Planed Finish VS Sandpaper - Microscopic View

[Normand Leblanc]

Lots of very interesting information in this thread.

One more question. Let's talk about gluing those surfaces. My feeling tells me that two planed surfaces glued together would be stronger than two sanded surfaces. When looking at the gluing area it seems to me that it should be the other way around, isn't it?

[Brian Holcombe]

Brian:

Wise choice. Sanded shoji would be disgusting.

Stan

Haha, indeed and along that same vain I left then unfinished.

[Paul Bent]

Stan, could you expand on the why's of this statement?

"A sharp blade is an absolute requirement, but beyond 6000 to 8000 grit, it doesn't make much difference in finish. A finer sharpening job does help the blade stay sharper longer. Interestingly, a highly polished blade takes more force to cut wood than a medium sharp blade."

[Joe Tilson]

Stan,
It may not be a very good video, but it sure got us some good information about wood reacting to the way it is worked.
Thanks guys, this is great!
Joe

[Stanley Covington]

Lots of very interesting information in this thread.

One more question. Let's talk about gluing those surfaces. My feeling tells me that two planed surfaces glued together would be stronger than two sanded surfaces. When looking at the gluing area it seems to me that it should be the other way around, isn't it?

I recall Fine Woodworking Magazine doing a comparative test of that subject a long time ago, back when it was a worth reading. My recollection is that the planed surfaces developed a stronger glue bond.

I agree with you. Planed surfaces definitely glue up stronger than sanded surfaces. I suspect the glue applied to sanded surfaces bonds to the sanding hair and does not penetrate as deeply into as many strong, undamaged zylem conduits as it does in planed surfaces. The glue bond is only as strong as the wood it bonds to, right.

Planed board edges feel like they glue up tighter than sanded ones, but it is just a feeling.

Stan

[Stewie Simpson]

Normand; you might find the following link of interest.

Stewie;

A note of caution on smooth surfaces: Burnished areas may be smooth, but will not bond. Burnishing causes the cellulose to change chemical characteristics and thus not bond to the polyvinyl alcohol portion of the wood glue. This can be tested by putting a drop of water on the surface of the wood, if it doesn’t soak in, the surface is burnished or sealed and should be sanded until cleaned of the burnishing. http://www.popularwoodworking.com/te...mooth-or-rough

[Patrick Chase]

Stan, could you expand on the why's of this statement?

"A sharp blade is an absolute requirement, but beyond 6000 to 8000 grit, it doesn't make much difference in finish. A finer sharpening job does help the blade stay sharper longer. Interestingly, a highly polished blade takes more force to cut wood than a medium sharp blade."

The first part of the statement is pretty straightforward, and I agree completely: Beyond a certain point you get into diminishing returns, and incremental improvements in polish don't buy you any improvement in performance for real-world woodworking. This is basically the same argument I made in the debate about whether the LV honing compound is actually 0.5 um or not - 0.5 um is 20000-30000 grit (depending on system) and is so far into diminishing returns that it just doesn't matter.

I would add two caveats:

1. I'd specify the threshold in terms of abrasive particle size rather than grit, because there's such variation in grit systems. I think that the point of diminishing returns is usually around 1.5 um, which would be ~7000 grit in Sigma's system and ~10000 grit in Shapton's.
2. The point of diminishing returns depends on what you're doing. Some woods are more sensitive than others, and some applications are more demanding than others. Kesurokai is a notable example of a demanding application, though I personally don't think that counts as real-world woodworking. With that said, I don't think there's ever much woodworking benefit from going below 1 um.

The second part of Stanley's statement is more conditional IMO. The argument that medium sharp blades cut more easily than highly polished ones is usually based on "toothiness" or "micro-serration" (Stanley, is that where you were coming from?). In my experience that's valid in some cases, and not in others. It's validity is greatest if there's a slicing component to the blade's motion, such that the blade is moving across instead of just into each fiber when it cuts.

[Stanley Covington]

The first part of the statement is pretty straightforward, and I agree completely: Beyond a certain point you get into diminishing returns, and incremental improvements in polish don't buy you any improvement in performance for real-world woodworking. This is basically the same argument I made in the debate about whether the LV honing compound is actually 0.5 um or not - 0.5 um is 20000-30000 grit (depending on system) and is so far into diminishing returns that it just doesn't matter.

I would add two caveats:

1. I'd specify the threshold in terms of abrasive particle size rather than grit, because there's such variation in grit systems. I think that the point of diminishing returns is usually around 1.5 um, which would be ~7000 grit in Sigma's system and ~10000 grit in Shapton's.
2. The point of diminishing returns depends on what you're doing. Some woods are more sensitive than others, and some applications are more demanding than others. Kesurokai is a notable example of a demanding application, though I personally don't think that counts as real-world woodworking. With that said, I don't think there's ever much woodworking benefit from going below 1 um.

The second part of Stanley's statement is more conditional IMO. The argument that medium sharp blades cut more easily than highly polished ones is usually based on "toothiness" or "micro-serration" (Stanley, is that where you were coming from?). In my experience that's valid in some cases, and not in others. It's validity is greatest if there's a slicing component to the blade's motion, such that the blade is moving across instead of just into each fiber when it cuts.

Sorry, I missed Paul Bent's question. Thanks for your astute comments, Pat.

The "point of diminishing returns" Pat mentioned is indeed the key factor. I think too many people get carried away with sharpening, myself included. But I don't have any sure way of determining grit size down to the micron, and so can only use the grit size of my stones as a reference. Therefore, I gave a range of 6,000 to 8,000 grit. Your mileage may vary.

When under time pressure on a jobsite, however, there is often not the extra five minutes to polish a blade all the way to 10,000 grit, so one experiments with getting by with rougher grits, and compares performance and durability with finer grits. If the grit size is too rough, say 1,000 grit, the plane cuts poorly, producing a nasty surface finish, and the blade dulls quickly, in my experience. A blade polished all the way to 10,000 will stay sharp a little longer when planing clean wood (cleanliness is a big qualifier) than one polished to 6,000 grit, but it won't cut much better or leave a discernably superior surface finish. If the wood (including the board's ends and sides, for instance) has any dust or grit on it (which it always does on a jobsite), the 10,000 grit edge will dull just as quickly as 6,000 grit edge. So in these jobsite circumstances, spending the extra time and money (stones aren't free and neither is time) doesn't yield a good return on investment. Ergo, "diminished returns." Does that make sense?

The phenomenon of the force required to motivate a plane increasing with a corresponding increase in a blade's polish is not something I pulled out of my fragrant fundament. It has been measured in laboratory conditions. The companies that make the blades for the Superfinisher planes and huge paper shears love these studies because it relates directly to blade edge longevity (how often the blade must be sent out to be sharpened, and resulting downtime), cleanliness of the cut (ragged, striated edges do not make for a nice book or magazine), and electrical consumption, all things their customers pay intense attention to. I believe it is the same phenomenon that causes the coefficient of friction between two surfaces to decrease as the surfaces become more highly polished, but as the level of polish increases up to a certain point, that trend reverses and friction between the two surfaces increases.

I know this is counterintuitive, but it is still absolutely true.

I know about superfinishers not because I own or use them, but because I buy lots of products that have been finished using them for my construction projects here in Japan. I have seen the machines in operation, discussed them with their owners and the manufacturers, inspected the factories, and read the specs. I know about paper shears because my relatives own several huge CNC machines made in Germany, and I have played with them (a 2.8m x 300mm blade is scary, actually), read the specs, and spoken with the sharpening company, and read the sharpening literature. I am confident the same principles applies to equipment in other industries.

This is most evident in handtools in the case of the Ooganna, the really wide planes where the force required to pull them is significantly higher than a regular sunpachi plane. So, for the planes we normally use, this point may be interesting and even detectable, but ultimately unimportant.

The micro-serrations Pat mentions is not a factor I was thinking of. As Pat pointed out quiet correctly, serrations really make a big difference with a skewed or slicing motion. Just like a serrated bread knife. Regardless of the fineness of the grit used to sharpen/polish, metallic blades (vs. obsidian, for example) always develop serrations to some degree or another. But for planes without skewed blades, serrations don't help much IMO. A plane blade acts very differently than a knife or scissors when cutting, and is a tool where sharpness has a much larger impact on performance.

I hope this makes my previous post a little clearer, Paul.

Stan

[Stewie Simpson]

Patrick; only can only relate to my own experience. and that is if you don't work your way up to 12000 grit on the stones before stropping on the Pure Chromium Oxide Paste, you will gain little in additional sharpness to the cutting edge. Little is gained by from using the PCOP if your stopping at 8000 grit on the stones.

As for the Chromium Oxide Wax Combo Bar I had in my workshop, it was last seen in the garbage bin a fortnight ago. R.I.P.

regards Stewie;

[Stewie Simpson]

8000 grit. http://www.sawmillcreek.org/showthre...-of-origin-USA)

12000 (+) grit. http://www.sawmillcreek.org/showthre...nder-the-stone

Stewie;

[Patrick Chase]

Patrick; only can only relate to my own experience. and that is if you don't work your way up to 12000 grit on the stones before stropping on the Pure Chromium Oxide Paste, you will gain little in additional sharpness to the cutting edge. Little is gained by from using the PCOP if your stopping at 8000 grit on the stones.

As for the Chromium Oxide Wax Combo Bar I had in my workshop, it was last seen in the garbage bin a fortnight ago. R.I.P.

regards Stewie;

I've used 0.5 um CrO film (3M 061X, which is pure and very tightly graded), 0.3 um AlOxide film (3M 266X), 0.1 um diamond film (3M 661X) and 0.1 um diamond paste (PSI). They all leave a better looking edge than an 8000# waterstone. The 0.1 um diamond film and paste yield amazing smoothness even under high magnification.

All of that is arguably irrelevant, though, because what matters (or should matter) to a woodworker is how the *wood* looks after you work it with the edge. That's where I think that diminishing returns come into play, and that's the point Stanley made as well in his follow-up post.

As I said the last time this came up, I'm sort of OCD-ish about sharpening and I usually do go down to 0.5 um or so, but I don't see any difference in my results when I stop at 1 um instead.

8000# is 1.2 um in JIS-compliant systems (Sigma, Imanishi) while 12000# is ~0.8 um. You should be able to easily and completely remove the scratches from either using 0.5 um CrO. IIRC you use 8K and 12K natural stones though, so we may be talking about vastly different abrasive sizes and therefore scratch depths here. Yet another case where "grit" is a fairly useless measure.

[Phil Mueller]

Please help me understand the significance of this. I get the smoothness issue. If a build up finish is to be used, wouldn't the final "smoothness" be determined by the finish? I've never read about any finish adhesion issues planed vs sanding. So unless the piece is to be left unfinished, or perhaps just a penetrating oil used, is there any other advantage to a planed surface? Not an argument for or against, just trying to learn something here. Thanks.

[Brian Holcombe]

Phil I find that a nicely planed finish offers advantage. I can maintain panel flatness which is tough to do with an ROS. I can keep clean edges and nice chamfers. I can also work without dust, the finest sanding dust of course being incredibly dangerous to have floating in the air.

[Patrick Chase]

Please help me understand the significance of this. I get the smoothness issue. If a build up finish is to be used, wouldn't the final "smoothness" be determined by the finish? I've never read about any finish adhesion issues planed vs sanding. So unless the piece is to be left unfinished, or perhaps just a penetrating oil used, is there any other advantage to a planed surface? Not an argument for or against, just trying to learn something here. Thanks.

Correct, if you're going to build up a thick film then that will obscure the surface of the wood. In addition to the finishes you list I think that the difference is visible under thin films (as from oil/varnish mixes or shellac, for example).

[Derek Cohen]

In 2006, FWW mag ran an article in which surface were sanded vs scraped vs planed, and then finished with boiled linseed oil, oil/varnish, and shellac.

As expected, the surfaces favoured the planed finish before finish. However, they concluded that differences were difficult to detect after a finish was applied. The oil/varnish mixture was contaminated by sanding between coats, but the other finishes were compared as is. The woods used were close-grained Cherry and open-grained Mahogany.

You may need to be registered to open this link:

http://www.finewoodworking.com/membe...38c7c7e835158d

Regards from Perth

Derek