Edge planing a board straight

[Mike Henderson]

How do you edge plane a board straight (with a hand plane)? I've never figured out how to do it except to take a known straight edge, such as some MDF cut on my table saw and lay it on the edge to check if it's straight. Then, I can focus on the high points and get the board straight to my satisfaction. I'm talking about a long board and not something 2' long. Maybe a 6' board.

So maybe I should ask the question in a different way, "How do you know an edge is straight when hand planing it?" I've tried looking down the board, and maybe my eye is not good enough, but it can look straight and still be bowed when checked against a known straight edge. Any alternative to checking it with a known straight edge?

Mike

[Sean Hughto]

Are you using a jointer - like a 7 or an 8?

I would think keeping an edge square to the faces would be a bigger challenge for most hobbiests than straight?

But to your main question, if the board needs to be straight to edge joint for example, I think testing the edges or using the bench top or what have you to check your result for "straight" is SOP.

[Preston Baxter]

Here is a method that I've adopted that almost always insures a flat edge without having to check with a straightedge. It works for me, but a lot may depend each individuals planing technique.

If the plane is long enough for the job, then handplaning will almost always result in a flat or rounded surface. Once I get to the point where I am taking a complete shaving from end to end; I pull back the iron for a very thin shaving, then try to plane the edge hollow. In other words, I start 1-2" in past the edge and plane to 1-2" in from the other end. When the plane stops cutting, the board is now very slightly hollow. I now take one or two full length shavings to bring the edge to flat, or near flat. You can judge by the shaving to how flat the board is getting.

At the moment the plane is taking a full length shaving again, the board is flat, any more planing from end to end risks rounding again. I prefer edge jointed board to have just the slightest hollow. This reduces the possibility of the ends of the glued up panels from splitting later.

[David Gendron]

+1 on Preston's description, that's the way I do it and it work realy well. One other way I use some time, if I know that one edge is "flat" is to scribe a line from it and plane to the line. If your scibed line is deep anought, you will notice when you get close to your goal, by having a small "feather" of wood at the edges of the board that will indicat to go easy and mabe to take a finer shaving! hoppe it make sense!

[Mike Henderson]

Sean addressed this, but my real question is, "How do you know if the edge is straight?" By this, I mean straight from end to end. For the purposes of this question, I'm not concerned with whether the edge is square to the faces of the board - only if it's straight from end to end.

Is there any way to know if the board is straight except to test against a known straight edge (or as Sean suggested, the top of your bench).

Mike

[Of course, if you check against the top of your bench, how do you know if the top of your bench is straight?]

[Tom Winship]

Preston, in para. 2, when you say you start in from the end and stop short of the other end, I'm assuming you are saying to start and stop the plane iron at these positions.
I also seem to have a problem with squareness of the edge. Any suggestions on that?

[Michael Faurot]

I also seem to have a problem with squareness of the edge. Any suggestions on that?


A No. 95, used in combination with your jointer, does the trick to help insure the edge is square.

139248

The one pictured is from Lie-Nielsen. But vintage Stanley or new Veritas would work also.

[Preston Baxter]

Sean addressed this, but my real question is, "How do you know if the edge is straight?" By this, I mean straight from end to end.

With the method I use, the length and inherent straightness of the jointer plane determines the board straightness. If the edge is planed hollow with stop shavings, then a final, full length cuts are made until a full length shaving appears, then the edge will be straight due to the flatness of the plane. That is of course assuming the sole of the plane is not badly concaved.

I think I picked this up from watching the Jim Kingshot and David Charlesworth bench plane videos.



Preston, in para. 2, when you say you start in from the end and stop short of the other end, I'm assuming you are saying to start and stop the plane iron at these positions.
Yes, I am referring to the plane Iron at these positions.



I also seem to have a problem with squareness of the edge. Any suggestions on that?
I keep a small metal square handy that I can quickly pick up and check at least three points along the board, for edge to face squareness. Having a slightly cambered iron on the Jointer plane really helps in squaring edges and keeping them square. With the camber, if you keep the board's edge centered under the plane, it will cut equally. It you let it ride off to one side, it will cut more on that side. So you check it with a square, mark or make a mental note of the high points, and steer the plane accordingly to take out the error. The Charlesworth planing video goes into good detail about this technique.

[Jim Paulson]

Mike,

To compliment what has already been said, I usually rely on my 4 foot aluminum level to check for high and low spots. It is handy and I've had go experiences using it. Whenever I'm edge planing much over 2 feet in length I'm reaching for my number 8 Bailey once the highs and lows are smoothed.

If the length is sufficiently long one can also check straightness with a chalk line held from end to end and snapped on the side of the board.

But now with that said, I have less interest in using a reference as well. So I am building a couple shooting boards so that my board edges will come out at right angles and straight. My low end power joiner usually causes me to touch up the edges anyway so I'm moving in the direction of relying more on hand planes. ;) I want to plane edges on drawer fronts and case sides and I'm making a four foot long shooting board. Charles Hayward in Cabinet Making for Beginners even talks about using a five foot long shooting board.

I think a lot of people think of shooting boards for trimming ends, but they seem to offer additional benefits with edge planing. Food for thought.:)

Jim

[Pedro Reyes]

Sean addressed this, but my real question is, "How do you know if the edge is straight?" By this, I mean straight from end to end. For the purposes of this question, I'm not concerned with whether the edge is square to the faces of the board - only if it's straight from end to end.

Is there any way to know if the board is straight except to test against a known straight edge (or as Sean suggested, the top of your bench).

Mike

[Of course, if you check against the top of your bench, how do you know if the top of your bench is straight?]

Sorry to be like this and answer with a question, but is straight better than having a slightly convex edge? I do what others have mentioned and plane a slight hollow, so I don't worry about measuring for a straight edge.

Is the straightness critical, perhaps I am missing something.

/p

[Hank Knight]

Mike,

This probably isn't a direct answer to your question, but I'm most concerned about "straight" edges when I'm jointing two boards for an edge glue-up. In this case, I use the two boards as references for each other. I joint one until it is "straight" and square as best I can tell (I do check with a straight edge and a square). Then I joint the other board so that it matches the first one. I really don't care if it's "Starrett straight", I care only that it matches the first board's edge. If I can balance one board on top of the other, edge to edge, with no rocking or slipping and there is no light between the two, I'm there. If the mate between the two is right, the balance feels solid. If the balance is the least bit precarious, if the top board wants to spin on a high spot or wobbles the least bit, I take another shaving or two. More often than not I can tell exactly where the problem is and direct my attention there with a bare shaving or two. Then I go back and take one full length, end-to-end shaving and try the balance again. I repeat this until I get the desired match.

I don't "spring" edge joints, but it's easy to do after you've reached this point. Just take a thin, less than full length shaving from one of the boards to let a little light through and then pull the joint together with clamps. If you have a good solid match to start with, a very thin, full width shaving from one of the boards shouldn't affect the match except to give you the "spring."

For most other situations, "eyeball straight" is fine.

Hank

[Randy Klein]

+1 on everyone who has mentioned stopped shavings, followed by through shavings until a full shaving appears. Assuming your plane is flat enough, this works every time.

Anyways, there is no way to have a perfectly straight edge and no real need for one, IMO. I simply shoot for the smallest convex edge that would become concave with one more through shaving.

[Bob Glenn]

Mike, it seems to me that if you took two boards and planned them together side by side until you were getting continuous shavings simutaneously off of both, you would be able to tell if they were flat by comparing the two edges against each other.

If you were low in the middle, there will be a gap when the boards are edge on edge. If the boards rock back and forth from end to end, you know they are high in the middle. If the boards tilt back and forth from corner to corner, you know they have a twist the length of the edge.

If they sit flat, I think then, they must be straight.

[Mike Henderson]

With the method I use, the length and inherent straightness of the jointer plane determines the board straightness. If the edge is planed hollow with stop shavings, then a final, full length cuts are made until a full length shaving appears, then the edge will be straight due to the flatness of the plane. That is of course assuming the sole of the plane is not badly concaved.
I keep coming back to my original question -"How do you know the edge is straight?" You can believe the edge is straight off the jointer plane, but how do you verify that?

The only suggestions I've read here are to use some reference straight edge.

When we do the face of a board, we use winding sticks to verify that there's no twist in the board. I was wondering if there's anything other than a reference straight edge to check an edge for straightness.

Mike

[Mike Henderson]

Mike, it seems to me that if you took two boards and planned them together side by side until you were getting continuous shavings simutaneously off of both, you would be able to tell if they were flat by comparing the two edges against each other.

If you were low in the middle, there will be a gap when the boards are edge on edge. If the boards rock back and forth from end to end, you know they are high in the middle. If the boards tilt back and forth from corner to corner, you know they have a twist the length of the edge.

If they sit flat, I think then, they must be straight.
I thought about that but I've had situations where I wanted a straight board and didn't need a second one. I guess I could use a sacrifice board just to check for straightness but I was wondering if there's any other way.

Mike

[Randy Klein]

I keep coming back to my original question -"How do you know the edge is straight?" You can believe the edge is straight off the jointer plane, but how do you verify that?


It seems you're delving into philosophical land, so I'll answer your question similarly.

Do not try to know the edge is straight — that's impossible. Instead, only try to realize the truth: there is no edge.

[Mike Henderson]

It seems you're delving into philosophical land, so I'll answer your question similarly.

Do not try to know the edge is straight — that's impossible. Instead, only try to realize the truth: there is no edge.
No, it's not philosophy. Let me give an example of the difference in using a technique that generally gives a straight edge and verifying that the edge is straight.

Someone uses some technique to plane an edge, then checks the edge for straightness. "Wow, that edge is straight!"

They do it again and have the same results. Eventually, they feel that as long as they use the same technique, they'll get a straight edge. But that's a hope. They don't know that they got a straight edge, only that when they used that technique in the past they got a straight edge. If something changed - maybe the use of different tool, or a slight difference in technique - the edge may not come out straight and they won't know unless they verify the edge.

But back in the beginning, they had to verify the straightness of that edge. How did they do that? That's what I'm asking.

Is there any way to verify the straightness of an edge other than checking it against a reference straightedge, or using the two board approach suggested by Bob Glenn.

[Maybe the answer is to use Bob's two board approach which will give you two reference straightedges. From then on, you can use one of those straightedges to verify the boards you plane.]

Mike

[Caspar Hauser]

a tight string or wire.

[harry strasil]

http://www.sawmillcreek.org/showthread.php?t=92355&highlight=shooting+board

[harry strasil]

they also make fences for planes to keep them straight, its easy to make one from wood to clamp to the side of your plane.

and if you have a known long straight edge you can clamp it to your board and use the fence as a depth stop.

[James Scheffler]

But back in the beginning, they had to verify the straightness of that edge. How did they do that? That's what I'm asking.

Is there any way to verify the straightness of an edge other than checking it against a reference straightedge, or using the two board approach suggested by Bob Glenn.

Mike

I think that the simple answer is that you need a reference of some sort - a straightedge, a flat bench, or another board.

If you do it enough, using whatever technique you prefer, you may reach a point where you no longer have to check. I will be surprised if I ever reach that point. ;)

Jim

[harry strasil]

tight string or small thin wire stretched tight is a true straight edge.

[harry strasil]

If you have a place that uses, sells aluminum close, go buy a piece of 3/16 by 3 6061 T6 extruded aluminum flat, its about the closest thing to a really high dollar straight edge you will find. I have an about 8ft piece left over from a job and its what I used with a router and edge trimming bit to joint the edges of the oak planks for my 5ft work bench.

[Wilbur Pan]

I keep coming back to my original question -"How do you know the edge is straight?" You can believe the edge is straight off the jointer plane, but how do you verify that?

You could use your jointed edge to draw a straight line on a piece of paper, flip the board, and see if there is any error in comparing the line to the flipped board.

[Gary Lockyer]

I may be missing the plot completely, but here's an article on checking straight edges for straightness http://www.hntgordon.com.au/checkingyourstraightedge.htm

I'd assume you could do the same for a planed edge

[Sean Hughto]

I think you need to ask how straight the edge needs to be in a given application.

Does it really matter if the edge of a table is not straight to stationary power tool tolerances?

As long as two edge jointed boards match up for an invisible joint, does it matter is they are straight in some abstracted sense?

A big part of what makes hand tooled pieces have a bit more soul is that some tolerances are up to the maker and everything is hand fit to other parts. It produces a result that can rival machined results in terms of fit, but has more hand in it at the same time. When a woodworker uses a plane to fit a drawer or door, he or she can account for out of square or out of straight openings to produce even beautiful reveals. The same sorts of things happen all the time in other joints and situations in pieces. It is the beauty of hand work. In short, straight might be called for in some circumstances, but well fit is more important in all circumstances.

[Mike Henderson]

You could use your jointed edge to draw a straight line on a piece of paper, flip the board, and see if there is any error in comparing the line to the flipped board.
That's an insightful suggestion, Wilbur. It would be good for producing a reference straightedge. In a way, it's similar to Bob's suggestion of using two boards - the paper replaces one of the boards.

Gary - the article you pointed to is similar to what Wilbur suggested.

Thanks,
Mike

[Preston Baxter]

I once heard that in woodworking, a straightedge will reasonably check a surface up to 3 times its length. I don't know if that is true on not, but most hand planed surfaces are spherical in nature, so a shorter straightedge will still show if the surface is flat or not, the magnitude of the indication will just be less.

As to the question of how the original straightedge developed, its easy, well maybe not labor wise, but in concept. You take three metal bars, pick a face side on each that will be your precision edge, then lap them against one another in turn. i.e. 1-2, 2-3, 3-1, 1-2 etc. Eventually all three surfaces will become flat. You can't do just two because this could generate a 2 spherical surfaces one convex and one concave. Three will cancel all errors.

To check if they are straight, Blue up the face of #1 with a very thin coat of prussian blue marking compound and press it against #2 . If 1 transfers enough points to 2 and then 2 checks fine against 3 , then you are there.

[Mike Henderson]

I think you need to ask how straight the edge needs to be in a given application.

Does it really matter if the edge of a table is not straight to stationary power tool tolerances?

As long as two edge jointed boards match up for an invisible joint, does it matter is they are straight in some abstracted sense?

A big part of what makes hand tooled pieces have a bit more soul is that some tolerances are up to the maker and everything is hand fit to other parts. It produces a result that can rival machined results in terms of fit, but has more hand in it at the same time. When a woodworker uses a plane to fit a drawer or door, he or she can account for out of square or out of straight openings to produce even beautiful reveals. The same sorts of things happen all the time in other joints and situations in pieces. It is the beauty of hand work. In short, straight might be called for in some circumstances, but well fit is more important in all circumstances.
I agree with you, Sean, and often (usually) wind up "fitting" my work.

I was jointing an edge the other day, using a straight edge to check the work. I started wondering if there was any other way to check if the the edge was straight and thought I'd ask for the wisdom of the forum.

Mike

[Sean Hughto]

I was just trying to channel the philosophical vibe in the thread. ;-)

[Tri Hoang]

It may sound odd/lazy...but after a while, I don't often check for straightness. After a routine full length/stop/full length shavings, I'd consider it's straight enough, especially on the shorter lengths (< 30"). If it need to be jointed with another edge, I'll know as soon as I put them together.

If you need to know for sure, there isn't any other way besides checking it against a known straight edge of some sort.

[george wilson]

We just got real practiced at it when I worked solely by hand,and fitted long boards with wooden jointer planes,no straight edges or squares. Harpsichords have some pretty long boards in them.

[Randy Klein]

tight string or small thin wire stretched tight is a true straight edge.


That actually depends on how you orient the string relative to gravity. No matter how tight you pull a string or wire, gravity pulls on it some and the resultant shape can be described by a hyperbolic function. Interesting enough, solving that problem is how Newton invented calculus, becuase the math of the day could not do it. But I digress.

So you would have to lay the board down and pull the string in such a way that the downward force of gravity does not impact the measurement.

[Pedro Reyes]

The whole question sounds like someone trying to measure temperature without a thermometer, waiting perhaps for someone to suggest a buoyancy vessel...all this when in reality perhaps the practical thing to do is step outside for 10 seconds and know exactly what to wear.

But to provide something that could be useful...

I think you can draw several very straight parallel lines close to one another using your straightedge (is that straight? How do we check this? ;) ). Once you draw these lines, say on the bench, simply superimposing the board should reveal any deviation from straight, the eye is an amazing qualitative measuring device.

just a thought.

/p

[Tom Henderson2]

How do you edge plane a board straight (with a hand plane)?

<snip>

Any alternative to checking it with a known straight edge?

Mike

Good question Mike, and one that I was wondering about, too.

I'm a newb and don't know much!

Many are recommending the "stopped shaving" method, and I've used that to good effect on shorter boards. But I believe that only works on boards that are no longer than ~1.5 times the length of the plane being used. So for a 22" jointer plane, that means about 33" or thereabouts.

So I'm not sure that method would work on the six-foot board you are contemplating.

But the factory edge of a piece of MDF would seem to be an excellent and cost-effective reference edge.

-TH

[Mike Henderson]

The whole question sounds like someone trying to measure temperature without a thermometer, waiting perhaps for someone to suggest a buoyancy vessel...all this when in reality perhaps the practical thing to do is step outside for 10 seconds and know exactly what to wear.

But to provide something that could be useful...

I think you can draw several very straight parallel lines close to one another using your straightedge (is that straight? How do we check this? ;) ). Once you draw these lines, say on the bench, simply superimposing the board should reveal any deviation from straight, the eye is an amazing qualitative measuring device.

just a thought.

/p
I agree that using straight lines could be one way to do it. But it's not very convenient. You'd have to take the board out of the vise whenever you wanted to check it. Making a straight edge using one of the techniques suggested earlier in this thread would be easier to use.

I don't understand the reference to trying to measure temperature without a thermometer and certainly don't see any analogy. I asked how to determine if an edge was straight and that seems to me like a very valid question. And several good suggestions were made in response to the question.

Mike

[I didn't respond earlier to the suggestions of using a long shooting board because that only begs the original question. Before you can use a shooting board, you have to make the shooting board straight. And if we assume use of hand tools, the question then is, "How do we plane the boards for the shooting board straight (and verify that they're straight)?"]

[Robert Rozaieski]

[Maybe the answer is to use Bob's two board approach which will give you two reference straightedges. From then on, you can use one of those straightedges to verify the boards you plane.]

Mike
Mike, I think you answered your own question here. While they may not have had Starrett straight edges (which are only straight to a certain tolerance mind you) in the 18th and 19th centuries, they certainly could and did very easily make them from wood. Simply match plane two long boards to what you believe is straight [enough]. Then check the two planed edges against each other by unfolding them like a book. When the two edges meet without any noticable gaps, the edges are straight. You then have two straight edges that you can use to check your work, IF it absolutely must be straight. However, I think in most situations, absolute straight probably isn't necessary to check for. If it's an edge joint, you simply plane until the two edges join in a seamless joint. Doesn't mean they have to be NASA tolerance straight. I think you're trying to apply contemporary/machine thinking to a not so precise operation.

To answer the question how do I KNOW it's straight...I don't. And I'm ok with that. When it's good enough, it's good enough. I don't need to quantitate it ;). It just needs to fit.

[Pedro Reyes]

I don't understand the reference to trying to measure temperature without a thermometer and certainly don't see any analogy...

Mike,

I work in an industry where we measure surfaces in Armstrongs. That's why the question seems strange. LIkely I'll show aything is faulty if I measure it like that.

The analogy referes to, why do I need ot measure in armstrongs when all I need is for the piece to fit? (both mechanically and aesthetically)

I believe in the "slightly concave" edges are better; theory for edge joints, most do, certainly better than a slightly convex scenario. Beyond that straight is only in reference to another mating piece more so than to any perfect surface, or in reference to what pleases the eye (as in the outer boards on a table). Greek columns are not straight, for a reason.

/p

[Narayan Nayar]

As Robert, Pedro, and others have intimated, "straight" only exists in theory. Much like (perfect) circles don't exist outside an equation, (perfect) straightness only exists as a vector between two points (also in an equation).

If you accept that fact, then there are only degrees of straightness. And the only ones that matter in woodworking (or anything else practical) are ones in which one non-straight edge matches another non-straight edge. In other words, the only degree of straightness that matters is one which is straight enough for the intended purposes.

And my intended purposes, at least in woodworking, usually revolve around simple heuristics: does light show under an aluminum straightedge? Does the piece fit? Or even, will the gaps in this edge affect its performance at all?

Frankly, worrying about straightness more than that might be an interesting intellectual debate, but that's about it.

[Michael Panis]

How do you edge plane a board straight (with a hand plane)? I've never figured out how to do it except to take a known straight edge, such as some MDF cut on my table saw and lay it on the edge to check if it's straight.....

...I keep coming back to my original question -"How do you know the edge is straight?" You can believe the edge is straight off the jointer plane, but how do you verify that?


See--you've answered you own question! Put the board on the tablesaw you used for the MDF. Adjust the fence so the blade just touches the edge and run it through. If you see any sawdust, it wasn't straight.:cool:

(Sorry--couldn't resist!)

[Mike Henderson]

Mike,

I work in an industry where we measure surfaces in Armstrongs. That's why the question seems strange. LIkely I'll show aything is faulty if I measure it like that.

The analogy referes to, why do I need ot measure in armstrongs when all I need is for the piece to fit? (both mechanically and aesthetically)

I believe in the "slightly concave" edges are better; theory for edge joints, most do, certainly better than a slightly convex scenario. Beyond that straight is only in reference to another mating piece more so than to any perfect surface, or in reference to what pleases the eye (as in the outer boards on a table). Greek columns are not straight, for a reason.

/p
I am certainly not advocating getting a board straight within an angstrom or two, or asking how to do so. I was simply asking how to get a board straight within woodworking tolerances.

I well understand the concept of "fitting" but even with that, we all strive to make our stock flat and square. My question is (was) - within woodworking tolerances - how to check the edge of a board to see if it's straight.

Several people have given reasonable and valuable suggestions on how to do that, and I appreciate their suggestions.

Mike

[Let me add one additional note: Michael Panis offered a humorous suggestion about using the table saw. But one reason I ask this question has to do with the table saw. At my age, I don't try to rip all my stock by hand. But before I can use the table saw, I need a reference edge, straight and true. I don't have a powered jointer so I prepare one edge with hand tools, and I want to make sure it's straight before I try to rip the board on the table saw. If it's not straight, all the pieces I rip will follow the curve of the reference edge.]

[Jim Koepke]

This has been an interesting thread that has many different good answers to the original question.

Trying to sight down the edge of a board used to be good, but now with bifocals, it is a bit more difficult as the board gets longer. It does seem easier if the corner where the edge and the face meet to make a line is viewed.

Recently another interesting discovery was uncovered. My #8 is set up to take a medium, 0.005-0.007", to thick shaving. It does not leave a wonderfully smooth edge. My #7 is set up to take a very thin shaving, 0.001-0.002". It can leave a very smooth edge with a sharp blade. After an edge was "trued" with my #8, the #7 was used to make a final pass. The only problem was the slightly low areas did not make a shaving. It took two or three passes to get a full length shaving. This leads me to feel that taking a thick shaving may allow for some bumps and valleys to remain that a plane set to a thin shaving will find.

This may only have validity in my daydreams, but it seemed like an interesting observation.

jim

[Randall Houghton]

Mike
Being a third generation carpenter I was taught to test edges for straightness with a braided fishing line and three spacers of equal thickness.
Attach one end of the line to the end of the board and stretch it down the edge of the board tight and attach the other end. Stretch the line enough to slip a spacer under it at each end of the board. Use the third spacer to check the gap between the line and the edge of the board to indicate the high or low spots on the edge. The more tension you can put on the line when you attach it the more your results will be. Alternately I made an 84" straight edge out of quarter sawed straight grained DF and finished with a couple of coats of shellac that I used to straighten door jams. I've had it about four years and check its straightness frequently and it has stayed within 2 thousandths even as the weather changes plus it's light weight and easy to handle. Practice will train your eye to be as accurate as any straight edge.
Regards
Randy

[Robert Rozaieski]

I well understand the concept of "fitting" but even with that, we all strive to make our stock flat and square. My question is (was) - within woodworking tolerances - how to check the edge of a board to see if it's straight.

I think that's the point you might be misunderstanding though Mike is that most of the time, many of us [or at least me :)] don't check the edge to see if it's straight. In many cases, we [or at least I] may not even care if it's straight or square. The mentality that every part must be straight and square on all sides is rarely practiced when you do a lot of the work by hand [again, at least in my own personal case] because often it simply doesn't need to be. This four-square (or six-square) mindset is one that comes from the machine world where you need all of those "perfect" faces for reference on a machine fence or table. This need simply doesn't exist if you aren't relying on a machine surface as a reference. So I think you are having a hard time getting a straight simple answer because many of us simply don't check at all.

However, there are of course occasions where one does need a straight and/or square edge and so I check just like anyone else would. There aren't any magic short cuts, I use a 4' straight edge (albeit mine is one I made of wood). It's not all that cumbersome to do so I see no need for any other kind of secret shortcut. I check my straightedge occasionally by using it to draw a line on the top of my bench and then flipping the straightedge on its long axis and checking it against the line I drew. Any deviation from straight will be doubled and easy to see so I then plane the straight edge until checking in this manner proves the straightedge to be true. My straightedge is then plenty straight enough for woodworking tolerances.

[Preston Baxter]

I forgot to mention that the rule of threes equally applies to wooden straightedges as well as metal. Get three lengths of some stiff, stable stock, and plane up an edge on each one. Keep checking them against each other, and plane out the errors until they all mate properly with each other then they will be perfectly straight...well perfectly straight for all woodworking purposes anyway.

Need to know if they are still straight before their next use on a project, it's easy, just check them all against each other.

[Jim Koepke]

Practice will train your eye to be as accurate as any straight edge.
Regards
Randy

When my job began in the last department before my retirement, the technical trainer often mentioned "using your calibrated eye."

Before leaving, one of my assignments was to write a technical manual on the rebuilding of the largest ticket transport we worked on.

As part of that, was an exercise to train the eyes. It involved using a caliper to measure spacers we had in different sizes from 0.001" to 0.040". After looking at the edge of each one before measuring, a person could get pretty good at judging small sizes. Kind of like a mechanic looking at a nut and knowing what size wrench they need.

The same applies to working wood and seeing if an edge is square. It is all in paying attention and knowing what you are looking to see.

I am getting away from measuring and "testing" and more to fitting and trusting my eyes and hands.

jim

[Mike Henderson]

I think that's the point you might be misunderstanding though Mike is that most of the time, many of us [or at least me :)] don't check the edge to see if it's straight. In many cases, we [or at least I] may not even care if it's straight or square. The mentality that every part must be straight and square on all sides is rarely practiced when you do a lot of the work by hand [again, at least in my own personal case] because often it simply doesn't need to be. This four-square (or six-square) mindset is one that comes from the machine world where you need all of those "perfect" faces for reference on a machine fence or table. This need simply doesn't exist if you aren't relying on a machine surface as a reference. So I think you are having a hard time getting a straight simple answer because many of us simply don't check at all.

However, there are of course occasions where one does need a straight and/or square edge and so I check just like anyone else would. There aren't any magic short cuts, I use a 4' straight edge (albeit mine is one I made of wood). It's not all that cumbersome to do so I see no need for any other kind of secret shortcut. I check my straightedge occasionally by using it to draw a line on the top of my bench and then flipping the straightedge on its long axis and checking it against the line I drew. Any deviation from straight will be doubled and easy to see so I then plane the straight edge until checking in this manner proves the straightedge to be true. My straightedge is then plenty straight enough for woodworking tolerances.
You're absolutely correct, Robert. I mix hand and machine tools in my work and I think in terms of straight and square. When I'm using hand tools, I strive for accuracy and perfection (which I rarely achieve to my satisfaction). I understand what you're saying and I appreciate you taking the time to express it.

Mike

[Augusto Orosco]

Interesting question. Just for the fun of the mental exercise:

1) Use a plomb line. You would need a steady hand to hold that board against the line, though!

2) Some light projected carefully using appropriate lens (like some of those inexpensive lasers sold at the Borgs)

2a) Much more involved than (2): Cut three identical pieces of any opaque material. stack them and make a small hole through them. Place one piece at the beginning and at the end of the board and project a line of light in between the two. Then place the third piece in between an move it all through the length of the board. If the line of light is interrupted at any point, the board is not straight. (the smaller the light, the more precise the measure).

3) Fill a long glass tank with colored water. On a perfectly steady surface, the water line should form a straight line.

4) Check with a straight edge twice: Flip along the edge the second time. This will not guarantee that the board is flat, but at least reduces the chances of error induced by a not fully straight straightedge

Given the impracticallity of 1, 2 and 3 and the imperfect nature of (4), This is probably not a lot of help to the OP, but I enjoyed thinking about it!

[Pedro Reyes]

Several people have given reasonable and valuable suggestions on how to do that, and I appreciate their suggestions.

Mike

Mike,

Sorry if my answers upset you in any way, I was trying to understand the need for straightness and to what degree. back to my suggestion, you could always draw your very straight parallel lines on the wall behind your bench, so as to have your board being edge planed project a shadow... just a thought.

peace

/p

[Mark Roderick]

Yes, this is the exact correct answer!

We don't CARE if a board is perfectly straight in woodworking. All we care is that it mate against the board next to it.

So we rip both boards, get them both as straight as we can using a template or whatever, then match them against one another. If they fit against one another with no gaps, we're done! If not, we use our jointer plane until they do.

This is just another instance in which "measuring" isn't really the point in woodworking.

[Mike Henderson]

Yes, this is the exact correct answer!

We don't CARE if a board is perfectly straight in woodworking. All we care is that it mate against the board next to it.

So we rip both boards, get them both as straight as we can using a template or whatever, then match them against one another. If they fit against one another with no gaps, we're done! If not, we use our jointer plane until they do.

This is just another instance in which "measuring" isn't really the point in woodworking.
Maybe I should have asked my question in a different way.

"Let's begin by making the assumption that I want a straight edge on my board. How can I verify that the edge is straight to woodworking tolerances?"

I appreciate all the comments about how I don't need a straight edge on a board, but I want one and was asking how to verify that the edge is straight.

Mike

[Pedro Reyes]

...straight to woodworking tolerances?"
...

Mike

Somewhat subjective agree?

I believe you also had a constraint of "alternate way to using a straight edge" which covers anything straight used (e.g. a board)

If we accept the premise that a decent straightedge is good enough for "woodworking tolerances" when it comes to edges.

Then I will also assume that you would like it to be simpler/faster/cheaper than using a straightedge, fair?

I don't think there is, I mean grabbing a straightedge and checking is very simple/fast and certainly cheaper than a lot of alternatives.

I only keep falling back to my lines on the wall with a crisp shadow. If we want to avoid drawing lines (since these could be off straight), you could always get two light beams, one level one angled, project two shadows on a wall (one dark one, being the overlap of both shadows and one lighter shadow) if that band of lighter shadow has perfectly parallel edges, then you should be straight, if not there is a curve. At leats in theory that should work. No need to move boards, you can check after each stroke... just a thought.

/p

[Bob Strawn]

With practice, you can quickly look down a board and see if it is straight. It is really simple, you line up the center of one eyes vision where you are looking from the corner near you to the corner at the other end of the board. Then you lift your point of view to allow you to clearly see the other end of the board. Then you slowly lower your point of view back to the surface of the corner. If the end disappears at the same time as everything else, then you have a straight board. It may take a few viewings, but you can even map out where the high spots are and do a good job of guessing where the low spots are.

With practice you can do this fairly quickly as well. If the end is obscured before any other spot, that spot is high. If the end is visible, but any spot is not, then it is low. If the board is shorter than your arm, you can slide your finger down while you view it, to find the exact spot.

A string stretched end to end is a good aid if you need to find the exact spots on a longer board.

Bob

[Mike Henderson]

I'm going to practice that, Bob. I've tried it in the past and have had difficulty telling if the edge is straight but maybe I didn't give it enough practice and time.

I'll try looking down the edge and then checking what I see with a straight edge. Eventually, I hope I'll be able to get the two checks to agree.

Thanks for the suggestion.

Mike

[Rick Erickson]

couldn't you put the board on a flat surface (face side down) and trace the edge with a pencil. Then flip the board over (other face side down) and trace another line right next to your first line. They should be parallel? This works for checking a try square.

[Mike Henderson]

couldn't you put the board on a flat surface (face side down) and trace the edge with a pencil. Then flip the board over (other face side down) and trace another line right next to your first line. They should be parallel? This works for checking a try square.
Just to clarify, I think what you're suggesting is that the board would be flipped over side-to-side and not end-to-end. I would probably use that technique (or some other) for creating a reference straight edge, which could then be used to check the work edge. My reason for that approach is that it's easier to put a straight edge on your work, than to take the work out of the vise and check it multiple times.

Thanks for the idea.

Mike

[Don Shirey]

I believe you could determine straightness by using one of the leveling type lasers, or perhaps constrictiong a holding jig and using a laser pointer to iluminate the surface you are concerned with. I have not given it enough thought to design the jig.

Cheers, from wet and cold Phoenix

[Don Shirey]

Sorry that should read "constructing"

[Mike Henderson]

Sorry that should read "constructing"
Just a forum tip - If you click on "edit" in your original post, you can make corrections, additions or changes for a period of about 24 hours after you made the original post. After that, the post is locked and you can't change it.

Thanks for your suggestion on how to check the edge.

Mike

[David Gendron]

What came first? the edge or the straight edge? how do you know that your "straight" edge is indeed straight and thrue?

[Mike Henderson]

What came first? the edge or the straight edge? how do you know that your "straight" edge is indeed straight and true?
Several people have suggested ways of verifying a "straight edge" - check back in the thread. Those techniques, along with some extra care, can produce a reference straight edge that's sufficient for woodworking applications.

Mike

[David Gendron]

IMO, you are waisting your time. That of course doesn,t apply to you Mike.

[Mike Henderson]

IMO, you are wasting your time. That of course doesn't apply to you Mike.
:confused::confused::confused:

Mike

[James Taglienti]

It is mindblowing that there have been over 60 responses to this question

[Eric Brown]

Here is another approach: Tape a laser pointer to the side of your plane and set the plane on top of the boards edge. Then put a mark on some tape where the laser is pointing. Any deviation while planing will result in the pointer moving up and down. The further the distance the more accurate it is, but it is also harder to see. (Might need help).

Eric

[Chip Nielsen]

How about holding the edge of the board up to a plumb bob line and checking with a flash light? If there is a gap the light will produce a shadow. Just a thought.

[Mike Henderson]

I settled on testing the edge with a known straight edge - which was what I had been doing.

Now, if I could just plane square... My edges always seem to slant to one side or the other.

Mike

[Jim Koepke]

Now, if I could just plane square... My edges always seem to slant to one side or the other.

Maybe a small level across the plane?

jim

[Burt Clarke]

I can see where MIke is coming from. I like Preston's answer. If your plane is long enough, and it takes a clean shaving all the way from one end to the other, the board is straight. I can't rely on a power jointer to give me a straight edge on an 8' board that's 5 in x 5 in. When wrestling a board that size, I run the risk of introducing more flaws than I started with. However, a jointer plane won't lie to me. And where do you draw the line in length? You can't always have a straightedge as long as every board you'll ever work with. Also, if I try sighting down a board, these no-line bifocals will put a curve in the end of the board every time. There's just something about taking a clean shaving all the way down the edge that tells you that it's going to line up nicely with an adjacent board that you did the same thing with. IMHO.:p

[Steve Branam]

Mike
Being a third generation carpenter I was taught to test edges for straightness with a braided fishing line and three spacers of equal thickness.
Attach one end of the line to the end of the board and stretch it down the edge of the board tight and attach the other end. Stretch the line enough to slip a spacer under it at each end of the board. Use the third spacer to check the gap between the line and the edge of the board to indicate the high or low spots on the edge. The more tension you can put on the line when you attach it the more your results will be.

I'm glad this thread got bumped up so I could see this. That's an excellent method! That's the kind of old-school experience I read this for. Good for those boards longer than 4' (reasonably good 4' straightedges being easy enough to come by, but not longer ones). Also a very practical on-site method.

[john brenton]

In answer to the question of their being an alternative, if you have at least one trued side you can measure from that all across the length of the board.

Or, even if you didn't have a trued side you could pencil in a line using a straight edge and measure from that.



How do you edge plane a board straight (with a hand plane)? I've never figured out how to do it except to take a known straight edge, such as some MDF cut on my table saw and lay it on the edge to check if it's straight. Then, I can focus on the high points and get the board straight to my satisfaction. I'm talking about a long board and not something 2' long. Maybe a 6' board.

So maybe I should ask the question in a different way, "How do you know an edge is straight when hand planing it?" I've tried looking down the board, and maybe my eye is not good enough, but it can look straight and still be bowed when checked against a known straight edge. Any alternative to checking it with a known straight edge?

Mike

[Trevor Walsh]

It comes down to whether "straight" means straight for edge joining two boards or "I'm making straight-edges" straight. For edge jointing a little hollow is no big deal and working from the plane will give you a straight enough joint.

If however the goal is to generate straight edges from scratch you either need to work with two boards and check (with special technique that prevents you from fitting two dissimilar profiles) or use a known straight-edge. Google "making accurate straight edges from scratch" for a great explanation of the technique.

[Mike Henderson]

It comes down to whether "straight" means straight for edge joining two boards or "I'm making straight-edges" straight. For edge jointing a little hollow is no big deal and working from the plane will give you a straight enough joint.

If however the goal is to generate straight edges from scratch you either need to work with two boards and check (with special technique that prevents you from fitting two dissimilar profiles) or use a known straight-edge. Google "making accurate straight edges from scratch" for a great explanation of the technique.
I Goggled as you suggested but all of the techniques I found required three pieces of stock which are placed against each other in different orientations.

What I wanted is a way of knowing when a board is straight when hand planed. I use a piece of MDF that I've cut straight on my table saw and that's close enough for my usage. The piece of MDF just has to be wide enough that the MDF doesn't sag or bend (much) when I use it.

This thread did generate some very interesting comments - thanks to everyone who participated.

Mike

[Burt Clarke]

I'm glad this thread got bumped up so I could see this. That's an excellent method! That's the kind of old-school experience I read this for. Good for those boards longer than 4' (reasonably good 4' straightedges being easy enough to come by, but not longer ones). Also a very practical on-site method.

I know that Mike came up with a reasonable solution, but now I've got to try this. Some braided fishing line and three fender washers ought to do it. Even when I'm sure it's straight, when it is all done, I still stack the boards I am going to join on top of each other and shine a light behind to see if I've got a good fit.

[Trevor Walsh]

That was just if you intended to make reference straight edges, I wasn't sure what your application was, so I went technical. I see it's more along the lines of straight for constructions or glue joints than making a setup tool.

Best of luck,

[Dan Sheehan]

Caspar has it. You could nail a tightly braided string (or dark fishing line?) to the end of the board and the wind it tightly around a nail at the other end of the board. Take two shims of equal dimension. Place one under the string at each end. Use a third shim, maybe a 32nd under the others, and run it along the board, looking for larger gaps or places where it rubs the string, I think that will indicate what's "straight" in our fallen world.

I don't do this in the shop, however, but a similar method for checking for straight works well in building construction.

Speaking of philosophical:

"A wise man seeks only as much precision as his subject matter permits" - Aristotle

[Ben Rivenbark]

This was an interesting and thought provoking thread. The idea of how our forefathers of long ago developed a reference edge made me wonder how I would do it. My solution for a reference edge of any length is as follows:

1. Find an area which will be free of interruption by pest or weather large enough to house a reference surface. You could make one thirty feet if you wanted.

2. Build a trough 1 foot wide by 1 foot deep by X feet long. Fill 3/4ths with water.

3. Allow to freeze solid.

4. Done! The surface of the ice should be flat enough for any civilization which doesn't have reference edges already...

I guess you could even go so far as to pour an extra half inch or so of water after the rest has frozen solid to reduce any errors resulting from expansion during freezing and let that freeze. This could also serve to resurface the reference similar to a zamboni...

Now obviously this.would be difficult to do in southern Florida and is limited by several other things, but its the best i could come up with, especially if a really long reference surface was desired.

Sorry for any typos. I wrote this on my phone.

[Johnny Kleso]

There is only one way to check if a board is straight.......
Check it......
Straight Edge or some other high tech method..

Like most other said the hand plane is made to flatten boards period..

I have some 1/4x 3 steel in different lengths I use as straight edges for cutting sheet goods and also use them for checking..

To inspect them you lay down a strip of blue painters tape on a bench use your edge and cut a line down the middle with a razor and fip the steel edge you just cut agaist the edge you just cut and it should be perfect..

[Gaz Palmer]

If you read up on boning rods, winding sticks and ranging lines, you'll uncover the means to check for straight and twist over long and short distances using simple tools and eyesight.

1. Boning rods. - three "T" shaped rods of equal dimensions, with one white, one red and one black. Place one rod (Red and black) at each end of the area to be levelled, with the top plain level. Using the white rod to check intermediate levels (The traveller), you sight across the top of either the black or red rod and raise or lower the intermediate traveller until it's crown coincides with the crest of it's counterparts. Accurate to within 1.5mm (1/16") over a distance of 50m.

2. Winding sticks. - I think most know of their use.

3. Ranging line and spacers/pegs. - As mentioned earlier, three equally dimensioned spacers, with one placed at each end of the piece being checked for true and a fine line stretched taught across their crowns. The third spacer is used as a traveller to check intermediate heights and a surface is considered true once the traveller can be snugly slid between the line and test surface without displacing the line. Accuracy = 10m (Depending on line gauge & tension, but limited by the effect of gravity and resulting sag in the line over distances exceeding 10m)

Numbers 1 & 2 can also be used for setting out both arched or concave surfaces, whilst number 2 can be used for laying out both flat and spiralled work. Each of the above have been in use for millennia by masons and carpenters.

Another method involves the use of water levels and the simplest means of producing a straight edge is by suspending a plumb-bob down the face of an upright length of timber before marking intermediate points and joining the dots. Remove waste just shy of the resulting line before planing the timber true and producing a straight edge. Accurate to the thickness of the resulting scratch line.

All five are ancient methods.

The final method, but my first choice, would be to use a long spirit level with machined faces. :)

[Burt Clarke]

Some very interesting posts in response to Mike's question. I thought I was old school. Well, I guess just old. :o Anyway, all of the suggestions will get you within a sixteenth for the most part. I tried a number of them, including the braided line and spacers. I didn't have much luck with the taught line. Even when I made the line sing, it always showed the ends high, even when they were low. Guess I just don't have the knack. All of them have merit for one purpose or another; most were good enough for choosing a "straight board" for a project; but not good enough for me for joining a quality table top or benchtop. Guess I'm a perfectionist, but 1/16th just doesn't seem close enough. Obviously, if the boards are short enough, most table saw tops or jointer tables are good enough. But, I don't think anything stacks up against fitting up against a known straight edge. I am not opposed to abandoning "old school" if it works. Here is a tip that I got from a Woodworkers Journal video and it works for me. Buy a 20 gage steel stud of a length longer than the board you are checking; you will be amazed at how reliably straight and rigid they are - in spite of the fact that they are very light and very cheap. I know that, if I can't see light between the stud and the edge I'm working, it's straight enough for jointing. They make wonderful saw guides clamped to a bench or grid, if you can't afford a festool. I have a 10 footer, and I cut an 8 footer to 5 ft. for cutting across plywood sheets. Obviously we are just talking straightness here; you still have to make sure the edge is square to the face along its entire length.

[David Christopher]

draw a strait line on a board, put the board on the bench behind the board beeing planed...plane a few licks, then squat down and eye the board beeing planed with the board with the line on it........

you wont have to take the work out of the vice..just squat and look