Thread: Destructive testing of Rietveld z-joints

Originally Posted by Rocker

Here are two pieces of 35x 40 jarrah, joined with a keyed mitre joint, glued with epoxy. The grain of the rock maple keys runs perpendicular to the mitre faces. I believe that this joint will be very strong. I shall be bringing it to the Bitingmidge gtg, to see whether anyone there can make the joint fail. The kerfs for the keys were cut with my tenoning jig.

Rocker

Have you got a flat-top saw blade, or otherwise how did you deal with the 'V' shape at the bottom of the kerfs?

Will you be able to measure torque during your Bitingmidge joint failure test? I guess the suitable quantity is (failure torque)/(mitre area) for a given stock thickness and timber/glue type.

Zenwood,

The kerfs were cut with an ordinary combination blade. But the process of paring the keys level with a chisel produces a slight tongue on the bottom of the triangular key, which hides the fact that the bottom of the kerf is V-shaped. It is not the sort of thing that a man on a galloping horse is likely to notice anyway

I doubt whether Midge has any scientific test facilities at his place; I was just going to invite people to belt the joint with a sledge hammer to see whether they can break it. I very much doubt that a vice could exert sufficient pressure to break the joint.

Rocker

Originally Posted by Rocker

Zenwood,
...the process of paring the keys level with a chisel produces a slight tongue on the bottom of the triangular key, which hides the fact that the bottom of the kerf is V-shaped. It is not the sort of thing that a man on a galloping horse is likely to notice anyway

Do you mean the tear-out-like phenomenon that pulls fibres from the key into the slight gaps around the 'V'? I thought it was just me that got that result

Here is a mitre joint reinforced by a 5/16" brass threaded rod, glued with epoxy into a 1/4" hole tapped with a 5/16" thread. I am sure the joint is very strong, but it looks tacky, because of the irregular outline of the oblique section of the brass threaded rod. I should have screwed the rod further in, and then plugged the hole with an ebony plug.

Rocker

How do you think the threaded mitre'll hold up compared to the keyed mitre?

The whole Z-shape concept has me fascinated but every time I consider the leverage involved I suffer brainfarts...

SC,

All will be revealed tomorrow at the Bitingmidge gtg, when I shall urge the members present to see if they can break the joints. My guess is that only a hydraulic jack will be able to break either joint. I doubt whether a vice would do it. I think the keyed joint would be stronger, and it could be made stronger still by drilling 1/8" holes on either side of the mitre and then gluing in unthreaded brass rod to pin the joint.

Rocker

Make sure some one takes a camera I want to see a few drunks with hammers in there hands Kinda gald I cant make it now sounds dangerous

Been following this thread with interest while on holidays. I'm thinking of trying to build an ultra-strong joint using concealed stainless steel wire rope as a tensioning member. Will post pics if & when it happens.

Hi Alex,

Welcome back.

I will be taking along tomorrow a gig that will hopefully put the trial joint into a position where we can apply known loads (body mass) at known location to check it's capacity. I believe we will reach failure - just a mattter of how many people will be standing on the plank loading the timber - wait for photos.

In relation to tension steel in timber I have two thoughts here, as I have spent the past 14 years designing tensioned concrete for building floors.

Firstly I am not sure how a short length of tensioned wire would perform as a small amount of creep in the timber will negate the tension.

Second you really need a reasonable section size to get the joint rigid even stressed as a small setion will extend more but also be flexural or like a rubber band not rigid in a joint. If the section is bigger it will only extend a small amount for a large load as such will reduce its tension a lot if the timber shortens only a small amount.

On the matter of strong and rigid joints really your best bet is to investigate using carbon or similar fibres that have a very high elastic modulous - that is that don't change shape much under load.

The joint Rocker has prepared if you had carbon fibre running vertically across the joint it would be awesome. Carbon fibre as a product set in resin is easy to work with in layers and drilling/shaping etc.

Cheers

AlexS,

I am not clear what advantage steel rope would have over threaded rod for reinforcing the joint. Look forward to seeing your pics.

Teejay,

Your carbon fibre idea sounds very interesting; I wonder whether it is available set in resin in thin sheets, which could be substituted for the rock maple keys in my joint. However. I think that actually the maple keys would not fail first - it would more likely be the glue attaching them to the jarrah, or the jarrah itself.

Anyway, it will be very interesting if you have devised a way to measure quantitatively the load that these joints can withstand before failing.

Rocker

I decided on a belt-and braces approach, and have now added a couple of pins cut from 1/8" brass rod to the keyed joint. I think this looks attractive and the joint must be very strong.

Rocker

Well the test has been conclusively resolved!!

Here are the pics of TeeJay's test rig and the whole testing catastrophe as posted on another thread!

Teejay's rig was a simple but very effective setup, the "vertical" member was screwed into place, and a plank clamped to the top of the horizontal bit.

Blokes were added to the thing till it broke!

The only flaw that I could see was that the length of the member tested was approximately two thirds the actual chair dimension, but the chair has also a surprising amount of deflection in each bit, which would probably more than make up for the difference in leverage.

In any case, the joints are strong enough for the job!

In the next post, I'll show the loading procedure!

cheers,

P

Rocker's Joint/TeeJay's test underway!

(Note to Southern readers: The blue stuff in the background is sky.)

Photos here are as follows

1) Rig with no load. Understandably, there is no deflection.

2) Now with two headless persons on board. (I suspect that they left their heads somewhere safe, due to an absence of appropriate headwear at the Home of the Biting Midge).

3) Deflection clearly visible with two headless persons on board, but no sign of any movement in the joint.

4) Four headless persons on board. RufflyRustic becomes concerned about what they may be thinking. Bitingmidge has seen a similar trick recently in China involving thirteen girls and a bicycle.

5) Failure!!! Well actually not quite, but we did hear a cracking sound and damaged the joint.

With four people (grown men) suspended on the joint, the plank was quite unstable, and considerable racking was induced. Note the outer parts of the joints only have cracked, and in my view a different glue would/should have survived longer than the timber. This is speculation, as the end grain joint is difficult to glue, and it may be on separating them that the timber did fail, but it looks suspiciously as though the glue failed.

Conclusion:

a) Balancing on a plank can be fun!

b) The Z chair joints each carry at best half the load each, so it would take eight headless people rocking sideways in unison to induce sufficient load to even cause damage to the joint.

c) Rocker's chair is a BEAUTY!! Much nicer in the flesh, and very comfortable. The Z configuration gives a nice amount of spring, and the joints are plenty strong enough: very cool!

Cheers,

P

Midge,

Fair go. The joint only started to fail when four people climbed aboard, giving a loading of about 300 kg. The glue started to gape slightly mainly as a result of the twisting force resulting from the fact that it was difficult to keep the plank level whilst the four testers climbed aboard. However, the joint is still intact. If I smear a bit more epoxy on it, no-one would know that it started to yield slightly.

Also, although the parts of the test joint are a bit shorter than the side rails of the chair, the centre of gravity of a person sitting on the chair would be about the same distance from the joint as the test board was in this experiment, so I think it was a valid test. I am very satisfied that the joint is amply strong enough for a dining chair.

Rocker