A random thought on plane flattening

[Pete W]

I'm a lurker under the Porch; I have subscribed but haven't yet figured out how to post a message there, which is why I'm posting here

Saw a post there from Bugbear on the dangers of the common method of plane-flattening (glass or other flat plate, wet'n'dry, etc) and realised that there's a body of evidence to support his opposition...

Among the other hobbies/interests I don't have time for, astronomy figures high on the list. From whence comes this bit of esoteric knowledge. If you want to make your own telescope mirror from scratch, you buy two round glass blanks (sort of like turning blanks), scatter appropriate abrasive material between them, and grind away in a circular fashion. After some time and effort you, rather miraculously, end up with one convex and one concave piece of glass (a parabolic, concave mirror being the ultimate aim).

So, as Bugbear argues, grinding your plane on abrasive paper is almost guaranteed to produce a concave (or convex) surface, if you do it enough (I don't have the experience to suggest how long is enough).

As an aside, the traditional final grit in the abrasive process of mirror-grinding is common talcum powder. Which always makes me smile when I read people recommending talc as a way of lubricating the surface of cast-iron tables.

 

[Philly]

Pete
You got it-the real knowledge is when to STOP flattening! :lol:
Cheers
Philly

 

[bugbear]

So, as Bugbear argues, grinding your plane on abrasive paper is almost guaranteed to produce a concave (or convex) surface, if you do it enough (I don't have the experience to suggest how long is enough).

ATM practice uses loose abrasive between 2 workpieces. I forget which one (upper or lower) becomes convex, which concave. I know that if the process goes too far, you simply reverse the lower/upper workpieces, and the process drives the other way.

In "normal" plane flattening (ha!) the lower surface is fixed abrasive, glued on a hopefully flat (ha! again) and rigid (ha! a third time) reference.

The plane certainly cannot be worked concave but may well (IME) become convex to some degree.

BugBear

 

[engineer one]

but surely the point is where does it change?

DC suggests that concave is not a bad thing, as long as it
hits at the front and rear and around the mouth. indeed one
might think that those older planes with ribbed soles are
trying to minimise this too.

surely the whole point about marking the sole with marker pen, and then
only grinding until the marks have all gone uniformly will
answer the problem.

bugbear is right, it is a question of when to stop :?: :?:

paul :wink:

 

[andrewm]

Among the other hobbies/interests I don't have time for, astronomy figures high on the list. From whence comes this bit of esoteric knowledge. If you want to make your own telescope mirror from scratch, you buy two round glass blanks (sort of like turning blanks), scatter appropriate abrasive material between them, and grind away in a circular fashion. After some time and effort you, rather miraculously, end up with one convex and one concave piece of glass (a parabolic, concave mirror being the ultimate aim).

Totally off-topic I know but what is about this process that ensures you get a parabolic mirror? Common sense seems to suggest that what you get is curved (as in a portion of a circle) such that it is the same radius at each point. But this obviously doesn't happen if this technique is used to produce parabolic mirrors.

Andrew

 

[bugbear]

Totally off-topic I know but what is about this process that ensures you get a parabolic mirror?

It doesn't. That's the "ultimate aim". A perfect curve of the desired radius is merely a stage in the process.

BugBear

 

[bugbear]

but surely the point is where does it change?

Only in the context quoted - 2 workpieces with loose abrasives. This is NOT the context of "woodworker's lapping" on a plane.

DC suggests that concave is not a bad thing

DC says that minor CONVEXITY is ok, because of the common tendency to plane the workpiece convex. A slight convexity in the plane provides a countering tendancy. C.F. Sprung joint, where one is aiming for a concave workpiece.[/quote]

surely the whole point about marking the sole with marker pen, and then
only grinding until the marks have all gone uniformly will
answer the problem.

This merely guarantees that every part of the sole has been touched by abrasive - this can easily be the case for a convex sole.

BugBear

 

[Anonymous]

After some time and effort you, rather miraculously, end up with one convex and one concave piece of glass (a parabolic, concave mirror being the ultimate aim).

So, as Bugbear argues, grinding your plane on abrasive paper is almost guaranteed to produce a concave (or convex) surface, if you do it enough (I don't have the experience to suggest how long is enough).

So if we put loose abrasive between two planes' soles, we can make compass planes! :lol:

As BB pointed out, with the abrasive paper method only the plane sole is being abraded. However, abrasives break down and the backing compresses so there is a limit to the degree of flatness obtained. Of course, there are also limits to the flatness needed for woodworking.

 

[bugbear]

However, abrasives break down and the backing compresses so there is a limit to the degree of flatness obtained.

You forgot swarf build up!

BugBear (who knows that Roger knows this)

 

[ydb1md]

After some time and effort you, rather miraculously, end up with one convex and one concave piece of glass (a parabolic, concave mirror being the ultimate aim).

So, as Bugbear argues, grinding your plane on abrasive paper is almost guaranteed to produce a concave (or convex) surface, if you do it enough (I don't have the experience to suggest how long is enough).

So if we put loose abrasive between two planes' soles, we can make compass planes! :lol:

So why do we rub waterstones together to flatten them? Mine don't come out dished. :-k

 

[bugbear]

So why do we rub waterstones together to flatten them? Mine don't come out dished.

Actually, they may well be dished. The question is - what's the radius?

To expand - nothing is ever "straight" or "flat" or "true" or "square" - precision engineers (who come closest) never use those words; they always talk about "tolerance", and guaranteed minimum deviation from those concepts.

Only woodworkers (*) think their Starrett squares are "perfectly accurate"; Starret make no such claim

BugBear

(*) yeah - I'm generalising

 

[Alf]

It is probable that the new plane will have its sole planed perfectly true. Try this with a straight edge. Surely this is as it should be - the face of a plane should be perfectly true? Well, the aim in planing is to get the surface of your rail or board true. If the plane's sole is quite flat this is less easily accomplished than if the sole from end to end is slightly round. From the mouth to the heel take a few shavings off so that, when tried with a straight edge it is round to about 1/16". I always treat my planes so.

Cheers, Alf

Flat Sole Society? Just say "NO!" :wink: :lol:

 

[bugbear]

In 1918 (or maybe 1928) The woodworker said

(checked - it was 1918)

Trueing Sole of Plane

E.S.W. (Henley) asks the correct method of trueing the
sole of a plane

REPLY.- The correct method of trueing up the sole
of a plane is by testing it on an engineer's surfacing
plate. We presume that you desire to true up the face
of a wooden jack, trying or smoothing plane. Take a
little powdered dry red lead and mix it with oil to the
consistency of cream. Smear this spareingly on the
surface plate and then rub the sole of the wooden plane
on the surface plate. The high portions of the the sole
of the plane will take colour, whilst the lower portions
will remain clean. Turn the plane sole-upwards and
grip it firmly in the bench vice. Take a finely set and
newly sharpened iron-faced plane and proceed to plane
away the portions which you have proved to be high. Then
again rub the sole of the plane of the surface plate,
and repeat the operation until a true surface is obtained.
Remember, however, to test the sole of the plane to see
that it is square with the sides by using an ordinary 6
in. try square.

Probably you may not have access to a surfacing plate, and
as a substitute you might use the iron face of a circular
saw bench or one of the tables of a power planing machine.
Should you not even have access to these you will have to
test your work with winding laths and a steel straight edge
and try square. If you have not got an iron faced plane,
you would be well advised to take it to the workshop of a
practical cabinetmaker, and he would compete the work in
a satisfactorymanner for about ninepence. The method of
thus up an iron or steel faced plane is by the same method;
that is, testing it on a surface plate and scraping it
down where required.

Maybe it's like the bible - you can find a quote in favour of anything!

BugBear

 

[Alf]

Maybe it's like the bible - you can find a quote in favour of anything!

Ah ha...

Of course the Earth is flat, but I like a nice banana-shaped sole on my plane. Does a good job, eh? <indicates large ark containing all the creatures of the Earth therein> Never would have got it finished if I'd listened to Ham and his flaming nonsense about inventing a saw blade that goes round in a circle while held on a table, just so he could use the table to flatten his Nile Valley and Lea Nileson fancy wooden planes. And don't get me started on Shem's idea of putting the big cats next to the gazelle pens...

Cheers, Alf

 

[MikeW]

"Flat" soles on a medium that can "move" over time and conditions, being used to work a medium that itself changes seasonally--and sometimes within a single day.

I think that trying to obtain relative flatness is all thats needed--and probably more important is making sure the plane's sole is not twisted. Obsessing over a tolerance only measurable with feeler gauges or prussian blue on referance plates is futile.

A plane flexes in use. The more the pressure, the more the flex. How does one "fix" that? Plan on covexity keyed to how much a particular person presses down on their plane? Different pressures and lengths of soles to account for resulting in lesser or greater amounts of convexity?

Well, gotta go out to the shop. Need to make room for...

Take care, Mike

 

[Anonymous]

Actually, they may well be dished. The question is - what's the radius?

To expand - nothing is ever "straight" or "flat" or "true" or "square" - precision engineers (who come closest) never use those words
(*) yeah - I'm generalising

The first thing we do is kill all the engineers!
(Hey if Alf can rewrite the Bible, I can mess with Shakespeare :lol: Very funny, Alf! (hammer) )

Even a straight line can be described as an arc with a radius of infinity.

 

[Alf]

Well, gotta go out to the shop. Need to make room for...

Yes...? I hate it when you do that. Of course that's why you do it. ](*,) :lol:

The first thing we do is kill all the engineers!

You sure we're not related, Roger? :-k :lol:

Cheers, Alf

Who likes engineers really. Some of my best friends are - actually, no, I tell a lie... :wink:

 

[Chris Knight]

Well, gotta go out to the shop. Need to make room for...

A Coordinate Measuring Machine of course. Everyone knows that a surface plate and prussian blue depend totally on smearing the right thickness of blue on the surface - a CMM has no such limitations - ask Tony.

Mike is only fooling about non-flat planes, his cabin in the woods could never have been built otherwise :wink:

 

[Anonymous]

"A plane flexes in use. The more the pressure, the more the flex. How does one "fix" that?
Take care, Mike

I agree. This is a question I would like to see answered. It would seem though, that the plane couldn't flex more than the stock is out of flat. Or put another way, as the stock gets flatter the plane would flex less.

In any case, I have a simple test. If the plane performs as needed, I leave it alone. If not, I tune it until it does. Most of my planes have not been flattened at all. Some I have flattened on a sander belt on float glass until marker applied to the sole was removed and some I have spotted on a granite surface plate. Just enough to get the job done.

 

[Anonymous]

You sure we're not related, Roger? :-k :lol:

Cheers, Alf

Who likes engineers really. Some of my best friends are - actually, no, I tell a lie... :wink:

Of course we are related, Cousin Alf!

Roger
Who has counts many engineers in his lists of "Most Admired" and "Most disliked"