Elbows

[Patrick Chase]

I'm having a hard time understanding the last sentence. In the case I mentioned earlier when I pointed out the smoothing plane texture in the 1920 Dining room, most people sitting at the table couldn't see it, even after I pointed it out. I thought about this all the time I was eating dinner tonight. I think maybe the people who couldn't see it were all wearing glasses. I'm not sure if it had something to do with wearing glasses, or something to do with eyesight. I don't see how it's physically possible to do any kind of hand planning without leaving some tracks or scallops, even with an iron taking a half thousandth shaving for a full width cambered iron.

You can come really, really close to "no tracks" with a mostly flat iron with a small amount of gradual relief/rouding on each corner (if the corners are square then it's a lost cause). You can always find the boundaries if you know what to look for though - even if the cuts are near-perfectly coplanar there's slightly different reflectivity/sheen where the cut feathers out on each side, that can be seen at high grazing angles, though certain finishes may obscure that.

[Jim Koepke]

Tom, If your eyes can resolve 0.0005", be thankful. I can see tracks from my planes when taking thick shavings. I am also able to take very thin shavings while riding a high track to remove the tracks. It may be my eyesight from old age not seeing what a younger person may see.

Though often I see things others do not. Some folks have never had a job like a printer or a photographer where seeing the details comes with the training.

jtk

[Steve Voigt]

Camber (the constant-radius kind as in the blades I posted in the other thread) is the key here. 15-20 mils is child's play with a r=6" blade, though of course the shavings are fairly narrow (1" wide at 15 mil cut depth, 1-1/8" wide at 20). Also cross/diagonal strokes are easier than lengthwise at those depths.

Steve, what blade profile are you using for that cut depth?

If you're crazy enough to rough with a flat blade or one with "turned-up corners" then yes, 15-20 mils will be rough going. It's doable and I've done it, but not my idea of fun.

EDIT: If you're using a scrub with a lot of camber then 15-20 mils is even easier. The resulting cut widths are 0.7" and 0.8" at 15 and 20 mil projection, respectively.

Hi Patrick,

Your numbers seemed off to me, so I ran them myself…for a 6" radius, I get:

.021" at 1" width of cut
.027" at 1 1/8" width of cut.

Can you doublecheck please? One of us is off; it could certainly be me.

Anyway, to answer your question, I use a 10" radius on jack planes. I never used to measure, but I kinda have to now.
If my calculations are correct, I get .013" at 1" wide, .016" at 1 1/8" wide, and .020" at 1 1/4" wide. that seems right to me. Just eyeballing, I'd say that 1 1/2" wide cut (with a 2" iron) is the absolute max for me in hardwood, and it's usually less. For David's example of hard maple, the 1" width would be plenty. For something like EWP though, I could take almost full width shavings.

[Patrick Chase]

.021" at 1" width of cut
.027" at 1 1/8" width of cut.

Can you doublecheck please? One of us is off; it could certainly be me.

Are you correcting for bed angle? I get almost exactly those numbers if I don't multiply by sin(bed_angle) after calculating the "raw" extension along the long axis of the blade. Depth of cut is the extension projected along the vertical axis (normal to the sole) so you do have to correct.

FWIW (not much without independent confirmation IMO, given the source :-) Brent Beach's online calculator gets the same numbers I did. It gets your numbers with the bed angle set to 90 deg.

[Steve Voigt]

Are you correcting for bed angle? I get almost exactly those numbers if I don't multiply by sin(bed_angle) after calculating the "raw" extension along the long axis of the blade.

Nope, I wasn't. You are absolutely right. Sorry!

So, I am taking a few thou less than I thought, but still in the same ballpark.

[Patrick Chase]

Nope, I wasn't. You are absolutely right. Sorry!

So, I am taking a few thou less than I thought, but still in the same ballpark.

I don't think it matters that much: O(15) mils either way, just a question of whether it's 10-15 or 15-20. EIther way you're going to want some camber on that blade :-)

[Patrick Chase]

Tom, If your eyes can resolve 0.0005", be thankful. I can see tracks from my planes when taking thick shavings. I am also able to take very thin shavings while riding a high track to remove the tracks. It may be my eyesight from old age not seeing what a younger person may see.

Though often I see things others do not. Some folks have never had a job like a printer or a photographer where seeing the details comes with the training.

jtk

There are at least two phenomena that can cause visible tracking: Flatness variations (ridges, scalloping) and surface texture/quality variations. As you point out the flatness variations can be made really, really small by taking thin cuts.

In my experience the texture/quality variations are a much tougher nut. A plane leaves a fundamentally different finish when it's cutting a clean shaving in the center of the blade vs where it's cutting a partial or feathered shaving as the depth tapers to 0. This is really quite easy to see once you know what to look for, though obviously species/grain/sharpness etc all play into it. Ironically the better or glassier the finish at the center, the more obvious the edges are. You can avoid this by using a flat blade, but then you have to contend with step transitions at the edges, and those are obvious at any depth.

You can mitigate the quality variation issue by managing the final sequence of planing passes such that each successive shaving hasn't quite tapered to zero thickness where it meets the previous one. That takes very careful blade preparation (you actually want a fairly small amount of corner relief, or else you can't avoid tapering to zero thickness) and even more careful "workpiece management" to make sure the strokes are coplanar to within the required (small) tolerance and meet as desired. I can't achieve it all the time, though I bet some very experienced folks like Warren, George, and Steve can. Even then you can usually see something if you know what you're looking for.

It's also important to note that the human visual system is incredibly, borderline-unbelievably good at picking out straight lines. I used to work in image quality and image/signal processing, and upon being confronted with an issue one of the first tricks we'd pull out of the bag was often some variation on "well, can we make it wiggly?". I could easily believe that somebody with full visual acuity could pick out a 1/2-mil scallop if it's straight and if the lighting is [un]favorable.

EDIT: Clarified.

[Brian Holcombe]

That is true they almost always show, I tend to stager my cuts so that it minimizes them, then back the plane out for a final pass on a show panel.











You can feel slight scallops on this surface. I can plane finer but it's just not practical for everyday use.

[Stewie Simpson]

To those that see fit to restrict themselves to using tooling that was only available to the craftsmen of the 18th century; a gentle reminder that it was most likely 20th and 21st century technology that was used to fell that tree; that milled that wood from the tree to its rough sawn dimension; that was then able to process that rough sawn feature into a DAR finish after being fed through an industrial sized power jointer and thicknesser; all before it reached your workshop.

Stewie;

[Patrick Chase]

You can feel slight scallops on this surface. I can plane finer but it's just not practical for everyday use.

I probably should have been clearer that I don't view the very subtle sorts of tracking we're discussing here as objectionable. While I think that Tom is right that it's very hard to avoid completely, I think that at the levels we're discussing here they're much less objectionable than, say, the aftermath of an ROS. For that matter I think that scraping to fix such subtle discontinuities is often counterproductive in the final analysis.

Wood is an inherently nonuniform medium. If somebody wants perfect uniformity then they should probably seek some other material, or just buy commercial furniture that's been stained and finished to the point of homogeneity.

[Patrick Chase]

To those that see fit to restrict themselves to using tooling that was only available to the craftsmen of the 18th century; a gentle reminder that it was most likely 20th and 21st century technology that was used to fell that tree; that milled that wood from the tree to its rough sawn dimension; that was then able to process that rough sawn feature into a DAR finish after being fed through an industrial sized power jointer and thicknesser; all before it reached your workshop.

Stewie;

Like you I joint by hand but often thickness with a lunchbox planer (DW735 with helical head). I sometimes do both by hand, though, not out of purism but because I enjoy it even though it's by far the most physically demanding part of neander woodworking.

I think that most others here come from similar perspectives - we're all looking for ways to do what we enjoy (working wood by hand) as much as possible without destroying our bodies. In that sense I think that this is a good thread inasmuch as it may help some folks who want to do more of their work manually achieve that goal.

[Graham Haydon]

I think that's a valid point Stewie. Then as now, I'm sure people would want to work with well sawn timber. If the local pit saw team were poor at their job I'd assume they'd go hungry quick, as would a supplier today if their boards were badly prepared. In both situations it would be causing both loss in time and money to have poor stock.
I might be a bit off here but the UK has been big importer of wood for hundreds of years. Most notably I think we imported, and still do, huge volumes of European Redwood. I'm not sure how early the Dutch had their wind powered mills but they were pretty early giving uniform "deals".
I think my point is the wood in earlier workshops was still well prepared and uniform enough to allow brisk progress. I'm not a purist, I would not like to live "back in the day" but I do enjoy discussing and having a go at working by hand in the most efficient and practical way possible, even if my boards do come off a resaw.

[Kees Heiden]

Cornelis Corneliszoon invented the wind powered sawmill in 1592. He invented a crankshaft to change the rotary action from the wings to the up and down motion of the saw frame. His invention was quickly turned into practice and the 17th century saw many appear, especially in the Zaan, just north of Amsterdam.

This one is still in use in my town Leiden:

[Graham Haydon]

Nice photo Kees! The use of this sawing method is earlier than I had expected.

[Kees Heiden]

So, if you are trying to relive the 18th century, then you can easilly explain away the use of presawn stock.

Personally I draw the line at the puffy shirts...