Table top wiggle

[Larry Browning]

I am building a large 6' round dining table or my own design. After initial assembly I discovered that the top had a wiggle. When I applied a small amount of side to side pressure it would twist a bit and seem sort of wobbly. I am thinking that the legs are twisting because they are not substantial enough. here is a drawing of the leg assy. Each leg is made from 8/4 material. I was thinking maybe they should be 12/4 to stop the twisting. Do you have any other suggestions to stop the twisting?

[Pat Barry]

That's a nice design Larry. How are the parts attached to each other, particularly the ones your arrow points at?

[Sam Murdoch]

Larry - nice looking base but lots of parts that are spring loaded by design.
2 potential issues come to mind.

1) There is not enough joinery tying all the base pieces together. They appear to be touching each other with no jointed connections - only secured at the ends. Are they just attached at the tops and bottoms with dowels? Also, are they sawn along the grain of the wood? I don't think thicker is needed but make them into a more integrated structure.

2) For a 72" dia. top I would plan on a minimum of a full 48" of support - and likely go out to within 7" or 8" to the edge of the top. This wold be the underside of the table top. It looks like your stretchers are long enough but again there appears to be minimal joinery. At the floor I would plan on no more than a 6" overhang. It looks like you have that relationship OK.

[Larry Browning]

I designed it so it could be broken down for moving, so those legs are bolted in place with lag bolts. The joint at the bottom also has 2 unglued dowels to help with stress and also 2 lag bolts. the top of the leg is also lag bolted the the table top cross supports. The legs are sawn pretty much along a vertical grain. I believe I angled the leg enough to maximize the width of the board.
Also, please remember this table already exists, I am looking to fix it, not redesign it.

Once again, my main question is whether or not a thicker leg would correct the twisting. I was also thinking that if the thicker leg would not correct, how about cross supports between the insides of the legs.

The table is otherwise very stable.

[Jerry Miner]

Larry--

That's a cool looking table. I think your issue has more to do with the fact that the legs are all separate and able to move independently than with their thickness. The real "fix" IMHO is some sort of bracing between the legs. The problem, though, is creating a brace that stiffens the table but also fits in aesthetically.

If you want to test the "thickness is not the issue" theory, you could cut some scraps to the shape of your leg and clamp a piece on each side of each leg--clamping hard, so it acts like a thicker leg--and check the wobble.

Then try a between-the-legs brace (rough-and-ready. This is only a test) and see what you get.

I think you'll find the brace is the answer. So what kind of brace? I don't know, maybe something in welded-steel scrollwork?

[Sam Murdoch]

Larry--

That's a cool looking table. I think your issue has more to do with the fact that the legs are all separate and able to move independently than with their thickness. The real "fix" IMHO is some sort of bracing between the legs. The problem, though, is creating a brace that stiffens the table but also fits in aesthetically.

If you want to test the "thickness is not the issue" theory, you could cut some scraps to the shape of your leg and clamp a piece on each side of each leg--clamping hard, so it acts like a thicker leg--and check the wobble.

Then try a between-the-legs brace (rough-and-ready. This is only a test) and see what you get.

I think you'll find the brace is the answer. So what kind of brace? I don't know, maybe something in welded-steel scrollwork?

This is a tough one after the fact Larry but I think Jerry and I are focusing on the same issue - more/better connection of the parts, i.e. bracing. My first thought was to build a couple of discs - one that could be fitted in the middle of the legs and then the other much bigger diameter to replace the top stretchers and to fit down and around the top of the leg sections so that they can't move independently. Don't know how to do this nicely or to guarantee that those would solve the problem.

Maybe a 4 legged spider type thing or an X. I think the ends of the X would need to straddle the curved pieces so that you really hold them.

Maybe turn a nice ball with notched/flat spots that can create good glue surface to the curves. The idea is to make those curved pieces into a whole.

Good luck with it.

[Larry Browning]

Hmm. I think I could make a X piece that would fit across the widest part of the legs. I am not so crazy about have it straddle each leg though. I still want to be able to break it down for moving. How about using a pocket screw to attach the X to each leg? I would not want to glue it either.
I know I am not making this easy. I want to try to keep the original look as much as possible.

[Sam Murdoch]

Sorry Larry - I don't think adding more sticks will do it. Can't you add to that top assembly so that it can come apart as one piece? I could be wrong but I'm pretty certain that the solution to the wiggle is to lock those pieces together - at least with joinery if not also with glue.

[Larry Browning]

Sorry Larry - I don't think adding more sticks will do it. Can't you add to that top assembly so that it can come apart as one piece? I could be wrong but I'm pretty certain that the solution to the wiggle is to lock those pieces together - at least with joinery if not also with glue.

Sam, I never claimed to be an engineer Ok, I guess the 4 legs could be a single thing. So keeping mind my previous comment. Why would not the X brace lock it all together?

[Sam Murdoch]

I'm not an engineer either - mostly a mock it up type of woodworker but after many many years and quite a few wobbly projects I'm developing an instinct .

The X brace wil lock it together if you have good bearing and secure fastenings. I just don't think that pocket screws will do it - after all the effort. At the least, I think you will need to notch the braces in - add glue and a screw or a through dowel too. I don't know why it needs to come apart as more than just the top and the base, but you will achieve the most dimensional stability if your base is an integrated whole not just a bunch of pieces connected at the points. I am now hoping that others will join this conversation. Are you ready to shoot this messenger?

[Larry Browning]

shoot you? Heck, I should be hugging you! I think I have conceded to the need for the legs to be glued. At least I can separate the legs from the feet. I will look into how I can tie the together. Maybe a notch in each leg. I am traveling this morning, so will not post until this afternoon.

[Larry Browning]

I am pretty skechup challenged, but I tried to draw a support piece that I could use. My wife thiks this will look better than a simple X shape. I also think this allows for maximum support. If this looks good to you, where should it go vertically on the legs? I was thinking about midway. Should I try to notch out the leg as well? A notch would add difficulty to making it fit properly, but if it will add a lot more strength I will attempt it.
support.jpg

[Jerry Miner]

Larry---

I don't see this as the right fix. The legs are kept apart from each other at a fixed distance at the top and at the bottom. Your issue, as I see it, is the "twisting" movement allowed with this construction. Adding another device to fix the distance between legs at or near the center doesn't change things much. Your new bracket would simply rotate as the table twists. (And if it's made of wood, there will be short grain that would be easily broken off with a little stress).

What you need is something to resist the movement of the top of the leg relative to the bottom of the adjacent leg---like an X-brace. I'm not an engineer either, but I've built lots of stuff. Maybe you could clamp on something temporary to test the concept?

[Larry Browning]

Larry---

I don't see this as the right fix. The legs are kept apart from each other at a fixed distance at the top and at the bottom. Your issue, as I see it, is the "twisting" movement allowed with this construction. Adding another device to fix the distance between legs at or near the center doesn't change things much. Your new bracket would simply rotate as the table twists. (And if it's made of wood, there will be short grain that would be easily broken off with a little stress).

What you need is something to resist the movement of the top of the leg relative to the bottom of the adjacent leg---like an X-brace. I'm not an engineer either, but I've built lots of stuff. Maybe you could clamp on something temporary to test the concept?

Jerry,
I am not following the logic that an X brace would work and this one would not. This would perform the same function as an X brace, except that it is more substantial. You could easily shape the bracket into a X shape by removing more material.
However, I have been playing around with making a template for this brace and have about come to the conclusion that getting it to fit tightly (and still look good) may be beyond my abilities. I think making an X brace that fits into notches in the leg may be within my abilities. I would make a half lap joint at the X. Also I am afraid the bracket would be very large. After doing some measuring it looks like that bracket would be almost 24" square.

[Sam Murdoch]

I'm still trying to guess my way through this one too. I agree with Jerry that before getting too involved clamping some strategically placed cleats could tell you a lot. Move them around until you get the best result. Here is a simple construction for testing what you have drawn Larry -

Screen shot 2013-01-20 at 4.59.52 PM.jpgThese are just 1 x sticks of wood that you could clamp to the sides of your legs. I propose adding a center block so that you can move this up or down the legs to find if any position mitigates the wiggle.

My thought is that something like you drew Larry could be attached at the very top to the stretchers. If you fit the notches tightly to the legs and they are glued you will at least get all the legs to be secure as a unit. THEN, perhaps add 4 glue blocks at the bottom of the legs fitted in between each pair of legs - these too glued in place - and again all to making the legs one assembly rather than 4 pieces.

It might take another set of glue blocks between the knuckles at the top of the bottom legs too but that depends on how rugged those joints are. This is a very difficult way to go about building anything. Hope something works for you.