Home Articles Forums Gallery Marketplace Videos

WoodCentral Forums

Knowledge You Can Build On — Est. 1998

Find a woodworking club near you

Springing a Smooth Symmetric Curve

Posts

Springing a Smooth Symmetric Curve

#1

Springing a Smooth Symmetric Curve

don stephan

Not sure "springing" is the correct descriptive, but I need to lay out a symmetric gentle curve for poplar clamping cauls 50" long. Much too gentle a curve for trammel points and a long pole.

Many times I've seen a suggestion to use any thin strip of wood, but I worry that changes in density, growth ring pattern, and more could cause the strip to bend more in some areas and less in others.

I have a six foot aluminum "yardstick" but not sure it will bend as much as I need. Do I simply look for the most consistent quartersawn grain on a board, saw off a narrow strip, and hope for the best? Anyone know how this was done 100 or more years ago?

Re: Springing a Smooth Symmetric Curve

#2

Re: Springing a Smooth Symmetric Curve

Bill Tindall, E.Tn.

I need to lay out a symmetric gentle curve for poplar clamping cauls 50" long.........I have never seen any need for clamping cauls. I have never seen them used either in a furniture shop or factory. I do add some thin sticks to prevent denting of the edge.

Many times I've seen a suggestion to use any thin strip of wood, but I worry that changes in density, growth ring pattern, and more could cause the strip to bend more in some areas and less in others....I have made several curves this way for furniture parts and they are always asymmetric for the reasons you provide. Solution , use half the curve and flip it.

Anyone know how this was done 100 or more years ago? 100 years ago clamps were not cheap and readily available from Lee Valley so cauls may have been needed. But that said they likely just planed the middle out of a piece of lumber and made the curve by eye. I think you are over thinking the precision needed.

Re: Springing a Smooth Symmetric Curve

#3

Re: Springing a Smooth Symmetric Curve

Lee Schierer in McKean, PA

I don't know about how it was done 100 years ago, but a thin strip of 1/4" luan underlayment will bend in a pretty consistent curve. If in doubt bend the piece of wood, draw your curve and the flip the wood strip end for end and bend it the same amount again. Any inconsistencies should be readily apparent.

Lee

Re: Springing a Smooth Symmetric Curve

#4

Jesper, Denmark

Re: Springing a Smooth Symmetric Curve

Jesper, Denmark

Don, make a batten from stright grain pine. It is important that it is square crosssection - say 1/4 x 1/4 inch or so. This will bend in a fair curve. If you can put a nail at each of the end points and lay the batten against this, you can bend the batten against the nails and draw the curve along the batten.

Standard boatbuilding technique - where fair non-cylindrical curves are needed all over.

Re: Springing a Smooth Symmetric Curve

#6

Re: Springing a Smooth Symmetric Curve

William Duffield

If all your pine is unfairly gnarly, all is not lost. Lee Valley carries a pair of drawing bows, one symmetric and the other asymmetric. They will give you an infinite variety of repeatable fair curves. Of course, a pine stick is a lot cheaper.

http://www.leevalley.com/US/wood/page.aspx?p=44631&cat=1,42936,50298

Re: Springing a Smooth Symmetric Curve

#7

One word, Benjamin*: plastics

Bill Houghton, Sebastopol, CA

See if there's a retail plastics shop near you. Here on the Left Coast, there's a regional chain called "TAP Plastics." You may find something similar in your area. Go in and describe what you want, and they can fix you up with a length of some suitable flexible plastic in your choice of colors, with which you can make a drawing bow. If I went into my local (next town over) TAP store searching for something like that, I'd probably spend $4-5, max; no idea what you'll find in your area.

-----------

*Sorry. Gratuitous reference to "The Graduate." Probably meaningless to anyone under the age of 55.

Re: Springing a Smooth Symmetric Curve

#8

Oh, I got it instantly.

Rod Peterson—Ormond Beach, FL

Of course I'm way over 55.

That was a seminal picture for me and people my age (guys, mostly). Who can forget Benjamin shaking the doors of the church and shouting, "Elaine!"

Katherine Ross' screen debut, I believe.

And Simon & Garfunkel. What wasn't to like.

Rod

Re: Springing a Smooth Symmetric Curve

#9

Wandering off topic: movie quirks

Bill Houghton, Sebastopol, CA

It's been pointed out that Dustin Hoffman drives both ways on the top deck of the San Francisco Bay Bridge (east to Oakland; west to San Francisco). Westbound traffic runs on the lower deck of that bridge*, but that would not have made a very good movie shot, so they actually arranged to shut down the bridge briefly for the westbound shots, so they could show him on the top deck headed to San Francisco.

*Soon, or maybe soon, to be decommissioned, now that the replacement bridge is finally opened; the old bridge failed rather dramatically in the 1989 Loma Prieta quake, dumping cars into the bay.

Re: Springing a Smooth Symmetric Curve

#10

Re: Springing a Smooth Symmetric Curve

Sgian Dubh

"Many times I've seen a suggestion to use any thin strip of wood, but I worry that changes in density, growth ring pattern, and more could cause the strip to bend more in some areas and less in others.

A piece of MDF does the job very nicely, and all those issues about growth ring pattern, knot placement and wood density disappear. All you need to do is cut a strip of suitably thick MDF about 20- 25 mm wide and ensure that you use the board faces to form the bend. It's rare to see a woodworker without pieces of MDF to hand that can be used for this purpose, so there's seldom a need to go out and buy anything new. Slainte.

Re: Springing a Smooth Symmetric Curve

#11

This is why I love Wood Central.....

Barry Irby

You got at least four choices, no need to do it, wood, plastic, MDF, and one more.....

The two sticks method. Take two straight rigid piece of wood or whatever that are each longer than the curve. Lets say two 1x4's 60" long. On your pattern put a nail at each end and another at the apex of the curve. lay the sticks on the pattern so that they overlap at the apex nail and both touch it and each touches one of the end nails, forming an extremely shallow vee. Clamp them together firmly. Remove the apex nail. Use a pencil in the vee and slide it left to right and right to left until you draw a curve from end to end. (Norm did this one New Yankee Workshop and you can probably look it up) Your curve may be almost too gentle for this method.

If I were doing this I would make half of the curve by eye, probably with a block plane or maybe a disk sander and then flip it. It would take some additional experimentation to find the sweet spot for the curve and I think that might vary a lot with what you choose to use for cauls, how wide, how thick, species, and how long.

Actually, I agree with all of the above. I have never used cauls like you describe and have never seen them used. Perhaps I would be a convert if I had, but so far they have not seemed necessary. If I am doing a large glue up I try for good prep work with pieces that will lay up nicely and not be unruly. I occasionally will ad about three biscuits in each joint just to keep them lined up momentarily if I am a bit nervous. (I bought a biscuit jointer and have to justify it, right?) I lay up the pieces dry, apply glue, clamps in a balanced fashion and wait. I have been known to persuade the occasional recalcitrant piece with a rubber hammer or perhaps to apply a caul to the end of a joint with a C clamp or two. I find most of it is in the prep work, both getting the wood prepped right and getting the clamps and area and glue set up (dry run) first.

Re: Springing a Smooth Symmetric Curve

#12

a reminder

Bill Tindall, E.Tn.

That is how I made the curve for the bow front chest of drawers and it worked reliably and easily. I had forgot till you reminded me.

Re: Springing a Smooth Symmetric Curve

#13

More about this curve (warning technical)

Bill Tindall, E.Tn.

The curve that results from bending a uniform (isotropic) stick (beam) fixed at each end with the force being applied in the middle is not any recognizable geometric function. It is not a section of a circle, ellipse, parabola, etc. The shape of the curve will depend on the distance between the fixed ends and amount of deflection at its middle, but also on the cross section shape of the stick and the stick's strength (elastic modulus).

Hence, it may not be possible for woodworker A to reproduce the curve of woodworker B unless they have identical sticks. And, if the shape needed is an arc of a circle, for example , bending a stick is not a successful route to achieve it.

When considering these curves realize that the curve produced from a bent stick flattens at each end. Will this kind of curve be suitable for the task at hand? I suppose "bow front" chests of drawers are named by the way their patterns were made. But for a table edge, or even your chest of drawer design, maybe a piece of an ellipse would look better, or some other curve. The point to be made is to consider various curves and not rely on bending the fist stick that seems flexible enough.

I became interested in this problem because I thought the curve was some common geometric function and therefor the points on the curve could be easily calculated and plotted without having to rely on finding a good stick material to make it. If I can get my mechanical engineering student neighbor to interpret the link provided below it may still be possible to plot. If there is a knowledgeable mechanical engineer in the house maybe they can enlighten us on the mathematics behind our bowed sticks.

http://en.wikipedia.org/wiki/Euler%E2%80%93Bernoulli_beam_theory#Three-point_bending see section 7.1

Re: Springing a Smooth Symmetric Curve

#14

Re: Wandering off topic: movie quirks

John in NM

I remember a similar situation with some horrible horror movie I watched in HS. It was a collection of Stephen King stories (IIRC) so they filmed each short story in and around Bangor, ME -- I believe King may have been involved in production, and of course, he lives there.

At any rate, one segment had the character driving the same section of bridge over and over, shot at different angles to make it look like a big city freeway. Us locals knew it was the only bit north of Portland that looked anything like the big city :D

Another one I remember was watching "Revenge of the Nerds", which was filmed at the University of Arizona. Had to watch it if you were attending in the 1990's. I remember there were some great examples of "creative geography" in that one. All movies must be like that, its just rare that we watch them with an intimate knowledge of the true geography.

Re: Springing a Smooth Symmetric Curve

#15

Re: Wandering off topic: movie quirks

Barry Irby

There is a TV show (The Following with Kevin Bacon) that is supposed to set in Richmond, VA and the surrounding area. Because it was locally set, I watched some of it. they used some local names like Chippingham Parkway and other street and county names. They made up "local" hospitals and business. I don't know if they really filmed any of it here, but I did not recognize any of it. They filmed a lot of the Lincoln movie here, but they use older buildings in very creative ways and film form odd angles to not catch a glimpse of the newer stuff in the area.

Re: Springing a Smooth Symmetric Curve

#16

Re: Springing a Smooth Symmetric Curve

Dave Bair

You can make perfectly tapered cauls using your table saw and jointer, according to one of the "Tricks of the Trade" segments in the August PWW (page 13). Use your table saw to cut a slot in each end of your caul stock (twice as tall as your final caul dimension), leaving about 10" of uncut stock in the middle. Place wedges in each of the slots to spring the 2 halves of the stock equally. Then joint each side flat and finish by hand sawing the 2 cauls apart. The jointed sides will be tapered more gently by using thinner wedges.

Re: Springing a Smooth Symmetric Curve

#17

This forum is a group of GREAT members

don stephan

What an interesting, educational, and thoroughly enjoyable set of responses. Thank you all!

Guess I'm an oddball, but I don't like using biscuits in table tops. Even though I put a lot of effort into preparing the boards to be equally thick and perfectly flat and straight, without gluing cauls I get nagging offsets that I can live without. Spent hours planing out nagging offsets once. Now that I know excessive clamping force on too curved clamping cauls is undesirable, I need to find a better curve. Now I have several approaches to compare.

Re: Springing a Smooth Symmetric Curve

#18

What is the problem the cauls are to solve?

Bill Tindall, E.Tn.

I was presuming these were to be concave cauls to go between the clamps and the edges of the boards in the assembly so as to put more pressure on the ends . It is a reasonable thing to do if one is "clamp poor". In principle one only needs one clamp with "sprung cauls", to invent a description.

However, your followup comment suggests the cauls, maybe convex cauls, are to go on the faces of the assembly to smash the edges into alignment, yes? If this is the application, why do they need to be sprung? I would think that dead flat would be best, unless they are going across the middle of the assembly and not the ends as is usually done.

If I am gluing up a top I do it one board at a time, slowly working the edge into alignment as I snug the clamps, starting at the middle and working to the ends. Takes a bit more time but I can get near perfect edges this way. The "broad" I am gluing may be an assembly of boards that I have glued and run through the planer to plane them into submission .

While I have not used cauls for this operation the idea has merit. Yellow glue, that most commonly used, has little shear strength before it is fully cured. Hence , under stress, like from a caul, the edges will be able to slip into alignment.

People talk of beating an edge down. It never worked for me. The yellow glue will flow slowly under stress for a half hour but it is highly resistant to an impact force even 30 seconds after mating.

I never found biscuits useful. I get better alignment without them. (It might be my biscuit slot cutting skill)

Re: Springing a Smooth Symmetric Curve

#19

Re: Springing a Smooth Symmetric Curve

Paul Leuba

I have constructed gentle curves by using a piece of 1/8" hardboard about 1" wide and long enough to form the length of the curve. Drill a 1/8" hole in each end and thread a long piece of string through the hole at each end, tying a knot in one end. Then draw the hardboard into the a curve with the desired radius and tie a knot in the string on that end. This will hold the curved hardboard in the desired form while lines can be scribed on the work piece, etc.

Paul in Hunt Valley, MD

Re: Springing a Smooth Symmetric Curve

#20

Rick L

Re: The easy way to curve a caul *LINK*

Rick L

Been using cauls for 40 years. one of the best, least known about, and least expensive tool everyone should have in your shop. I make the curve by laying the caul on a good straight surface and put an 1/8" small thick scrap of wood under the middle of the stick Example 1/8" thick and 1" square. I Then clamps the ends of the caul down and I have my initial curve. I then scribe from the flat surface and when I release the clamps I have my curve drawn perfectly. I have an edge sander but you could do it a number of ways from disc sander to hand plane. I typically make a master caul and pattern route pairs. I found consistent size to be preferred. No one mentioned this technique. Very simple and yields a good even clamping pressure. I can't imagine buying these. There's some PT Barnum selling cauls made on his CNC claiming he has the perfect curve. Cauls have been around for hundreds of years. Interesting why someone doesn't see the value of cauls.

Cauls have many uses. For gluing panels they are such time savers. Milling stock carefully and using cauls I can glue up nice flat panels myself. I do it in one shot and have done 4' x 8' panels myself. I don't waste time using biscuits either. Make the cauls a convenient length. I like 4' but see what works for you.

Mike's blog doesn't go into much depth on the use of cauls but it's the basis of a simple veneer press or being able to glue up 24" wide cabinets. Such as a 24" wide and deep 72" tall cabinet with six shelves dadoed in the carcase and not using screws as you don't want to ruin the clean look of your joinery. It's an inexpensive way to extend the reach of clamps


Caul Tutorial

Re: Springing a Smooth Symmetric Curve

#21

Re: The knots

William Duffield

These are the knots I'fe found to work best for this application.

The first end should be a bowline. Everyone knows how to tie one of these. The rabbit (not rebate) comes out of the hole, goes around the tree,, and goes back down in the hole. Easy to tie, easy to untie, won't slip.

The second end should be a taut-line hitch (or tent hitch). Easy to tie, easy to adjust the size of the loop, holds tight when you let go of it, easy to untie. http://en.wikipedia.org/wiki/Taut-line_hitch

Re: Springing a Smooth Symmetric Curve

#22

Rick L

Re: convex, not concave...

Rick L

A board at a time!!!!! You do need to learn about how to use cauls.

Re: Springing a Smooth Symmetric Curve

#23

Rick L

Re: Cambered caul video... *LINK*

Rick L

Here's an easy way...


Cambered caul video