface was, being a thin granite tile, which was only partially supported, under seriously work loading.

BugBear

 

[Sgian Dubh]

... seeing this chap physically twist his Stanley 06 or 07 while truing an edge.

That sounds very similar to something I said in my first post in this thread-- something like three posts down from the opening question. It's an old and valid technique. Slainte.