New WoodChat!
Home Articles Forums Gallery Marketplace Videos

WoodCentral Forums

Knowledge You Can Build On — Est. 1998

Nearly 100 Exotic Wood Species, Direct-Imported Since 1972. Shop Slabs & Turning Stock.

Sprung joints

Posts

Re: Sprung joints

#51

Pocket protectors

wilbur

I don't own a pocket protector. I buy pens that don't leak. ;)

Re: Sprung joints

#52

But that's not the whole point

Bill Houghton, Sebastopol, CA

A good pocket protector maintains the shape of the pocket, too, so it won't sag; and it makes setting the pens aside at night far more convenient (just one thing to remove).

I find myself wondering what wood species would make a veneer strong enough yet flexible enough for a good woodworker's pocket protector.

Re: Sprung joints

#53

Chris Scholz

Re: Just a ballpark figure...

Chris Scholz

Elasticity, plasticity, viscosity, stiffness, bending, 2nd moment of inertia, none of this is needed (or even useful) for this simple problem.

Here is how to do it:

1) Draw a free body diagram

FBG of Sprung Joint

E.g. F1 through F4 are clamping forces, F5 and F6 are joint forces.

2) Add all forces, they must balance (Newton's 3rd law*)

Here 0 = F1 + F2 + F3 + F4 - F5 - F6.

If all clamping forces are the same and the joint forces are symmetric, ... you do the math.

3) Divide the joint force by the contact area (indicated in red). This is the pressure.

* In Sir Isaac Newton's honor forces have units of N, personally I prefer dyne/furlong.

Re: Sprung joints

#54

But...

John in NM

It is desirable to exert pressure along the entire sprung joint, which is where all that bending complexity comes in. It would probably be adequate to assume a very shallow triangular profile rather than a curve, more or less how you've drawn it.

You're analysis is exactly what I was thinking of for a simple straight joint though, except I'd imagined the force distributed over the entire joint face. I chose force in pounds due to pressure for this application being expressed in psi rather than kPa, that's part of the reason I ended up mixing units up in the initial reply.... yes, we have SI for a good reason :D

Re: Sprung joints

#55

Re: joint testing

Pam Niedermayer - Austin, TX

Thanks, Bill, I thought that might be what you were talking about. I find it somewhat odd that whenever I test a glued joint, usually Titebond I or II are the only ones I test, the joint does fail in the wood, but the wood just on either side of the joint. And I have to wonder, not having applied all that much pressure to test it, usually just hand pressure, why this is. It seems to break wood in places that would never have broken if not glued, say if you had a piece of wood as wide as the glued piece. So for years I've wondered whether the very act of gluing somehow weakens the wood on either side of the glue.

Pam

Re: Sprung joints

#56

Re: But that's not the whole point

Pam Niedermayer - Austin, TX

Since we all use computers instead of pens and pencils these days, I think the days of the pocket protector are long gone. I'm always surprised when I have to sign a document and have to dig deep to remember how to write.

Pam

Re: Sprung joints

#57

Re: Just a ballpark figure...

Adam Cherubini

Nice picture Chris. My concern was that if you tweaked each clamp as Franklin recommends, you'd get more pressure at the ends exactly as you have shown. I wondered if the pressure could be enough to starve the ends of glue where we most want a tension capable bond.

Not to be stubborn, but I'm not convinced this can't happen. I could well imagine bondline pressure well in excess of 1000psi. I could imagine 3000psi pressures. I know that sounds like a lot of load, but if you've ever crushed wood with a clamp, you've likely produced 3000+psi.

For John, of course the answer to the pressure distribution in the joint is non-linear. The only good way to figure it out is with a finite element model like NASTRAN or PATRAN. I don't believe it can be done by hand practically.

Adam

Re: Sprung joints

#58

Chris Scholz

Zero force --> zero pressor -->

Chris Scholz

--> so the profile that is not in contact at the joint is irrelevant what joint pressure is concerned.

Re: Sprung joints

#59

Chris Scholz

Yes, you can make it as complicated as you want

Chris Scholz

but the free body diagram (really Newton's laws) is still valid and will get you 95% there. Finite element methods (boundary element methods, finite difference methods or pretty much any computer aided analysis) have the nasty property that if you don't understand the problem the results that your computer spits out will be useless.

Re: Sprung joints

#60

Here's what I was thinking...

John in NM

Take Chris's drawing, imagine there are 5 equally spaced clamps providing the load on a 5' long joint. Lets give each clamp a 2 square inch face. Tighten each clamp until it starts to crush the wood. Lets say the wood in question crushes at 3000 psi. Each clamp is therefore exerting 6000 pounds of force (I'm not converting it to Newtons! :D), for a total of 30,000 pounds force on the joint.

Case 1). Straight joint 60" long by 3/4" wide is 45 square inches. Divide the 30,000 force by 45 square inches and you get 666.67 psi on the joint. To reach 1,000 psi we'd have to add another 1.5 clamps :D, so its getting to be a crowded glue up, but not too unreasonable.

Case 2). Sprung joint. For simplicity, we'll just assume its short bits of straight joint at the end and a gap in between. Though we'd never tolerate a joint sprung so much as to gap, we're just trying to figure out if Franklin's 1,000 psi is a reasonable stopping point. So lets assume that the straight bits at the ends are each 15% of the total length, giving us a 15 square inch joint area. With the same 5 clamps on the 5' glue up we then get 2,000 psi at the ends.

Based mostly on case 2, I'd agree with you that it needs to be tested at a higher pressure. I'd be willing to bet that Franklin's engineers didn't consider a sprung joint, falling into the "you don't need to spring them, darn it!" camp. I'd guess that if you really modeled it accurately, the pressure at the ends would be somewhat below 2,000 psi on a reasonably sprung and fully closed joint.

Re: Sprung joints

#61

Re: But...

Sgian Dubh

The following is a general comment and not addressed to anyone in particular, even though I clicked on the reply button in John's post.

I think some people are getting overly excited about how much pressure is required to close up an edge sprung joint. No more pressure should be required to close the joint at the centre of the spring than that used to hold something like a large city telephone directory spine down between the thumb and fingers of one hand.

If you need more pressure than about this amount you need to redo the joint. The additional cramping force required at the centre point over that used at the ends of the joint is therefore minimal. Slainte.