Table top wiggle

[Jerry Miner]

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.

Larry-- This is where I think you need an X-brace (like I said, keeping it aesthetically pleasing is the challenge)

Larry's table x-brace 1-20-13.JPG

[Larry Browning]

Larry-- This is where I think you need an X-brace (like I said, keeping it aesthetically pleasing is the challenge)

Larry's table x-brace 1-20-13.JPG

Oh, I see what you are saying. I also now understand your logic. However, I really see no way to do that and have it not look terrible. Tonight I am going to try Sam's idea with scrap pieces of wood. This should tell me if it will work or not. Maybe if I can just reduce the twisting a bit.
As it turns out, I have a store bought dining table that has a similar design. It features an X brace in the center. These legs are metal, but the design seems to be basically the same.

[Sam Murdoch]

In summary Larry - you have built a young and healthy person with good flexible joints - now it's time to age your creation - introduce some arthritis and sclerosis.

[Pat Barry]

I don't get it Larry. If its twisting, then something is loose. Find whats loose and tighten it up. I suspect its the joints, not the members themselves. You can't fix the problem if you don't identify the problem.

[Larry Browning]

I don't get it Larry. If its twisting, then something is loose. Find whats loose and tighten it up. I suspect its the joints, not the members themselves. You can't fix the problem if you don't identify the problem.

Pat,
I really think I have identified the source of the twisting. My 1st thought was also a joint or joints, but after careful observation I discovered it is not a joint at all, it is in the legs themselves. I believe the legs are flexing. The top is very large and heavy which allows a small amount of side to side pressure applied to the outside edge of the table to produce a huge amount of torque. I can actually see the legs flex. The solution needs to at least reduce that flexing. To me this is why some sort of a brace is going to help. Like Sam says, the whole leg assy needs to be connected together like a single unit and not 4 separate pieces. I am also afraid that Jerry M. is right and that I will need to connect the top of the leg to the bottom of the next leg, which will be completely unacceptable to the aesthetics.

[Larry Browning]

In my original post I had the idea of remaking the legs to be 12/4 instead of 8/4. It still seems to me by doing that, it would make the legs less flexible eliminating the twisting. Tell me again why this will not help?

The foot assy is made up of a half laped X which is 12/4. Each leg is attached to that, so I could make the legs the same width and still have it look decent.

[Pat Barry]

Starting over at the top. OK Larry, so the legs are twisting. Are you calling the legs the members being indicated by the arrow in your initial posting? If so, what is the cross-sectional dimension of those members?

[Larry Browning]

Yes, those are the things that are twisting. If I understand your question correctly, the legs are made from 8/4(1 13/16 inches I believe) cherry lumber. Since I have mentioned this before, I am not quite sure if that is what you mean by cross sectional dimension though.

[Larry Browning]

Ok, I have tried Sam's brace using scrap pieces clamped to the legs
photo.jpg

This has made a HUGE difference. There is still a small amount of wiggle, but I think it is acceptable. I also think that a bracket like the one I drew may add even more support. I am very encouraged. Next I am going to make a bracket out of plywood to see how that goes. I will keep you posted.

[Sam Murdoch]

Not to say that allllll that we have discussed so far is not viable, now having seen a photo and thinking about this for 3 days or so, I offer a few more variations of a possible solution.

These sketches are not to scale for your table but are just to clarify my ideas.

Option # 12) The bottom right one is a flat rail to place between the legs approximately where your trial jig is clamped. It can be 3/4" to 7/8" thick made up of two thinner pieces laminated together for extra rigidity then bandsawn to shape. In each case I encourage tenoning into the legs to create a shoulder and a much better glue joint. I show a center tenon but that could just as well be a 1/2 tenon. Only needs to dado the legs about 1/4" deep. The outside curve would be especially nice if it followed the angle/curve of the legs.

Option # 17) The second idea on the left - You have a table top apron now add an apron to the top of the base too, and so this vertical rail, also about 7/8" thick. These should be taller than your stretcher so that they overlap the leg to stretcher joint by at least 3/4". Should be kept to the outside (maybe set in only about 1/4" to the outside face of the legs) and create a full curved apron at the top of the leg assembly.

Though this represents the most work as you will likely need to laminate these curves (maybe you can saw from thick stock with the right lumber) I think this is the most elegant solution. Again, I encourage a tenon joint of some configuration. In this case it should be easy to dado your stretchers and that should be sufficient as long as you get an overlay to the legs and achieve a tight fitted glue joint between the legs.

Hope you are still having fun Larry



Screen shot 2013-01-22 at 10.05.17 AM.jpg

[Larry Browning]

At this point, I think I am going to pursue the 1 piece bracket like my drawing where it can straddle each leg. I will attempt to cut a slot into each leg for the bracket to fit so I can glue and screw to each leg. I can say it is a shelf for the cat to lay on.

About your new ideas:
It seems to me the key to the strength of this bracing is to link the legs in the center. Until I actually added the clamps in the center there was no change. My 1st attempt last week was to add a brace between each leg around the outside. I clamped those in place and it seemed to have very little or maybe even no effect. My understanding of your new idea is add bracing around the outside of the legs and not linking in the center. If the one piece bracket doesn't work out, I will pursue other ideas. But I think it is going to work.
Additionally, the actual table is a little different than the drawing. I did not make a circular apron, it is 12 straight pierces. I gave up on trying to make it circular. I think it looks fine this way.

This table has been quite a design adventure, and what I am ending up with is quite a bit different than the original idea. In some ways I think it is better.

[Pat Barry]

Larry, that picture was worth well more than a thousand words. After seeing, those members are massive and I just cannot see them actually flexing to cause your twist. I would be further convinced from this that the issue is the joinery. I would still look closely at the joinery but the cross-brace of one type or another will strengthen this significantly.

[Larry Browning]

Larry, that picture was worth well more than a thousand words. After seeing, those members are massive and I just cannot see them actually flexing to cause your twist. I would be further convinced from this that the issue is the joinery. I would still look closely at the joinery but the cross-brace of one type or another will strengthen this significantly.

I had my wife apply the twisting pressure while I studied the movement very closely last night. There may be some very slight movement in the joint where the leg attaches to the foot assembly, but it is very slight. I may end up gluing those as well. There is significant flex of the legs themselves in that they are twisting. The bracing helped significantly. Even though the legs look pretty hefty in the picture, when compared to the table top, they actually look small and delicate. The top is a full 72" round and weighs close to 100lbs. A small amount of pressure will produce a very large amount of torque.

[Pat Barry]

Larry, I trust your observation. How much movement do you have in the top? ie: 1/2 inch rotation, or more than that? Also, out of curiosity, what type of wood is it? Also, how are the bottom attachments to the base done? Dowels? If so, how many dowels per leg?

[Larry Browning]

Larry, I trust your observation. How much movement do you have in the top? ie: 1/2 inch rotation, or more than that? Also, out of curiosity, what type of wood is it? Also, how are the bottom attachments to the base done? Dowels? If so, how many dowels per leg?

Pat,
That is a good question. I have not measured the amount of rotation or even thought about doing so or how much it was. So this is just a guess. Without any of the bracing I would say it was rotating maybe as much as 1 to 2 inches(just a guess). I would say that after I added the bracing it reduced to maybe 1/2 inch (again just a guess). I feel that is pretty acceptable and I plan on moving forward with making a prototype brace from plywood.

I really do appreciate your interest in solving my problem, but I have already answered you other questions in previous posts. I will answer again. The wood is cherry, and the legs are attached to the foot assembly with 2 unglued dowels to help with alignment and 2 lag bolts in each leg. I want to avoid glue so it can be broken down for moving.