Plane Whispering: of bevel, angle, and frog

[ivan]

Thanks again for your thoughts. I am both mostly self taught and still learning. My interest is an expression of curiosity, and hope for improvement.

Is it any use? Only for a smoother, if you love the just-off-the-blade finish. If there's never a need for anything better, why do some folk persevere with ancient Japanese hair shirt planing technology, or go weak at the knees at the sight of an infill plane? Just because it's unaffordable?

Doesn't the measuring device wobble about? Well, the results are repeatable, and the dial gauge is clamped to the plane with a heavy, switchable magnetic clamp designed for engineering use.

The blade movement isn't large, but quite significant when compared to the thickness of a shaving. Because the blade is bending under load (as a loaded beam) the cutting edge is digging in by about the same amount as the blade lift.

Thus the Stanley blade can dig in by up to double the shaving thickness. This is reduced by a factor of ~4 by fitting a 2 piece cap iron, or a standard LN blade (3.5mm)

As DC says, the 2 piece cap iron provides its extra clamping point at the 'hinge', about 30mm up from the cutting edge. This shortens the unfixed blade length from 100mm (edge to lever cam) to 30mm, a factor of about 3. Blade stiffness WRT bending is proportional to length cubed, so a properly clamped Clifton 2 piece stiffenes the blade by 3x3x3=27 times. The downside is the lever cap is a bit too long to press in the right place (too much at edge, too little at hinge) as in practice the increase in stiffness is around only ~4 times. A couple of old lever caps are coming my way to experiment with.

Skewing the plane reduces the load on the blade by reducing both load angle and shaving width.

For comparison, a japanese blade is very much thicker/stiffer, it is held firmly along the long edges by the body (not the chipbreaker) the

chipbreaker is sharpened to 75 deg and set just a hair from the edge ...the mouth is set no thicker than a sheet of paper.

(FW Dec 2000)

That's about 4 thou - I thought a 15 thou mouth was tight. Alas I have no japanese plane to examine, but despite a lower bedding angle the shaving is still being turned through ~10+25+75=110 deg within a couple of thou of the cutting edge.

 

[MikeW]

Doesn't the measuring device wobble about?

Well, the results are repeatable, and the dial gauge is clamped to the plane with a heavy, switchable magnetic clamp designed for engineering use.

I am fairly sure the magnetic base is affixed securely, or at least can be. If this is a dial gauge then there is basically a rod which at its tip has some sort of end, often a ball end. Once it enters the body, there is a dampening spring aside from the means of driving the needle. In the ones I have seen disassembled, it is akin to a screw and gear. Rack and pinion basically. Then there, among the other bits, often is another spring much like on a clock.

My point is, that while there is a mechanical connection between the measuring pointer and the needle, there are tension springs. These along can induce a wobble in the needle. There is a certain inherent range of error.

Too, they are made to be fixed mounted and the thing being measured pass over/under them...and or if the base used is affixed to a movable surface plate or other moveable device, one has to take into account the amount of error induced by both the transport mechanism and the surface vagaries of the thing being measured.

In this case, the measuring gauge is mounted to a thing which moves across a fixed or semi-fixed object and there are many variable which would induce error in the measurement. Pressure gain and release on the movable thing--a hand plane--the cutting of this fixed or semi-fixed surface--the wood.

Now, that may sound negative. It is not meant to be so. It is that in any measuring system one needs to account for error induced by the method and tools. Without having a firm grasp of how much error is or is not induced one does not really know the validity of the measurements. So what I do not know from the thread is where you have identified error and accounted for it.

What I do know experientially is that thin blade do chatter. I do think this chatter is because of blade flex. This is one reason I do not use thin blades. And again, experientially, I do note that discernable chatter is not an issue because using thick blades.

But I don't know that scientifically. So that is one reason why I believe what you are doing has validity. At the same time, I think the method needs refined.

Take care, Mike

 

[ivan]

I see what you mean, MIke W. The pointer movement can also be seen statically, if you extend the blade for a silly deep cut and push with typical planing pressure. I wouldn't claim an accurate result, but probably well within an order of magnitude. It seems to confirm Jim Kingshott's comment (in his toolmaking book)that Norris blades at 5/32" are the thinnest he'd reccomend, suggesting something nearer to 1/4" would be optimal if you have put a lot of work into making your plane. It remains to be seen if a standard LN or Clifton blade can be made as rigid by my modified LN chipbraker*, or if as Steve suggests, you may need infill-firm blade clamping too.

I have been reading Odate's book on Japanese tools over the holiday. Interestingly, he says it's not the usual plane that's used for fine finishing cuts, it's similar, bevel down, but has no chipbreaker. To enhance the finish (and control tear out if present) the wood is wiped with a damp rag to swell and so stiffen (especially softwood) the fibres to improve the cut. I wonder if a damp shaving may also be more flexible and so less likely to promote forward splitting .

For really difficult timber he talks of a bevel up plane, which also supports most of the bevel so reducing vibration. This is bedded at ~35deg with a bevel of~30deg (for hardwood)

He shows planes with a mouth that is a few mm long, which he says is just a little thicker than the shaviing, and parallel with either the chipbreaker's second bevel (if fitted) or the blade surface (if not) Thus bevel up planes would have a mouth at the same, or slightly greater, angle as the blade bevel.

* LN chipbreaker ground flat (existing turn over removed) and a 3mm rolled burr dressed tight to the blade instead. Examination shows the standard LN lever cap will clamp this chipbreaker and blade to frog in intimate contact.