Got an email from a potential client. He’s looking for a bench, all wood, about 60x18x20, with the top being sort of a lipped tray. The tray is to contain a bunch of loose ball bearings, on which the free-form seats are to move.
I have a few concerns. One is that the ball bearings will dent the top surface of the tray. Any thoughts on wood/finish for this? I’m wondering if steel sheet, or aluminum, or maybe even plastic laminate, might be a better choice. Concerned about wear, esp. with p-lam, though. Don’t know if something like teak, oak, or maple would hold up. (I’m a carpenter, not a furniture builder.) I’d hate to build this and then have depressions in either the finish or its substrate.
The other is that if he fills the tray with bearings, he won’t get the kind of slide action he seems to be looking for, ’cause they won’t move much. I’ve suggested using those infeed type multi-directional rollers mounted in races, mounted either under the seats or on the top of the tray.
I’d welcome your thoughts and suggestions.
Also, where would I get those bearing/race combos without buying and cannibalizing a bunch of roller supports?
Replies
Tom,
Does your client want whatever is placed on the bench to roll around on ball bearings? If so, you want a surface as dent-resistant as possible. Otherwise, the ball bearing will embed themselves into the benchtop and render themselves useless. I don't know if these would suit your client's needs, but have a look at "transfer balls".
Chris @ www.flairwoodwork.spaces.live.com
- Success is not the key to happines. Happiness is the key to success. If you love what you are doing, you will be successful. - Albert Schweitzer
What is he actually trying to accomplish? That is, what is the function that this is supposed to fulfill, without getting into the details of how that function might be implemented?
You basically don't have much chance of making this work with a wooden surface without marring it. Have you ever looked closely at a wooden floor in a public building? You can easily see the marks from women's high heels, and those distribute the load over a larger surface than a ball bearing does. Ipé has the highest indentation resistance of any commercially-available wood. I just did a quick back-of-the-envelope calculation and determined that even with ipé, you'd get tracks, although they wouldn't be very deep (less than 1/64").
And it's completely unworkable from a mechanical point of view, too. If the tray is filled with ball bearings, they won't have room to roll, as you note. And if there is enough space so that they have room to roll, then they'll just roll around until they reach the edge of the seat, and then get squeezed out; in no time at all, the seat will be laying flat on the tray.
-Steve
His email described "several amorphous wood seats on ball bearings, making the seats very 'slidey' ".
I suspect he wants a conversation piece/sculpture as well as seating.... frankly, I doubt that, if it worked the way he envisions it, it would be very comfortable. I don't think I could sit on something like that for more than a couple of minutes.
Regarding the denting and tracking of the bearings into the wood, I suggested steel, aluminum, or maybe p-lam as a bearing surface. I know it doesn't take a lot to dent, say, an oak floor, much less pine or fir. Or any plywood I've worked with.
And even 1/64" indentation in ipe, while not sounding like a lot, would be very visible given the the right (or wrong) lighting angle.
After my initial thought of, "wow, that's kind of a cool idea", It came to me that a tray full of bearings wouldn't produce any sort of motion in the seats.
Actually once all the surface is dented, it will be smooth, the denting is because the wood fibers have been compressed. Theoretically if all the fiber were to be compressed evenly, then the surface would be smooth. But, getting it all evenly compressed would be the problem.
You might be able to laminate a veneer over the top of a steel substrate and get a smooth durable surface with the look of wood.
DGreen's solution of an air flotation system will get the desired action, but not the look.
Have flat bottom seats and float them on an air bearing.
Giving money and power to government is like giving whiskey and car keys to teenage boys.
-- P.J. O'Rourke
Seems like you would need the bearing rollers attached to the bottom of the seat rather thab loose in the tray. If you had enough of the bearings on the seat and a very hard wood you might get enough combination of weight distribution and hardness of surface to make it work. Charge a lot of money for this one.
Sounds slippery and dangerous, that is if your customers idea would even work. Though, I guess there would be some type of stops built in. As far as denting the wood goes, it reminds me of that magic trick where someond lays on a bed of nails. The quanitiy of nails is supposed to make it seem more dangerous when really it helps distribute the load on the epidermous to the point where it is tolerable. The more bearings the less the wood dent, and it sounds like not a really visible part of the bench. But I think that you can find a better design. You should be paid by the hour to make different kinds of mechanical mockups, including the bearing thing. This is not a simple project and if the customer thinks it should be or they are not willing to pay you to figure it out, get out now!
Brian
Not sure about your functional requirements of a "recessed top" created by adding a lip around the perimeter. I'd go with something like UHMW as a top (same as those furniture movers that fit under heavy furniture to slide it on a floor. A thin sheet provides a slick surface and is easily replaced.
Sounds like this is an assembly table not a workbench.
Greg
•••••••
Exo 35:30-35
Don't know if your customer would consider it, but probably the best way to get the motion you're describing is to build the seats with a bottom of UHMW polypropylene (similar friction properties to teflon, but is a lot more resiliant), and put a smooth melamine laminate on the bench surface. Believe me, you'd slide like crazy with that combination (maybe slide too much).
In reference to the cost estimate - several posters are right. Get him to agree to pay by the hour for the prototyping necessary, including building a couple of small models, and have a lawyer draft a release that says that the design is entirely the client's, and the materials and workmanship were approved by him before it's put into service.
Then talk to your insurance agent and buy something called an "Umbrella Liability Policy" for no less than $3 million. It'll cost you about $200/year, but it means that your possessions are (possibly) a small fraction of the insurance policy, and the plaintiff's lawyers will go after the insurance company, not you.
I've been carrying a general liability policy for years, $1M/$2M aggregate. Been GCing for years, as well as carpentry/trim subbing. So I think that's covered, but I appreciate the thought.
He said he'd send me more info tonight, I'll post whatever might be useful.
Hard to dent Purpleheart!
This forum post is now archived. Commenting has been disabled