Precision with hand tools

[Brian Eaton]

I’m working on a split top saw bench and have finally gotten all of the lumber dimensioned. Unfortunately, all of the various pieces do not match exactly in dimension. I started by cutting to rough length, flattening/straightening I face and the straightening and squaring an edge. Once I got those steps finished I use a marking gauge to scribe a line for thicknessing the board and then saw about a 1/16th away from the board and then plane to the lines. Before finishing the opposite face I joint the second edge, referencing from the original face.

invariably, once I’ve planed to the lines and checked for flatness, I’ll check with a square (off of the reference edge) and the face opposite the reference face isn’t square. In adjusting with a plane to get it square I end up going past the line a smidge and then the boards do t match exactly in dimension.

For my saw bench project everything seemed “close enough” that I’m going ahead with it but I’d really like to be able to get multiple boards dimensioned exactly the same.

Any ideas, tips, etc. for how I might improve my accuracy? I have a set of Veritas winding sticks that I use as a straight edge for shorter pieces, and I have a longer level that, as far as I can tell is straight enough for longer pieces. Also, I know my two combination squares (6” and 12”) are accurate as I’ve checked them against a know straight/flat edge/surface. Lastly, as far as I can tell I am making accurate and consistent marks with my Veritas marking gauge (pressing the face of the gauge into the surface and pulling towards me to make the piece).

Am am I just being in a hurry to learn this skill? I’m a bit type a and one of the reasons I got into more precision woodworking was because I got so frustrated with how inconsistent/out of square box store lumber was when I was making more basic projects.

[Bill Houghton]

Mostly, it's the same thing as how you get on stage at Carnegie Hall: practice, practice, practice.

For getting a square edge, you might try scribing both faces, and watching as you plane for the "feather" that will appear as you plane to the scribe lines.

[Brian Eaton]

Thanks, Bill. I know it is likely just a lot of practice - just trying to learn good habits and not have to unlearn so many bad ones. I should have been a bit more specific about marking for thickness. I do mark both edges and the end grain and do look for the feather edge as I’m planing. I suppose a better way to say it is that I get to that feather edge on both sides, check for flatness, and then when I go to check if the second face is square to the reference edge it tends to be out of square though I know I’ve planed to the lines on both sides.

I shall all keep coming up with projects to practice with!

[Brian Holcombe]

If you are making the reference face slightly concave then the resulting check for squareness will read inaccurately unless you use a square which is long enough to touch across the entire face.

[James Pallas]

Brain, I don't think you mentioned how far off you are. Is it an eighth a sixteenth or just light under your square. I believe sometimes we just have to work towards the finished work. On your saw bench are you expecting the splits to be exactly flat after you do your joinery and put it together? IMO you would be very good or very lucky for that to happen. When you have to plane the top to get it flat is the bottom side going to be exactly parallel to the to the top? Same thing when you have to plane the top edge of a drawer or level the dovetails on a drawer side. Use your best judgement.
Jim

[Patrick Chase]

I’m working on a split top saw bench and have finally gotten all of the lumber dimensioned. Unfortunately, all of the various pieces do not match exactly in dimension. I started by cutting to rough length, flattening/straightening I face and the straightening and squaring an edge. Once I got those steps finished I use a marking gauge to scribe a line for thicknessing the board and then saw about a 1/16th away from the board and then plane to the lines. Before finishing the opposite face I joint the second edge, referencing from the original face.

You saw 1/16" away from your line and then plane to it? That's a *really* effort-intensive way to dimension lumber. I'd suggest taking a small percentage of the time you're spending on that, and spend it practicing accurate ripping. You should be able to get within 1/32" or so without too much practice (and ultimately much better than that), and will end up spending less time total that way. One thing that helped me was to practice resawing thick offcuts into ~1/4" boards. You can of course slice thinner than that, but if you do then the "give" of the thin offcut may enable you to cheat and twist the saw back on-line in a way that you can't when making deeper cuts, so it's not good for practicing IMO.



invariably, once I’ve planed to the lines and checked for flatness, I’ll check with a square (off of the reference edge) and the face opposite the reference face isn’t square. In adjusting with a plane to get it square I end up going past the line a smidge and then the boards do t match exactly in dimension.

As above it sounds to me as though the "original sin" is in your sawing technique, and everything else stems from that. Practice standing dead square to the work, and focus on maintaining a very light grip on the saw handle. You can't really twist it if you're not grabbing the saw to begin with, after all.

I would also suggest fixing "gross" flatness issues before thicknessing.

This is probably obvious, but make sure that you mark off your lines for thickness starting from a flat reference face. That's why we always joint one face before thicknessing the other.

[Brian Eaton]

James, I usually end up between 1/32 and 1/16 off between two pieces in thickness, prinarly at one end (varies with the piece which end) - I.e. one board ends up a bit on the thin side at one end. The same issues occurs somewhat less frequently with width. Per calipers I can sometimes be as much as .01-0.02 thinner at one of a board as compared to its other end.

[Warren Mickley]

There are a few technical considerations that have not been addressed, so I will add a few notes.

When you plane your first face, you have only one concern: making it flat. And assuming you have sufficient thickness, you keep at it until satisfied. With you reference edge, you have two considerations: it is to be both flat and square to the face. If it is flat and not square or square but humped you can make adjustments, and keep trying over again as long as you don't run out of width.

With your last two sides everything has to come together at once. You do not have second chances. For the last edge, you want to be flat in width, flat in length, not in wind, square to the face, and also right on your gauge line. Once you hit the gauge line, however, anything you do is going to make problems somewhere. At that point you have to decide what is most important and either stop or keep going. It sounds to me that you maybe would do better hitting the gauge line and calling it good enough. You want to be checking for square as you near the line, not waiting until after you hit the line. There are several reasons we make measurement from just the face side and face edge, but one is that you have had the luxury to work these initial surfaces until the are good and flat and reliable.

[Brian Eaton]

Warren, those are good points. I think I’ve pretty much got the hang of a flat face with no wind and a square/flat edge. I’ve got two more pieces to dimension for my saw bench so I will check for square more frequently as I plane down to the gauge lines on the back side and call it good as long as it is pretty close. I was watch a video last night and it was a good reminder that the reference face and edge are what are important (the guy was making a small cupboard) and was using the reference faces to make sure the inside of the cupboard was square) and said several times that he wasn’t terribly worried about the outside being perfect because he was more concerned with the inside being correct since that was what the shelf’s would slot into. At the end of the video the cupboard looked pretty good to me!

[Jim Koepke]

Per calipers I can sometimes be as much as .01-0.02 thinner at one of a board as compared to its other end.

One of my earlier woodworking epiphanies was that trying to use machinist precision in measuring caused more problems than it was worth. Fitting pieces to each other worked better than trying to fit them to marks on a ruler.

Most of the time now my only need of a ruler is to find the sizes a finished piece has to fit into. More often this can be carried to the work with a story stick.

For length my pieces are compared and then adjusted as needed on a shooting board.

If two pieces need to be the same width they are held together and checked in all ways possible and then if need be adjusted, often together. If two pieces need to be the same width and one is a touch larger, clamping them together so the proud one can be brought down with a plane until a bit of shaving appears from both is one way to make sure they are the same.

jtk

[Brian Holcombe]

You can literally do more accurate work by hand than machine, I can flatten large panels to a very fine degree of flatness and squareness by hand. However beyond a certain degree of flatness one finds that they need to find tune the result to the project. I like my references to be very accurate and parallel surfaces to be to the gauge line. If a precision fit is required, such as to fit a panel into a groove, often I will make the panel thick and cut a rabbet into the hidden surface.

Its important to work with hand tool approach rather than machine tool approach, accuracy when needed rather accuracy for the sake of accuracy. Machines work off multiple reference faces and so parallelism is far more important than it is in hand tool work where we work off of one reference face or corner.

[Brian Eaton]

“Its important to work with hand tool approach rather than machine tool approach, accuracy when needed rather accuracy for the sake of accuracy” <- this is a good reminder. I can get lost in the quest for precision and forget why I’m doing what I’m doing if I’m not careful!

[Steven Mikes]

I just started building a split top saw bench too! Going through the same process, but I'm not worried too much about getting the parts all exactly identical. As long as mating pieces all fit together well it's good enough for me.

[Darrell LaRue]

I’d really like to be able to get multiple boards dimensioned exactly the same.


I dunno how to help you with this. When I work to specific dimensions, the measure of "exact" is that the reference faces are all square and the ends are shot to the same length. Thickness is not usually an issue, as differences are thrown to the off-sides, and can be planed flush if necessary, or just let them be, depending on what it is. I have a small table that has raw riven surfaces on the inside of the aprons. But the joinery is fine because it is all laid out and cut from the reference (outer) faces. The underside of the top has some unplaned bits, where I planed the top just enough to get it to sit flat on the frame and you don't see the ugly parts.

Darrell
working on a pole-lathe turned chair that will be every bit as characterful as the 13th C original

[Brian Eaton]

Your advice is similar to a few of the other responses that people have posted and it is starting to sink in. I’m nearly done with the dovetails for this bench and so far it seems like it is going to go together well. I’m excited to take some of this new perspective to a more “fine” project - my wife wants a small wall cupboard or shelf she can put her essential oils on and it is on my honey do list. Thanks!

[Jim Koepke]

[edit]
I’m excited to take some of this new perspective to a more “fine” project - my wife wants a small wall cupboard or shelf she can put her essential oils on and it is on my honey do list. Thanks!

Here is a cabinet built for my wife to hold her many boxes of tea:

https://sawmillcreek.org/showthread.php?224747-Tea-Cabinet-Latest-Project&highlight=

Maybe a change of dimension on the shelf spacing to better fit the essential oil containers would be in order.

The case is dovetailed which are then covered by molding.

jtk