Question for Lyn related to planing and finish

Re: Question for Lyn related to planing and finish

Lyn J. Mangiameli

>A question I actually have considered regularly in my own work. For cocobolo and thuya burl and some other woods of natural oil and high density, I usually leave them unfinished, but for high speed buffing and sometimes some caranuba wax buffed on.

For most other woods, like blackwood acacia, I use the LV/Sutherland Wells Polymerized Tung Oil Sealer. It penetrates deeply, dries quick and is very slow build, which is good so I never overdo it. It is probably my favorite finish, and on some woodturnings I've sometimes applied as many as 20-30 coats, then power buffing them out.

When I just have to have a film finish, I use shellac, often after a coat of PTO sealer.

I have the ability to spray lacquer and varnish, but I only use the former over/as paint. I only use varnish (usually water based urethane) for shop furniture where it is just disgusting how much it muddies the surface.

Can't say there isn't a better way, but this is what has come the closest for me.

Re: Question for Lyn related to planing and finish

Frank D. in Montreal

>Hi Lyn,

I'm a big fan of yours and appreciate your input to this forum.

I think a very important question as regards to finishing and planing is whether a planed surface gives finished wood a better appearance than a sanded surface. If not, then it seems to me that all the care given to the final surface by hand planes is overestimated. I know this is a relatively naive question but I was wondering if you have noticed a difference in your experience.

I promise in the next couple days to test a few boards and compare my results; I have put of this simple test for too long. I usually finish everything I do with 220 sandpaper (hand sanding) to render the surface uniform, because I don't have the skill necessary to plane, assemble, and clean up without sanding down to make the final surface uniform. I was wondering about your thoughts on this (or anyone else's who has taken the time to compare).

Frank

Oops,

Frank D. in Montreal

>Just to be more precise I'm talking about the difference in appearance of a planed/sanded surface finished with a film like shellac or lacquer, or an oil like tung or BLO. These are common finishes, and while finish certainly varies depending on the wood and the use a piece will see, I think it is most helpful to base comparisons on these finishes. Of course ebony may not require a finish, or just some carnauba, but I think most woodworkers use domestic hardwoods and commonly available finishes.

Please don't take this as a question to put you on the spot, and if you have any advice to help me in my upcoming comparisons I welcome them. I thought of starting by achieving a planed finish on a board, then by sanding half of it by hand with 220 sandpaper. I would then apply finish and could compare the results side by side.

Re: planing and finish

Lyn J. Mangiameli

>This is a perennial question, and I think a good one. I’d like to take this down to some fundamentals. We all know that both planes and sandpaper can remove wood, but I think it is helpful to consider just how this is achieved.

The plane holds a cutting edge at a fixed distance from a reference surface (blade extension), fixes that cutting edge at a specific angle (effective cutting angle), and maintains a specific gap between the cutting edge and the trailing edge of the toe (mouth size). If the blade is prepared relatively flat across the cutting edge (not crowned significantly, but maybe the corners dubbed), the above factors are constant for the width of the blade. While some blade movement and deformation occurs under load, it is very slight.

Now let’s look at sandpaper. The individual grains form cutting edges. The primary control of depth of cut is selection of grain size. Depending on the size and grading the grains have more or less variable distance from the backing (thus differing extension into the wood). The backing (reference surface) is not rigid both vertically and horizontally, compared to the sole of a plane, allowing the extension of the cutting edge into the wood to differ according to hardness and density of the wood. The surface is prone to loading that also modifies cutting depth. The grains are of varying effective cutting angle (though almost entirely at some high effective cutting angle). The grains are also of variable sharpness and there is almost no control of sharpness except by choosing paper that has seen lesser or greater use.

These factors must then be considered in relationship to the wood characteristics: open or closed grain, varying densities (knots, major differences between seasonal rings, etc.), cell size and structural integrity, etc. Sometimes sandpaper will be superior for dealing with these characteristics, sometimes a plane, sometimes both, and rarely, neither.

What a plane does best, is provide/maintain a flat surface and develop open cells and structures (i.e, under the best circumstances it cuts through the cell wall/structure rather than crushing the sides over and filling them with dust). It deals well with variable densities. This happens because it has so much control over uniform blade extension and removal of waste wood.

Let’s look at how sandpaper can prove problematic in these areas. Sandpaper does very poorly at maintaining uniform removal of wood fibers. Except with sanding blocks of cumbersome size, it lacks the reference surface most soles provide. But most of all, the usual lack of rigidity of the abrasive/block assembly allows for differential removal of wood dependent on differing densities. Wenge is a good example of a wood that has pronounced variation in density between its annular rings. If you sand or even power buff this wood for very long, you will get a rippled surface where the dense dark rings rise up because the softer wood has been differentially worn away below them. Same thing happens with knots, sand much in the area of a knot, and you will soon find the knot becomes a raised island as the softer wood around it has become differentially worn down about it. This also applies to may highly figured woods and spalted woods, where the denser areas of figure, or the softer areas of spalting, differentially wear away when sanded.

Which brings me to the number one criteria I have for surface finish following planing: the surface must be as free of tearout as possible. If you attempt to remove tear out by sanding, you are having to remove a fair amount of wood. That sanding will often result in a loss of flatness due to differential removal of wood according to density variations.

Now if little wood requires removal (like only needing very fine sand paper), this is not an issue, and sand paper can do much to remove fuzziness and improve the sheen of the wood surface. But even here you have to be cautious. Sanding can tend to burnish the cell surface, fill pores with dust, and generally crush over the cell walls. At best, this diminishes clarity, at worst it can greatly affect the wood’s ability to evenly absorb stain and finish. Anyone who has worked with maple much know how these problems can develop if you sand too finely. Areas “close up”, burnish, and you can’t get even acceptance of stain.

Now I’m not against sanding, indeed I regularly sand my woodturnings up to 800 to 1200 grit, and sometimes then power buff them with even smaller abrasives. But I do what I can to minimize sanding on large flat surfaces, especially those of widely varying densities. When I do sand such surfaces, I use a half sheet block, attempt to be able to start at 400 to 600 grit, and sand as little as possible.

So, one’s success with sanding is going to depend a lot on how little wood needs removal, how large and rigid a holding block you will be able to employ, and the uniformity of density of the wood surface. If the surface can be kept flat, and the surface is sanded to the higher grits, the end result from sanding can be stunning, and often as good as a smoothing plane can achieve, but that’s not an easy or sure thing to accomplish.

FWIW, I’m not afraid at all to take very fine Abralon abrasives to a planed surface if I don’t’ think the sheen is all that it might be, or if I’m going to be staining (rarely) and think the stain is going to raise the grain.

Much more could be said, but I’ve got to see a patient right now.

What about scraping?

Thomas Skaggs, Foothills of Mount Level

>How does a scraper factor into the equasion of this question? Just curious.

Tommy

Not mentioned....

bill tindall

>Not mentioned is the ability of a film forming finish to clarify a surface. Sanding tears up the wood surface and leaves it dull looking. A film forming finish flows among these micro tears and reduces the reflections from them. Much like adding liquid to a white powder can turn the white to transparent the finish can brighten the surface of the wood. Some film formers result in a brghter film than others depending on the refractive idex of the film. Nitrocellulose or butyrate lacquer makes an exceptionally bright film while most water based finishes do not.

A question to be considered is whether the obvious superior appearance of a planed surface will look any different from a sanded surface after one's finish of choice is applied. If not, then many of Lyn's criteria for choosing a plane become irrelevant and only avoiding grain tear remains. And one could then consider alternate criteria such as edge durability, or ease of use, or even cost. So it would seem that the best plane may also be tied to the finish that will be used. Hence, plane evaluations can guide a personal choice of what to own, but can't substitute for individual evaluation of the multitude of individual situations.

No one can tolerate grain tear so this factor should be an important factor in all evaluations.

I don't see the problem Lyn discusses about sanding and levelness. I don't disagree with the potential, it just doesn't seem to be a problem with an experienced sander- proper substrate, proper succession of grits (people starting too fine are more apt to induce unlevelness). The only time extreme levelness becomes a critical issue is on high build gloss finishes (what i call piano top finishes) where one needs a uniform reflection from the surface. Irregularities of a few thou. on a somewhat dulled finish can't be seen, but they sure can be annoying to try to plane to a good surface (but easy to sand).

Re: Not mentioned....

Frank Mutchler in Colorado Springs

>Bill, what do you do to prevent the irregular (non-flat) surface that results from moderate to heavy sanding? My experience is that the softer wood sands away much faster than the harder areas leaving an irregular surface.

Re: Not mentioned....

Frank D. in Montreal

>Hi Frank,

I'm not Bill but I have this problem when I work with BC fir. Like with most other woods if you plane to as final a surface possible you only have to give a very light hand sanding. I almost never use 150 grit, only 220 is needed if there are no machine milling marks to take off. I also use the best sandpaper I can find (SIA 1960 or 1962, which cuts like a knife). This doesn't really address your question about moderate to heavy sanding but working with hand tools gives you the advantage of not needing to sand nearly as much. With such a light sanding the rings stay level.

Frank

Re: What about scraping?

Lyn J. Mangiameli

>Ah, scrapers.

The thing about scrapers is that scraping really encompasses two different approaches to material removal. What I primarily consider scraping is to use a blade of extremely high effective cutting angle (like 75-90 degrees) to force shearing of wood fibers. This is basically what happens when you use a scraper plane without putting a burr on the blade, which is what I usually do. In these situations, the cutting edge functions most like a plane, though a lot of sand paper likely has a lot of individual grains with cutting surfaces of similar angle.

Then there are the scraper blades that have a burr rolled onto them. This functions a lot like a high angle blade but with a huge steep chip breaker attached. You basically heel over the blade past 90 degrees to set up the burr to achieve a high effective cutting angle. I'm not super fond of this condition as the burr is weak and virtually impossible to produce uniformly. Still, somtimes it works great. Though in both circumstances, woods of higher density tend to scrape better, just like high effective cutting angle planes.

There are some things about many scrapers that are not found in a plane, like the ability to bow the blade to localize the scraping action. Effective but at the price of localized depressions in the wood surface.

Frankly, I use scraper planes very little since I set up several blades for high effective cutting angle. I use card scrapers more often, but even then, more for woodturning that flat work.

Twas the first day back to work in a long time, and I'm kind of tuckered out, so I'll end here, but feel free to ask if there is something you think I've not touched on adequately.

Re: Not mentioned....

Lyn J. Mangiameli

>Bill comments

"Not mentioned is the ability of a film forming finish to clarify a surface."

Yah, not much said by me about film finishes as I personally don't care for them an avoid them whenever I can, except as and over paint. I much prefer oil and wax finishes that less mask the characterisics of the wood.

Bill going on:

"Sanding tears up the wood surface and leaves it dull looking."

Yep, in the coarser grains, but get into 2-4000 grit "papers" and the micro abrasives found in automotive polishing compounds and you don't have any more ragged a surface than a blade will leave, sometimes even less.

More Bill:

"A film forming finish flows among these micro tears and reduces the reflections from them. Much like adding liquid to a white powder can turn the white to transparent the finish can brighten the surface of the wood. Some film formers result in a brghter film than others depending on the refractive idex of the film. Nitrocellulose or butyrate lacquer makes an exceptionally bright film while most water based finishes do not."

Yep, the most common way we see this is to put a fresh coat of wax on our cars and watch the swirl marks go away---for a little while. Gives great depth too, but at the cost of creating a layer between the underlying surface and the user. Fine for some things, but unappealing to some of use who don't like the "plasticy" surface that removes us one step farther from the wood. Note that I almost never use lacquer except as and over paint. In those instances, I like it a lot. And while the lacquer/other film does reduce apparent scratch marks and tears, it can't increase the clarity of the figure.

More Bill:

"A question to be considered is whether the obvious superior appearance of a planed surface will look any different from a sanded surface after one's finish of choice is applied."

Well not to be too facetious, but yah, if I put black paint with a high solids content over both, I won't be able to tell the difference. :-) More seriously, just conduct your own experiment with a highly figured wood like cocobolo, then report back with the results.

Still Bill:

"If not, then many of Lyn's criteria for choosing a plane become irrelevant and only avoiding grain tear remains."

Tearout and pitting are the single factor that was weighted most highly, and was the most apparent as a means to distinguish between plane configurations. With highly figured, tear out prone woods, cocobolo being one of my favoriate examples, the differences are not subtle.

But I don't agree that it stops there. Clarity of the figure and details of the wood is the second most important to me. No blade makes a perfect cut, but at its best, a blade cuts thought a cell or cellular structure fairly cleanly, sandpaper tends to shred the surface instead. The later leads to sections of the walls collapsing and diminishes the detail available in the wood. Nothing beats a blade cut for maximally revealing the detail of wood features.

Going back to Bill:

"And one could then consider alternate criteria such as edge durability, or ease of use, or even cost."

One should anyway, though as I've tried to point out earlier, other than early failure due to extremely low included bevel angles, there really isn't much difference in edge durability for a given alloy and preparation.

Bill again:

"So it would seem that the best plane may also be tied to the finish that will be used."

I don't think it is so much the finish, as it is those aesthetics you most prize. Finish is just another means to achieve them.

Quite correctly Bill states:

"Hence, plane evaluations can guide a personal choice of what to own, but can't substitute for individual evaluation of the multitude of individual situations."

As it always should be!!!.

And Bill hits the mark again:

"No one can tolerate grain tear so this factor should be an important factor in all evaluations."

Yep, for me and mine, it was the predominant one.

And finally Bill asserts:

"I don't see the problem Lyn discusses about sanding and levelness. I don't disagree with the potential, it just doesn't seem to be a problem with an experienced sander- proper substrate, proper succession of grits (people starting too fine are more apt to induce unlevelness). The only time extreme levelness becomes a critical issue is on high build gloss finishes (what i call piano top finishes) where one needs a uniform reflection from the surface. Irregularities of a few thou. on a somewhat dulled finish can't be seen, but they sure can be annoying to try to plane to a good surface (but easy to sand)."

Wow, we really diverge on this one, and I'm actually very surprised given your experience as a woodturner as well as flat worker. Sandpaper is simply notorious for causing unequal removal of material due to variations in hardness and density. Even with very fine grits, it can be a problem with softer spots like spalt patterns. I suppose if you slop on enough plastic on top of the surface there will be some leveling, but it takes an awfully thick film to do so.

But as always, go with what brings satisfying results for you.

Thanks

Frank D. in Montreal

>Thanks Lyn for such an elaborate response. I'm off to bed so I'll save my questions but this is an issue I intend to pursue.

Re: Thanks

Lyn J. Mangiameli

>Happy to be part of a continued discussion, but might I suggest you consider setting up the discussion in a new thread under a new heading. I'd like to see a wider range of input on the subject and I think something that didn't have "Questions for Lyn" in the lead title might be better.