Yet another sharpening question

[Alex Liebert]

Here's a quote from Paul Sellers' blog:

Firstly, and I have already said this elsewhere, plane irons need not be dead flat at all.
Secondly, they need not be replaced with any other make of iron and certainly not thicker ones.
Thirdly, if they are bellied they need not be abraded to flatness.
Fourthly, if they are hollow they are ready to go and need minimal restorative work beyond minor abrading and polishing out along the back of the cutting edge.
There, I have just saved you an hour or two’s work.

Could someone explain his point here? If the back of the blade isn't flat, how can it be polished completely across its bottom edge?

[Stewie Simpson]

Alex; the following is from Paul Sellers blog. Forum rules prevent me from including the link. Controversial topic.Nothing further to add.

Look at this. I took a clifton plane iron from a brand Clifton plane and took a straightedge to the iron with the StaySet cap iron on the blade. The bend in the iron was then 1/32″. I did the same with a Lie Nielsen and that one bent a little more also. Loading them in a conventional plane and adding normal clampdown pressure from the lever cap they both conformed to the flatness of the bed of the frog regardless of the frog type. No plane iron is flat once the cap iron is applied with normal pressure of tightening the setscrew as tight as you can to unite the cutting iron assembly.


So, when you look at the normal pressures applied by the lever cap to the cutting iron assembly (the cap iron and cutting iron assembled), and notice that when the lever cap is indeed set and clamped down, the whole of the assembly is conformed to the frog. This is the same no matter the frog type, the maker or the thickness of irons. The only reason we need the flat face of the cutting iron to be near to flat is that it is easiest to get the cutting edge to the abrasive. Beyond that the full assembly of the iron in the plane conforms it readily to the task of planing perfection. You don’t technically or practically need dead flatness to the large flat face of your irons so I think that is quite freeing.

[Simon MacGowen]

Simply note Paul's use of "need not dead flat" or "near to flat." He is not saying any unflat iron can work.

When woodworkers demand tools satisfying tolerance to the tune of "0.0015"or even higher (and they are willing to pay the price or time to get that for their woodworking purposes), I laugh.

We have too many woodworkers willing to be led (or misled) by engineers or marketers as to how woodworking should work. Or, how wood actually works. They spend 5 times or even 10 times or more money for that premium rule or square or whatnot, although they know their work wouldn't be twice better. They put 5 hours (if not more) trying to set the tablesaw 0.001" better in tolerance even though they know they won't see any difference in their cuts, even if they achieve that.

I am a woodworker and not a slave to tolerance or flatness or squareness or straightness. If my plane iron is not dead flat, I wouldn't know it as long as my work doesn't show it.

Simon

[Pat Barry]

Simply note Paul's use of "near dead flat" or "close to flat."

When woodworkers demand tools satisfying tolerance to the tune of "0.0015"or even higher (and they are willing to pay the price or time to get that for their woodworking purposes), I laugh.

We have too many woodworkers willing to be led (or misled) by engineers as to how woodworking should work. Or, how wood actually works.


Simon

Right, engineers should stick to engineering, not woodworking, or how wood as a material actually works. We all know woodworking has been going on much longer than engineering and engineering has contributed little to woodworking, right?

[Andrew Seemann]

To take the engineering analogy further, the woodworking "gurus" have a habit of specifying things like flat, level, or square in prose without giving any tolerances.

You see phrases like "absolutely flat" or "perfectly square" or "completely level". Yet when you look at their pictures or videos, they aren't using micrometers, surface plates, ground straight edges, etc; they are squinting one-eyed at their hand planed bench with a couple of sticks (winding sticks, if you will) on it, using cheap rulers, pieces of granite, glass, that might be +/- 0.005" The "guru" might know how flat something really needs to be (or they may not, some of them are not as knowledgeable as we or they think) but they don't always remember or know how to to say it.

[Jim Koepke]

Could someone explain his point here? If the back of the blade isn't flat, how can it be polished completely across its bottom edge?

There are two dimensions involved when discussing blade flatness. One is from cutting edge to the top of the iron. This doesn't need to be perfect. Many planes that have come my way have had the tension between the cap iron and blade increased. This can cause the blade to bow and not fully mate to the frog. My choice is to undo this by working on the cap iron to decrease this extra tension. If there is too much bow in a blade, that can also be worked out with care and a little force.

The next dimension is edge to edge. If it mates to the cap iron without gaps to catch and jam shavings, then it is good enough. If one cap iron is getting swapped around on different blades, it helps if they are all 'relatively' flat, at least flat enough for a 1/4" or so to mate with the cap iron.

jtk

[Osvaldo Cristo]

Simply note Paul's use of "need not dead flat" or "near to flat." He is not saying any unflat iron can work.

When woodworkers demand tools satisfying tolerance to the tune of "0.0015"or even higher (and they are willing to pay the price or time to get that for their woodworking purposes), I laugh.

We have too many woodworkers willing to be led (or misled) by engineers or marketers as to how woodworking should work. Or, how wood actually works. They spend 5 times or even 10 times or more money for that premium rule or square or whatnot, although they know their work wouldn't be twice better. They put 5 hours (if not more) trying to set the tablesaw 0.001" better in tolerance even though they know they won't see any difference in their cuts, even if they achieve that.

I am a woodworker and not a slave to tolerance or flatness or squareness or straightness. If my plane iron is not dead flat, I wouldn't know it as long as my work doesn't show it.

Simon

Very well said, Simon!

BTW I am an Engineer... :-)

A good Engineer will know the precision required for each task. I developed micro chips at 1980s and it required different precision than my tasks in the woodworking shop.

All the best.

[Warren Mickley]

I don't think Sellers' standards are very high. After planing the wood he sands at 240 grit, then applies a sanding coat of finish to raise all the damaged grain and then sands again with 240. With a plane in good order there is no damaged grain and no sanding necessary.

The iron does not need to polished on the back if you don't care how easily the plane works or what kind of surface you obtain.

[Ted Phillips]

Right, engineers should stick to engineering, not woodworking, or how wood as a material actually works. We all know woodworking has been going on much longer than engineering and engineering has contributed little to woodworking, right?

I usually keep my "Woodworking Hat" in the basement. When I get home from work, I take off my Engineer Hat and put it in my briefcase for the night. I put on my Dad/Husband Hat for a little while... but when I get to the basement workshop, on goes the Woodworking Hat. It doesn't fit to within a 0.001" tolerance for my noggin (unlike my Engineer Hat), but it is comfortable and gets the work done...

TedP

[Andrew Pitonyak]

Here's a quote from Paul Sellers' blog:

Firstly, and I have already said this elsewhere, plane irons need not be dead flat at all.
Secondly, they need not be replaced with any other make of iron and certainly not thicker ones.
Thirdly, if they are bellied they need not be abraded to flatness.
Fourthly, if they are hollow they are ready to go and need minimal restorative work beyond minor abrading and polishing out along the back of the cutting edge.
There, I have just saved you an hour or two’s work.

Could someone explain his point here? If the back of the blade isn't flat, how can it be polished completely across its bottom edge?

If you use the "ruler" trick. You want two edges to meet. You are not, however, registering the back of a blade against your work like you might do with a chisel.

I prefer a flat back, but, I have one plane blade that after lots of work, one corner is still there. So, I did the ruler trick with it and moved on with my life. Great plane, crappy finish on the blade.

[Simon MacGowen]

I don't think Sellers' standards are very high.

That may or may not be true. But what is true is that 95% (if not more) of modern woodworkers I have come across do not have standards higher than his.

Unless elitism is the issue, Sellers' standards are good and high enough, not to mention that his work is in the White House's collection. For a fact, I know Sam Maloof's work is also among it.

Simon

[Alex Liebert]

I think I just find Paul's language hard to interpret, especially in written form.

If he's saying the blade doesn't need to be straight along its length, or if he's saying that only the very tip of the back where it meets the bevel needs to be flat, that makes perfect sense.

It sounded like he was saying that a cup in the back of the blade need not be flattened at the tip, and if that's the case I don't see how that could be. He doesn't advocate the ruler trick, so if he's polishing the edge of the back of the blade but it has a cup in it, surely only the corners of the edge would be getting polished at all.

[Alex Liebert]

I don't think Sellers' standards are very high. After planing the wood he sands at 240 grit, then applies a sanding coat of finish to raise all the damaged grain and then sands again with 240. With a plane in good order there is no damaged grain and no sanding necessary.

The iron does not need to polished on the back if you don't care how easily the plane works or what kind of surface you obtain.

Interesting, and I think it speaks to what I'm getting at.

It seems Paul often hints at the notion that, the focus on precision flatness in tools is a fairly modern thing and not as critical as the web might make one think. I'm playing with a few planes I picked up very inexpensively at a flea market recently. I'm looking at the cutting edge of the blade on one and the back is quite rough, and the bevel is all over the place. It's also almost worn down to the slot, so if nothing else, this tool was used extensively. Was this tool used to produce fine work with an approach to maintenance that's less finicky than today's but still very effective? Did it only ever produce results that as you suggest were not to a high standard?

Bit of a tangent, not trying to stir the pot or get specific restoration tips here- just an example and I'm still curious about where the line is between the precision tuning and fettling in articles and videos, and just how much of that, if any, was ever being done 50 or 100 or more years ago (when beautiful furniture was clearly being made by hand, one way or another.)

(If it matters, my education is in fine art and my professional background is software engineering...puts me a bit at odds when evaluating anything like this)

[Andrew Seemann]

I don't think Sellers' standards are very high. After planing the wood he sands at 240 grit, then applies a sanding coat of finish to raise all the damaged grain and then sands again with 240. With a plane in good order there is no damaged grain and no sanding necessary.

The iron does not need to polished on the back if you don't care how easily the plane works or what kind of surface you obtain.

I think the main thing is whether his standards are high enough for what he is trying to accomplish. You can always have higher standards, but you get to a point (sometimes quite quickly) where higher standards don't make the work better, they just make it more expensive. That whole law of diminishing returns.

[Patrick Chase]

If you use the "ruler" trick. You want two edges to meet. You are not, however, registering the back of a blade against your work like you might do with a chisel.

True, but to get the most out of a BD plane you *do* need to register a cap iron against the black of the blade such that it doesn't create a fiber trap, and IMO that creates tighter requirements for transverse flatness than does a chisel.