Feet Planed with Differing Abrasive Sizes

[D_W]

I did some controlled testing to compare a bunch of irons because nobody has been considerate enough to plane to failure in a controlled test, and also because nobody knew how well V11 stacked up against its claims. Some of us (me included) were suspicious and thought that we could get new data and probably knock down the advertising claims of V11 a little at the same time. My specific issue was that I didn't think MDF was a good proxy for what would happen in clean wood that wasn't as abrasive. I still don't know that MDF is a good test bed (it seems like a really lazy way to test things to me), but the results held.

One of the other participants (I tested everything, but a few people provided ideas regarding what to test) wanted to know if he could go to a simplified more coarse sharpening setup, and if he did, what would the compromise be?

Steve Elliot tested this long ago and charted information, but it never hurts to do it a second time. Steve's conclusion was surprising not that finer edge finishes make for longer planing, but how much they do, and he found at the time that edges all the way down to fraction of a micron diamonds did better than coarser abrasives, measurably so.

I didn't go to sub micron, but I did test a 1200 grit ezelap plate (which is finer than the number would suggest), a 5 micron diamond powder (which is quite coarse for an edge), 2.5, and 1 micron, as well as a dan's black finish stone (the best arkansas stone I know of, and most consistent from one to the next) as well as my favorite washita.

These tests were controlled using a guide for angle, the same test piece of wood, and wire edges were stropped off with confirmation of a clean edge under a metallurgical microscope.

The wood used is beech heart, the iron used is O1 (this is after testing the other alloys, and I had no interest in potentially planing 2000 feet to dull each time with V11 or M4). Shavings until failure are approximately 2 1/2 thousandth thick, something a typical user might work with on a smoothing plane until the last pass. Here's the footage planed:

* 1200 ezelap on the bevel, back bevel with 1 micron diamond powder on cast iron - 527 feet (dull surface from the start and high planing resistance)
* 1 micron diamond on both sides of the bevel - 840 feet
* 5 micron diamond on both sides of the bevel - 568 feet (dull feeling and resistant, like the ezelap and back bevel combination)
* dan's black stone front and back - 700 feet

Those results were done with a 35 degree final bevel, and a request was made to run some at 30 degrees. I see quality issues with small chips on some irons below about 32/33, so I generally freehand around that. 35 was chosen for other testing because it's easy on a hard-to-hone iron to prescribe three bevels to ensure that the last stone actually works the very edge (grind, 32, 35 are the steps). I learned that from a charlesworth video when I started, and figured anyone who didn't like the results I posted could very easily run the test on their own

At 30 degrees final bevel, the additional longevity in a hand plane showed a little bit as follows:3

* 1 micron diamond on both sides of the bevel - 908 feet
* dan's black stone front and back - 738 feet

I threw out the 2 1/2 micron results, as i recall, they were unexpectedly short and I didn't feel like redoing them. But I can't remember for sure what the reason was other than something didn't work out right. The sharpness was also unimpressive.

It was a pain to do all of these with a microscope and a guide (to set everything identically and then confirm everything visually under a scope. about half of the irons had some kind of edge issue after sharpening with a guide that one would never see with the naked eye. I don't think they would affect much in longevity, but part of the testing was to judge surface quality, and it's not quite fair to put an iron with a defect on wood and then comment about the lines that it leaves behind).

The perfectly white reflective smooth pictures of one micron diamond are shown on other posts. There's some sentiment that natural stones create a longer lasting edge, but I'd suspect that they generally do not as there are no natural stones that leave an edge as fine. It's easy to get a good edge on an oilstone. It's quite hard to get one visually perfect with no defects after stropping.

Here is a finished black arkansas stone edge (you can see that the stropping moves the edge around a little bit):

5 micron diamond (it actually looks great, but the edge itself is rounded over after stropping on bare leather - the toothy bits left by 5 micron diamonds just round off on a bare leather strop)

While i also thought that this was worth looking at, I have to admit that the cycle time freehand for finishing with a finer abrasive is probably 20 seconds longer at the most (1:20 vs 1 minute), so there's no great reason once you master freehand to use coarse abrasives to finish. Doing all of these tests at the same time gives you a much better feel for the difference in cut resistance and surface quality - it's stark. 1 micron diamonds also do a nice job of prepping an edge for the strop - there just isn't much there to have to strop off - noticeably less even than a black arkansas of the finest quality).

A fine washita makes for results very similar to those of the black arkansas stone (washitas work fine with V11, but not M4 and 3V - I'm beginning to think that vanadium carbides burnish novaculite too much. V11 slows the washita down a little bit, but the edge is actually finer off of a washita with V11 than it is with O1 - not a horrible side effect for finishing).

 

[D_W]

One of the lovely things about O1 and ward's original water hardening steel and other things of the like is that it will wear evenly even if the edge doesn't start finely. When you get into steels that have big carbides in them, that's not necessarily the case (also, the lower toughness conventional high speed steels also dont "heal" small defects with wear as well).

The newer powder HSS steels do tend to forgive some minor defects and wear evenly, too, but the edge has a different look (actually, the edge under the microscope tends to be very different even for the same steels depending on whether the cut is end grain or long grain, and by the type of wood. Hard maple will cause some of the powder steels to look visually similar to O1, smooth and not gritty looking, but on beech, the powder steels all look more gritty).

 

[ED65]

D_W, thank you so much for slogging through the testing and for posting some of the results here. This is going to prove invaluable, not least for the various take-homes for different people who use a variety of tooling.

I am having a hard time trying to reconcile part of what your test shows against some previous information (e.g. in the extensive amount of knife test-cutting data from the likes of Cliff Stamp) but it'll take me a bit of pondering to generate a few meaningful questions.

One quick thing though, this jumped out at me from the first read-through:

1200 ezelap on the bevel, back bevel with 1 micron diamond powder on cast iron - 527 feet (dull surface from the start and high planing resistance)

Do you have an explanation for this? Is this more about the back bevel than the coarseness of the abrasive?

 

[D_W]

In order to make the results repeatable and a photo perfect edge attainable from end to end regardless of test damage, I put a 0.5mm rule back bevel on everything.

The reason for the additional planing resistance was just dullness due to the abrasive coarseness. it was a little surprising how big of a difference it made in footage planed, but a sharp plane is so much better at staying in the cut.

It's really not apparent just how big the difference is in resistance, surface brightness, etc, unless you do all of these things in a row. I couldn't tell you what point the 1 micron edge felt as dull as the 5 micron edge at start, but I'd venture to guess that at 0 feet, the planing resistance from the 1200 grit combo edge was much greater than the 1 micron edge (both sides of the bevel) at 500 feet. After just having the feel of the fine diamond edge staying in the cut nicely, it was really disappointing to get the feel of the 1200 grit or 5 micron edges.

As a huge fan of the washita, I'm an advocate of anything that's one abrasive and leather strop right off of the grinder - especially for someone learning to freehand - I was hoping that the 5 micron diamond would be a decent substitute for that kind of thing for people who use steel that's not tolerant of washitas, but it just wasn't close.

1 micron back bevel was used with the 1200 plate for a very simple reason - the wire edge from the 1200 plate is extremely difficult to remove with bare leather. a 1 micron back bevel makes it come off a lot easier. I think a 1200 plate is something like 20 microns. 5 microns is deceptive in diamond grit - when it's embedded on cast or brass, it cuts very deeply.

 

[D_W]

I have a fairly fine washita stone that also planed 903 feet, but it did so when the stick of wood that I was using was transitioning from heart to sap (american beech), so I took it out of the results. I don't know what it would've planed if I'd have had all heart like the other tests. Probably something closer to the 700 that the black ark stone made.

That fine washita stone is still what I use day to day, though. I haven't taken any of these stones yet and charged them with 1 micron diamond abrasive to see what longevity would be, but 1 micron diamond is cheap, and stones have a nicer feel than cast when you're freehanding. If any contaminant at all gets on cast iron (even loose floating wire edges), it can get embedded or even just loose, it can notch the edge edge in little tiny notches that will show up on a planed surface.

 

[D_W]

Translation of the above:

* fine abrasives yield much in edge life in reasonable shaving thickness (at the low end, the "just use a king 800/1200 and get on with it" crowd, edge life loss vs. a much finer abrasive can be 35-40%)
* effort in use and quality of result also falls well in support of fine edge finishes
* the fine edge finish used in these tests is not a full bevel, even at a tiny kiss to the edge, the effect is obtained
* there is no support of the theory that natural stones make a longer lasting edge than fine synthetic abrasives (boo hoo, i am a fan of the natural stones)
* fine edge does not mean significant additional sharpening time vs. the simplest of regimes
* fine edge finishes can eliminate or significantly decrease the need for stropping (technique is the issue here - light alternating front and back is plenty to remove wire edges of any significance)

* if you have reasonable clearance (10-15 degrees) on a common pitch plane, you may gain some longevity by creeping toward higher clearance, but less is to be gained here than is gained by decreasing abrasive size. you will need to explore the point where edge issues negate additional clearance.

 

[ED65]

D_W, rather than possibly muddy the waters unnecessarily I've decided not to draw on anything related to cutting performance in knives. It might not be directly comparable for a few reasons so best not to try to extrapolate from that. The first reasons I can think of are A) it's not wood that's being cut and I don't know how much (if any) difference the material makes B) no cap-iron effect, and last but not least C) I was just reminded that there is a veritable ocean of information in this area to wade through, far too much of it contradictory!

In order to make the results repeatable and a photo perfect edge attainable from end to end regardless of test damage, I put a 0.5mm rule back bevel on everything.

Okay, makes sense.

The reason for the additional planing resistance was just dullness due to the abrasive coarseness.

With a back bevel on everything that's what I expected you'd say, but I am surprised you felt the difference was so stark given the wide variety of honing routines out there. Everyone honing coarser can't be experiencing that much greater resistance surely? Maybe they are. But still surprised given your first example was 1,200 grit, not coarse at all by old standards. Yes I know diamonds don't work the same way as other abrasives do even if the numbers match up, but it's still not coarse coarse.

My impression as of two years ago (closer to when I had compared this a lot) was that regardless if I was using my preferred diamond plate or another stone, stropping or not, I could generally detect no difference in planing resistance. So I thought I was stropping mostly for OCD reasons on plane irons (and also I figure it takes so little extra time who can be that lazy?) but your results suggest strongly that I should never skip this step. I'll have to do some more testing, making sure to take out as many variables as possible, paying closer attention to the feel to see if I think that still holds true.

Now the clarity of the surface. This is the total head-scratcher for me because there's little chance of a major confirmation bias going on. I'm wondering if the wood is more at fault here.

I can achieve a bright/clear surface that feels glassy from an iron straight from my worn-in 1,000-grit diamond plate. But also from an old, dense SiC stone (Carborundum?) that I've estimated in the past is approximately equivalent to P600, and ditto from the fine side of an India/Crystolon combo stone which as you probably know is rated significantly coarser. I'm sure I've looked at surfaces directly from various other stones too, including an old Washita from a Buck honing kit (bought 30+ years ago to date the source of the Washita if that's significant).

 

[D_W]

I'm not sure the buck stones were washitas. I think they're soft arkansas stones.

As far as stropping comparing different abrasives, are you stropping with compound or just bare leather?

I never noticed this before, but what probably made it stark is that I was shooting for the same footage and the same shaving weight at a time, and it made comparing resistance then very easy. I probably would just work a slightly thinner shaving with an iron that started more dull to equalize it, but the reality is that's getting less work done.

As far as surface clarity - it's ultra difficult to get a picture of a reflective surface because my phone sees such a thing as a defect and the software eliminates it by focusing on the area where the glare came from. There is some relative issue here as far as brightness goes - 1 micron diamonds with no defects will leave a surface that looks like it's been waxed when viewing in raking light. A very clear defined image of the background will show up in the wood.

With 5 micron diamonds and more coarse, there is no such thing - there's enough surface dullness to prevent clear reflection. Someone who is going to scrape and sand won't care.

There's a curious side part to this, though. AS the 1 micron diamond wears, it will travel from being uniformly bright to getting more and more dull. When I'm planing, I don't care as long as defects don't appear as the wood will get some kind of finish. But I was disappointed to see that even with stropping, the edge off of a 1200 grit stone created a surface brightness similar to the 1 micron irons when they were about 95% expired- it may have actually been worse.

I didn't test stropping coarse edges vs. stone work as I ran out of initiative for ancillary experiments and was running out of good quartered wood.

Beech is favorable for the surface brightness check, though. It's dense and doesn't have much for organized pores. Hard maple is also good, but hard maple has so many mineral defects and other issues that I couldn't dull irons before they just got damaged instead.

Like this:

Here's the culprit. Sometimes, the mineral streaks or inclusions are harmless. Sometimes they have bits of what looks like silica in them. no iron survived repeated passes over this spot:

 

[thetyreman]

interesting DW, I had a few 'spots' like that when making my beech chopping board, and thought I was imagining the blade going blunt ridiculously fast, makes sense now.

 

[ED65]

Right, just finished a small and unscientific comparison using three no. 4s with their irons honed respectively on the SiC stone (~600 grit), my 1,000-grit diamond plate and the last done my normal way which involves finishing on a loaded strop back and front. No stropping of any kind on first two. Also I made sure to trash their existing edges so no trace of the original honing remained and the apex had to be formed completely from scratch.

Resistance first, no difference that I could detect but I wasn't planing a large surface or for very long. I'm sure the longer you plane and the harder the wood the more noticeable this is likely to be.

There IS a difference in quality of the planed surfaces. Significant but though.

The surface left by the diamond-honed iron and the one produced from a stropped edge are similar enough that you can only really notice it by direct comparison despite the gigantic gap in the grit size (13-15 micron and <0.5 micron respectively).

And nobody could legitimately call the surface left by the coarsest-honed iron dull; it's definitely not as bright but still plenty of glint from the cells in the wood and it feels as smooth as glass just like from the other irons. Most people would be perfectly happy with it – it's streets clearer and brighter than you get from routine sanding for example. This shouldn't really be a surprise since the surface should be at least somewhat like it was sanded to 25 microns, and sanding is usually taken to 50-60 microns.

So again I'm wondering how much effect the wood has on this. I can leave all three planes with their irons set up this way so I can continue testing across a range of species over the next while.

 

[D_W]

Did you keep the shaving thickness even across planes? I would've compensated for the coarser edges by backing off. Because I was weighing shavings to keep the test fair, I couldn't do that - they had to remain consistent thickness.

As far as comparing to sanding, yes - you have to and and burnish to match any planed finish or even get close. if you're going to scrape and sand, you're not going to care about the level of surface brightness, but it's a good indicator of edge quality, and I was observing it less for relative brightness (that's more of a surprise that it was as different as it was from one to the next). what I was looking for was the emergence of defects while planing to see if any iron type was better at avoiding defects, as well as whether sharpness level prevented defects.

The surface off of V11 is so bright that you actually have a counter issue to deal with - it is reflective like a surface that's been finished. The slightest irregularity in grain direction that doesn't plane as bright instantly shows up. It's easier to get a uniform surface with slightly less brightness.

I found that all of the irons held up well in clean wood. Carbon steel, from what I could see, was the best at developing the least defect lines in it, but carbon and powder would "heal" if there were small lines that weren't big enough to feel. Plane some more wear to the edges and as the surface dulled a little bit, the tiny defects would be worn off the edge.

A2 worked in the opposite direction - no defects for quite a while, but then after about 2/3rds of its footage, it started to shed the edge for lack of a better way to put it. I won't bore with pictures, but the progression actually shows the edge going from what looks like burnished to flakey looking and granlular.

The defects on the tsunesaburo iron made the surface a little duller from the start, and then it kind of stayed like that. Another big surprise - I expected that iron to yield the brightest surface due to the steel being pretty plain (blue steel is a lot like O1, but a little cleaner and a little more carbon), but it gave up little balls of something right away. Not big enough to leave lines, but enough to dull the surface.

All that said - planing resistance - can your three irons work in the same plane? I was very careful with this test and some of these things, like resistance and ability of the plane to stay in the cut, I didn't really expect to notice. The lack of variation in plane/cap iron/etc and weighing the shavings is probably what made them noticeable. They're not key parts of my observation, though - footage planed and condition of the edge as it wears were the two factors that I was looking to test.

As far as the plane staying in the cut (these tests all had the cap iron set at about a hundredth or so), at the very outset of the test, you could've planed every shaving with one hand. It's hard to notice that the plane is losing clearance relatively linearly (but it is) - just you notice at some point that the resistance from shape and angle on the bottom of the iron makes the plane stay in the cut a little less well, and then a little more, and then at some point, you notice that you have to actively influence it.

The less you sharpen a plane, the less the clearance has to be limited before that starts. Small things such as that are little nits in day to day work that aren't that easy to notice, but they all lean toward sharpening to a finer degree (and figuring out how to do it without adding time) and sharpening more often.

The super bright surface will be scraped and sanded off by most, but it'll never hurt anyone. The plane staying in the cut on its own longer can be the difference between chatter at the beginning an ends of cut or not - not short chatter from the iron oscillating, but the plane bounce that some people get when they start at the edge of a board.

 

[G S Haydon]

Thanks for the test. Very interesting reading and good to see you had to be open that good, fine, synthetic stones are as good as the natural stones you enjoy using.

Also, it was interesting for me to read that you are essentially getting the best results using the same system as David Charlesworth does, although freehand. Fair point?

Lastly, I have just finished a desk with a spalted beech top. I wanted absolute clarity to best show the grain. My normal edge, india stone, left a smooth surface surface, much smoother than abrasive. However when viewed with raking light the finish was not good enough for me. I went finer with a boot sale charnley and got a finish I was satisfied with.

 

[D_W]

Yes on the charlesworth-ish method. focus on the very tip of the iron and don't get obsessed with the entire bevel. Just no ruler trick. While doing this testing, I actually viewed the condition of the back as it's easier to get a picture of with a microscope. If the back is clear and the edge is crisp, then sharpening is probably complete. In order to remove the full amount of wear from the back bevel, you get stuck chasing the back bevel larger. This is, of course, not necessary - if the wear is uniform at the edge, you can let it go a little. Since I was testing irons, I didn't have that luxury, and when the test was done, I was glad to hone off the growing back bevel.

I was also testing with the cap iron set, which can generate more back wear when there's no damage, but the length of the wear with a 1 hundredth set appears to be about 2-3 thousandths long.

re: sharpness for a smooth surface, if the plane is cutting at all and the cap iron is set, the surface should be good. The advice to plane a clear surface is almost always to go with more sharpness, but that's only useful on the very last passes. I find the same as you - it's the surface brightness that suffers from lack of sharpness, though someone who reads that may interpret that as being an issue of tearout (it's not - it's very uncommon for wood that tears out with a cap iron to set to plane well at all with anything, and it's also uncommon to have wood where any of that is an issue at all).

Steve Elliott posted long ago about the durability of an iron with various honing media. I think all of his work is accurate, and he's got the disposition (no ego) to do testing as he's not invested in trying to make the results match prior statements. I figured I'd find what he found, but it was worth a shot. I knew Warren would object to what I found, but when I get off my duff and finally publish all of the testing that I did, it'll be very easy for anyone else who is remotely competent to do any of it. They'll find what I find if they're honest. At any rate, Steve found that durability of the smoother's edge continued to increase all the way to 0.25 micron diamonds (as in, a measurably better result than even 1 micron diamonds or half micron). I stopped at 1 as it wasn't the point of the test, and I don't really know what point I'd need to add an extra step. With 1 micron, there's no extra step, and there's measurably better durability than any natural stone. Based on steve's testing, I think something like a norton 8000 would fare less well than my oilstone and probably as bad or worse than my washita, so chasing edge life with synthetics does require dealing with really expensive synthetics (gokumyo, shapton 30k, etc) or changing to oxides (slow) or very fine diamonds.

Most folks who work a full bevel and insist that it results in greater edge life probably wouldn't like to see their work under my microscope, and they also subsequently wouldn't like the results of actual practical repeatable testing.

 

[ED65]

Sorry D_W, forgot there were pending questions here.

Did you keep the shaving thickness even across planes?

As much as I could, yes. Obviously I had to go by look and feel, and did check transparency as I dialled in the shavings on the second and third planes to match.

The surface off of V11 is so bright that you actually have a counter issue to deal with - it is reflective like a surface that's been finished. The slightest irregularity in grain direction that doesn't plane as bright instantly shows up. It's easier to get a uniform surface with slightly less brightness.

Slightly off-topic but I don't see this as a problem since finish will be going on. Finish can equalise gloss across wildly divergent surfaces – burying the difference between planed and normally sanded surfaces for example – so this slight difference might vanish with just the first coat (possibly even with an oil finish, because of the burnishing... assuming the finishing process is as per tradition).

All that said - planing resistance - can your three irons work in the same plane?

Sadly no. I wanted to do this from the start to take out the impact of the plane bodies, not least because the soles on the two non-vintage ones were ground to vastly different finishes in the factories (one looking like it was only taken to 80 grit and the other to a finish comparable to that on a premium plane).

But the cap irons are proportioned differently enough that I couldn't fit all three iron sets to one plane and get the necessary range of adjustment. And I can't swap the one cap to both of the other two irons without some light showing at the leading edge in at least one case.

The super bright surface will be scraped and sanded off by most, but it'll never hurt anyone.

Aye. And I don't want to suggest that users should hone less finely when there's the option to go further (and with so little additional time or effort). But it is nice to know you can and get good results, for those who don't have that option [yet ] or when you're in a situation where you have to make do (so you don't feel like you're compromising on surface quality).

The plane staying in the cut on its own longer can be the difference between chatter at the beginning an ends of cut or not - not short chatter from the iron oscillating, but the plane bounce that some people get when they start at the edge of a board.

Good to know. This alone may be enough reason for some to go the extra 'mile' (half inch )

 

[D_W]

Yes on the heavy finish - once the finish is level, only horrid stuff still shows up. Tearout free generally looks good. Once in a while, I like to try to burnish and wax, which is the only case where surface issues in terms of dullness show up. Even a burnished surface needs to have extremely little tearout, but for most, as you say, it's academic. Sealer, knock, back grain raising and then top coats negates it all.