How do I stiffen up this metal table?

[Animag771]

TL;DR Summary

My DIY ironing board-like table is springy and wobbles when touched. Is there any way to stiffen the pipe to beef it up enough to keep it from flexing?

So here's what's going on... I've gone and made an attempt at building a height adjustable table.

Originally I was going to use stainless steel for the pipe, but after finding out that it would more than quadruple the price, I ended up with SCH40 Aluminum. I used three 48" long, 1" ID pipe for the legs which criss-cross and attached them to two 36" long, 1.25" ID feet.

I put it all together and mocked up a table top and everything works great... Then I added some weight to it do simulate the weight of the actual top I plan to use (70lbs) and the pipes start to flex and it makes the whole thing kind of springy. If you set a glass of water on top you can see the water moving back and forth as the table bounces (jiggles?) until it finally settles and stops.

Is there any way to stiffen this thing up enough to make it more solid? I thought about drilling into the long pipes and filling them with concrete, but what happens when that starts to crack? I also thought of cutting the welds and inserting 1" solid steel rod (or flat steel oriented vertically) into the pipes. Or there's always the option to add pipe.

Would any of these options work? Any other ideas? Or am I out $100 and need to redo this with larger diameter steel?

 

[phinds]

you could put flat rods into the length of the pipe, oriented vertically. A little JB Weld at both ends would keep them from slipping out of position.

You should be able to get those at any hardware store

 

[hutchphd]

You clearly never jumped off an aluminum springboard at the pool...they are indeed very "springy". I also once had a wonderful set of aluminum x-country skis.!
I believe the primary problem is that the natural resonance frequency of the system (top plus springy legs) is a few Hz...just what you don't want. You could stiffen the legs or lighten the top (increasing the frequency). You could put some kind of shock absorber on each end to dampen the oscillations. If you tied both the top and bottom rigidly horizontally it might work, but the adjustment would be harder.
I like the shock absorbers...

 

[jrmichler]

You could add a series of notches underneath the table top. Each top cross bar would seat into a notch. That anchors the top bars, which will make the entire leg system stiffer. You can test the idea with a couple of C-clamps. It would still be height adjustable, just in discrete steps.

Then add rubber pads to the bottom cross bars so they don't slide on the floor. That should also help stiffen it. Crutch tips are cheap, available in several sizes, and might work.

If after doing all that, the top rocks too much in the sideways direction (in and out of the paper in your photos), then the rocking is caused by the (lack of) torsional stiffness in the legs. You then have several options:

1) Start over, but with two sets of legs welded near the ends of the cross bars.

2) Start over, but make the legs from larger diameter tubing.

3) Start over, but use steel tubing.

A general rule when making something from tubing, and where the tubing needs to be stiff in bending or twisting is that tube diameter is more important than the tube wall thickness. Also, steel tube is three times stiffer than aluminum tube of the same dimensions.

A good source of low cost thin wall steel tube is electrical thin wall conduit, also known as Electrical Metallic Tubing (EMT). EMT is galvanized, so will not rust (except for the welds). You can weld EMT, but be very careful to not inhale the fumes, or you will learn the hard way about metal fume fever (search the term). Metal fume fever is a very miserable way to spend a few days. I speak from experience here.

 

[Nik_2213]

IMHO, you have two problems.

First, the small diameter of the pivot which, combined with the halves' long leverage, is seriously 'un-stiff'.

Second, the lengths of the slim frame will so make this bouncy. I reckon you need vertical or near-vertical struts between the floor and board horizontals. But how to make these adjustable ? Nest tubes to be telescopic, per walking stick or elbow crutch, fit spring 'cotter' clip through matching holes...

disclaimer: some years ago, I used this 'telescopic' trick to thwart 'sadly declining' MIL who persisted in releasing the tubular framed safety sides of her home-hospital bed and falling out. Night after night after night...
I devised a variation of 'lift shutters up into deep slot, lower into shallow'. A length of (IIRC) 1½" plastic drain-pipe had ample flex to easily feed up over the tubular frame's head-end leg, stand loosely on the floor. Readily removed without tools by MIL's rota of carers, it could be kicked clear in extremis for eg CPR access. A couple of finger-widths below the frame's raised leg, I drilled an M6 / ¼" clearance hole across the pipe, fitted nut & bolt. If we found the frame resting on this with MIL still safely in bed, we knew she'd tried to go 'walkabout' again...

 

[Mark44]

First, the small diameter of the pivot which, combined with the halves' long leverage, is seriously 'un-stiff'.

This was the first thing I noticed... Any slop in the fit at the pivot would increase the effect of twist around the axis of the pivot.

 

[256bits]

If you set a glass of water on top you can see the water moving back and forth as the table bounces (jiggles?) until it finally settles and stops.

Several modes of vibration for the top
Translational - the top moves
1.front to back
2 left to right
3 up and down
Rotational - the top has a twist
4 in the horizontal plane ie about a vertical axis ( left end in, right end back )
5 in the vertical plane into the page ie about an axis front to back ( right end drops, left end rises )
6 in the vertical plane of the page ie about an axis left to right ( front end rises, back end drops )
Diagonal modes
such as the back right corner will move up, the near left will move down.

Just by inspection, it looks that all modes are possible, some more predominant than others, either due to rigidity, or lack thereof from the members, and torsional rigidity or lack thereof mainly about the pivot ( being pointlike ).

If you had ever noticed some of the collapsible chairs have the legs diagonally placed, which kind of alleviates the translational modes left/right, front/back. You could achieve something similar by extending the pivot out a good several inches out in both directions both front and back and attaching a member from the extension to each corner of the crossbar for the top and the bottom. That should alleviate some of the torsional mode of twist also.

For up/down vibrational motion the legs should be stiffened either internally as mentioned, or externally with a flat, or another round. Not sure if that will totally fix the problem.

Before adjusting your table, it might be a good idea to buy some wooden dowel, maybe as small as 1/8 inch ( as that is easier to cut and hot glue and undo ), make a model and try out adding additional pieces as mentioned. Note that the vibrations will not be as with metal, as wood has better damping characteristics.

A tight pivot makes the whole assembly attempt to act as a rigid body as possible. Some slop there will act as damping for the vibrations, but then the static stability is compromised somewhat, so there is a trade off. Rubber bushing maybe.

 

[Nik_2213]

FWIW, we had problems with a similar design of shop-bought ironing board, albeit from an unexpected direction.

One of our tabby cats took to charging into laundry room, taking a flying leap onto the ironing board. This was okay when she was a cute kitty, less so when, as a stocky yearling, each claws-arrested landing toppled it backwards, ironing, iron and all.

I still had some of that plastic drain pipe, so slid a length over each of the rear-facing legs' rubber bumpers to create stabilising 'out-riggers'...

 

[JBA]

A principle source of the springiness is the sections of the cross legs below the pivot point, particularly the single tube crossmember. Above the pivot there is a rigid triangle consisting of the leg sections and table top with the stop strips on the table bottom that places those leg sections in compression rather then bending under loading. The lack of similar stops for the bottom sections places those tubes in pure bending that is exacerbated by the flexibility of aluminum. One modification that might give some improvement would be to connect the bottoms of the opposite legs to hold them from sliding apart and therefore stopping the bending loading flexing just as the table stop strips do for top leg sections. In order to retain the assembly folding capability you might use a steel cable attached between the legs; but, be sure not to use too small a diameter cable or it will stretch under load and defeat its purpose. Note: the cable connections should be in the center of each of the bottom tubes where the crossmember tubes connect for the best effect.

 

[Rive]

To be honest I think this design is very hard to save. I would try something like this instead, to provide adequate support on multiple directions with short (!) components/bars and also, I would switch to square hollow sections instead of round pipes.

 

[Nik_2213]

Um, vertical props ??

 

[Animag771]

WOW! Lots of suggestions. Thanks I'm still reading through then and will give replies later today when I have some time, but I just realized I forgot to include a link in my original post.

Here is the idea for the table's operation and design inspiration.

I'll also be installing a locking gas spring attached to the table bottom and one leg, so the table can be raised and lowered to any position. The idea is to have it be adjustable from 17" to 36" so it can be used as a coffee table, dining table and kitchen island. The top will be a 3' x 5' x 1.5" walnut butcher block. I live in a pretty "tiny house", everything has multiple uses.

I've noticed some members have suggested vertical supports, but I REALLY don't want to do that. If I were going to use vertical supports, I may as well just scrap this and build a traditional 4 legged table and use gas springs or telescoping legs. Would it work? Sure, but it's just not the look I'm trying for.

Thank you everyone for your replies, I'll be back later to give more thorough responses to your suggestions.

 

[anorlunda]

I'm not sure how much it would help, but straps between those horizontal arms that pull tight when loaded are worth a try. In fact, you could test it for just pennies in cost and minutes of effort with duct tape. If that doesn't help, then you haven't lost much.

 

[JBA]

Now that I see the design you are following I also see that you cheated and instead of using two tubes on both diagonals as shown on the original you only one tube on one of your diagonals, not a good idea because that makes that diagonal only 1/2 as stiff and twice as flexible in bending than the two tube diagonal.

 

[DannoXYZ]

Couple of differences I noticed between your design and the demo video:

1. they used larger diameter steel tubing vs. your aluminium. Since stiffness goes up to ^4 power of diameter along with 3x stiffer steel vs. alloy, their design (if identical) would be easily 128x stiffer than yours.

2. Table-top is attached to top legs creating triangulated section. Not sure how they did it, and they were careful not to load the table. Look into how ironing-boards lock the upper legs. If you can settle on several distinct heights rather than continuously variable, you can incorporate "claws" that latch to upper legs and form rigid triangle. This will go long way towards improving rigidity.

3. pivot. As mentioned, any looseness at pivot is magnified at end-points. Look at using larger-diameter sleeves in the tube with tight fit to attachment bolt. Brass or nylon-lined with tight-fit perhaps. Or weld in mounts for bearings. Press-in roller or dual-row tapered needle-bearings can be adjusted to remove any looseness at pivot while allowing rotation.

 

[Animag771]

Sorry for the late response, it's been a busy weekend. So, it seems that the general consensus is that this is scrap and I need to redo it with steel and space the legs (cross-members?) farther apart?

Ok so for the steel, how does 1" SCH40 steel pipe sounds? I was at the hardware store yesterday, so I went to the plumbing section and grabbed a 48" piece of 1" galvanized pipe, propped it up against the shelving beam (at an angle) and attempted to bend it by putting my foot in the center and leaning my weight onto it. Not a very scientific test, but there seemed to be no visible give. Think it'll do?

As far as the rest of the table goes... The feet (parallel to the floor) are 3'. I originally had the middle leg centered and spaced the outer legs 1.5" off center on both sides. That obviously didn't work out. So how about instead of using only 3 supports, I use 4 (like in the video) and space them 12" off of the center of the feet?

Lastly, the pivot. With the additional spacing on the legs, would it be best to use 2 pivots? Or would I be better off using a long rod that runs all the way through? Also just an FYI, my existing pivot is a 3/8" steel rod which fits fairly snug inside of a 1/2" aluminum pipe and I've used steel spacers pressed over the 1/2" pipe to keep things from having play. I've attached a close-up picture of the pivot point.

 

[Animag771]

Any opinions or ideas on my previous post?

 

[Rive]

With the additional spacing on the legs, would it be best to use 2 pivots?

Check on the design you provided as a reference previously: it is not just the number of pivots, but the distance between them...

 

[Animag771]

Check on the design you provided as a reference previously: it is not just the number of pivots, but the distance between them...

That's the reason I said that I'd be increasing the spacing to 12" off center for the legs. It's hard to tell from the video, but that seems to be close to what they've done. They also used a single rod for they're pivots, but I was just curious if that actually makes a difference vs using 2 separate pivots so there wasn't a long rod going between the sets of legs.

 

[JBA]

The companies that sell this type of table have been through the learning process of what design works best so follow their lead and take advantage of their experience.

 

[essenmein]

FYI don't put concrete inside the ally tube, concrete is caustic and will cause all sorts of corrosion issues with the Aluminium.

As far as the structural issues, I would agree the pivot looks small, and to be honest I think the alum tube is on the small side as well. The single tube on one side won't help. As mentioned already, if that was steel tube with the same wall thickness it'd be ~2.2-3x stiffer depending on alloys, conversely if something was intended to be steel, you need about 2.2-3x the material if doing the same thing with alum (rough guide anyway).

Then can't quite tell, but the "feet" look like they have bent due to the weld contracting, if that's not an optical illusion then that might significantly contribute to the rocking and general feeling of instability.

 

[Animag771]

FYI don't put concrete inside the ally tube, concrete is caustic and will cause all sorts of corrosion issues with the Aluminium.

As far as the structural issues, I would agree the pivot looks small, and to be honest I think the alum tube is on the small side as well. The single tube on one side won't help. As mentioned already, if that was steel tube with the same wall thickness it'd be ~2.2-3x stiffer depending on alloys, conversely if something was intended to be steel, you need about 2.2-3x the material if doing the same thing with alum (rough guide anyway).

Then can't quite tell, but the "feet" look like they have bent due to the weld contracting, if that's not an optical illusion then that might significantly contribute to the rocking and general feeling of instability.

Good points, didn't think about the caustic nature of concrete. The feet aren't bent, it's just the picture playing tricks.

What does everyone mean by the pivot being small? Are you all saying that it needs to be larger in diameter? Also, why?

Thanks for the responses from everyone. I'm trying to be patient and hold off on buying the steel to redo this until I've got a solid game plan. I don't want to repeat the same mistakes, but I'm also tired of eating dinner off of a wooden box :P

 

[essenmein]

Good points, didn't think about the caustic nature of concrete. The feet aren't bent, it's just the picture playing tricks.

What does everyone mean by the pivot being small? Are you all saying that it needs to be larger in diameter? Also, why?

Thanks for the responses from everyone. I'm trying to be patient and hold off on buying the steel to redo this until I've got a solid game plan. I don't want to repeat the same mistakes, but I'm also tired of eating dinner off of a wooden box :P

The pivot is not just pivoting, it has to provide stiffness in basically all axis here, so you have some small diameter steel rod trying to be bent by quite a long lever. Then the small pivot would put larger stress concentration on where the tube meets the pivot, and ally being soft as butter, will over time, probably not even that long "ovalize" the hole, which then rocks more, causes more ovalization and its a down hill spiral from there.

You could take the same material as the little sleeves you made for the pivot, drill the tubes out the OD of the sleeve, cut length to dia of tube and weld into the tube, now the pivot has the whole length of sleeve to work against rather than holes in the wall of the tube.

 

[Animag771]

The pivot is not just pivoting, it has to provide stiffness in basically all axis here, so you have some small diameter steel rod trying to be bent by quite a long lever. Then the small pivot would put larger stress concentration on where the tube meets the pivot, and ally being soft as butter, will over time, probably not even that long "ovalize" the hole, which then rocks more, causes more ovalization and its a down hill spiral from there.

You could take the same material as the little sleeves you made for the pivot, drill the tubes out the OD of the sleeve, cut length to dia of tube and weld into the tube, now the pivot has the whole length of sleeve to work against rather than holes in the wall of the tube.

Awesome! Thanks for the quick response. I really appreciate that you not only gave me the "why", but also provided a way to improve on the current design. I'll be sure to improve the pivot on the re-disigned steel one.