Bed design help, mattress suspension?

[Brett Viren]

Hello,

I'm designing a bed and am looking for some advice on how to deal with suspending a king size futon mattress with out resorting to using a fifth center leg.

The basic support structure is a box the same width and length as an "eastern" king and ~8" high. I plan on using 3 internal, lengthwise beams in addition to the 2 side rails of the box to support many slats running crosswise. The box will be constructed using cross halving joints. It will be topped with a large "picture frame" and will rest on

4 legs which, finally, rest on casters. The lengthwise beams will attach to the head/foot boards with through and wedged M&T and the slats will attached by sliding into mortises dadoed into the side rails before the "picture frame" is laid down. The slats will be at least pinned to keep them from working loose but I'm also thinking to wedge them to provide some tensioning.

So, some questions:

What wood and dimensions can people suggest for the beams and slats? Any rule on how large of a gap between the slats can be tolerated before being felt though the futon? Should I use more than 3 beams? Could less be used and still provide the support?

Is the tensioning of the slats a good idea? Will the wood stretch over time (I would guess so) and if so, can this be chased by periodically pounding in the wedges or replacing them with larger ones? Are some woods better suited for being tensioned?

Thanks for any comments!

-Brett.

[Andy Dingley]

Ash, maybe hickory in the USA. They should be straight grained, split coppiced timber is good (and cheap) if you can get it.

I'd make the slats about 4" across, but this isn't critical. The gaps aren't important for anything else other than allowing the slats to move freely, so bunch them up tight if you're worried about feeling them. I'm not a lover of futons, I prefer latex foam. One doesn't mid gaps, but will let you feel them. Latex itself doesn't like big gaps and will wear badly. I use 1/2", maybe 1" max.

I'd just use one, maybe two - although past experience of working on beds for Americans has been that they need a ridiculous amount of mechanical strength and still fail. For normal size mattresses on an English bed, I don't use any internal beams and just run the slats full-width.

Centre beam design on beds is tricky, if you aren't going to make the beam telegraph itself through. I'd use a single beam for a bed that was effectively a twin, a pair of beams for those couples who both sleep in the centre.

Typical joints if you do go for a central beam are to either run the slats one-piece and full-width, or to narrow the ends of each slat down to about 1/2 width symmetrically and interleave them (the slats are staggered). This also works for Orthodox Jewish double beds, where they have to be separable (the "beam" is the doubled side frame of the two halves).

Not in my book. I mount them as loosely as possible, to avoid squeaking.

I just mount my slats by loosely placing them over a locating pin. Generally I make bed frames from welded steel and this is just 1/4" rod, welded onto the frame.

No, but it may take some permanent curvature. Softwood will do this more readily, especially if it's fairly new timber..

[Stephen M]

If I follow correctly, this is all fixed joinery. This is going to be a beast move (as in down staircase, around a corner and through a door).

Bed bolts a great because it allows the structure to be broken dorn into pieces. It also allows you to snug up the joinery at a later date if the bed is subjected to too much racking force :-)

If I were you, I would construct it like a normal bed (headboard, footboard and rails that detatch from one another) and build a torsion box that sits inside the rails.

FWW has an article a couple of months back on the anatomy of bed construction. You may want to pick that up.

-Steve

[Paul C. Proffitt]

I made a queen size bed for my son last fall. There is no center support (leg or beam). I made the slats from 3/4" poplar. There were 9 slats total and it uses a foam mattress. Each slat was made using 3.5" wide boards (1 x 4). I ripped a 1" section off the side of each board, then turned it 90 degrees and glued it back to the rest of the board. This gave each slat an 'L' cross section which substantially increased the stiffness while only slightly increasing the overall thickness. It seems to be working fine for the past four months.

[Brett Viren]

Just about everything is to be held together with wedged, through tenons and the large cross halving joints for the head/foot-to-rails joints. The only glue will be the used to construct the legs and to lay down the parts of the "picture frame" (there must be a better term) to the head, foot and rail boards. This latter will be done so that everything can still be broken down.

BTW, the inspiration comes from a Japanese bed I once saw. Here is a diagram I think I found posted to this ng a while ago (thanks to whomever is hosting it) that should show roughly what I mean.

'd like to have some nice program like that which made this drawing. as all I have now are pages and pages of notebook entries that probably would only make sense to me. But anyways, my variations are mostly design and not not structural.

-Brett.

[Brett Viren]

Thanks, Andy. I thought hickory might be suitable, I'll look in to ash also. I'm still trying to track down a good source of hardwood (in eastern L.I. NY) so having two possibilities will help.

Heh, maybe I should stick with 3 but just put them all on my side of the bed!

[ ... other helpful things snipped ... ]

Thanks again,

-Brett.

[Stephen M]

snip..

Snip...

Brett, Brett Brett.

Did you not see me winking? ;-)

All the best,

Steve

[Patriarch]

Brett Viren wrote in news: [e-mail]:

I see two contradictory statements here:

[New to woodworking] + [description of a joinery task which would challenge an old chinese master]

And you want to make two pair of the legs which are mirror images of each other, don't you?

I'm guessing that you're an engineering student.

Make prototypes of inexpensive material. And find something else to sleep on, while you learn how to build this very ambitious project.

Have fun with the exploration.

Patriarch

[Brett Viren]

Physicist, actually, but my student days are long gone.

Yep, our floor is very firm. So far no complaints from the queen of the house, so I'm taking my time!

Thanks,

-Brett.

[Brett Viren]

Ah, I see. I thought that was just a nervous tick. I didn't want to stare! .

I guess I still don't have an appreciation of what I'm getting myself into with this project. When I was in college I did a bit of machining as part of my job and a joint like this would be relatively easy to fabricate. Of course that was using a huge, expensive and accurate mill working on stable materials.

In fact, the first bed I made was partially built on that mill. I called it the tennis racket bed because it used a lacing of nylon rope to support the mattress. Dumbest thing I could have done. Every month I'd have to re-tighten the damn thing.

Anyways, I'll be sure to post some pics when I get further. They should be good for a laugh or two!

Cheers,

-Brett.

[Morris Dovey]

Brett...

Please do post pix. This is an interesting project you've taken on, and pictures of the process will be most appreciated.

[Nate Perkins]

Brett Viren wrote in news: [e-mail]:

...

...

Hi,

Have you seen the article in the Feb 2005 issue of FWW? There's a lot of discussion there, some of which involves methods for supporting box springs or matress.

Hope that helps.

Nate

[Stephen M]

That sounds like a "feature" to me :-)

bounce, bounce bounce....

[Brett Viren]

Instead of doing the taxes today like I should I've been playing around with Blender and drew up the hardest part of the design I'm shooting for. Just getting the angles right in software was tough enough. I can't guess what it will be like once woods hits the saw. But playing with this design in Blender really helped to understand the how this all goes together.

If anyone is interested in the result, see here for a couple screen shots:

only show the basic leg assembly and how the head, foot and side boards join in. The border ("picture frame") isn't shown but it will be wide enough to just cover the 3 external sides of the leg. Its corners will be mitered to follow the leg outline and will only be glued to the uprights to allow it to be broken down.

The part that I think I'm going to like most about this is the mitered and beveled board ends that just poke out of the legs. I think they should take on a darker color when finished because of being partial end grain (right?). If so it will be a nice contrast to the leg faces.

-Brett.

[Patriarch]

Brett Viren wrote in news: [e-mail]:

So does your program figure out the angles of the compound miters for you, to take to the saw?

Test your end grain amd mitered grain stain theory on scrap. Wood species and finish types have differing results.

In this drawing program, this bed almost looks possible, with the right tooling and sufficient space.

Patriarch

[Brett Viren]

It might, but I haven't ran across how. Blender isn't really a CAD program, it is more for 3D object modeling. It's rather fun to play with once the basic controls are learned but as far as I can tell it lacks the usual dimensioning and other measurements that CAD provides.

I know how to calculate the miter/bevel angles used in the leg faces. I've found various online calculators or formulae that give this. Here is one:

drawn, the leg faces have a 15 deg tilt (90-15=75 deg for that web page).

I haven't thought about the end cuts yet. At first guess I think they need to be different than the leg faces. Compound miter cuts are still very confusing to me. Another reason this modeling helps. And, this is all definitely something I'll test on scrap before cutting good wood.

-Brett.