I have already posted this elsewhere, so I hope it is OK to also post here...

I need your input on the build of the two desks in the image below. ===>> All hardwood construction: Quarter-sawn white oak, and (possibly) black walnut panels (too much...???), or place the walnut somewhere else).

Concept sketches:




1. The executive knee-hole desk on the right is the original design. It has a 30" x 60" top, 1-inch overhang on the sides and the front.

2. However, some new constraints *may* require that the drawer cabinets be mobile, i.e., on wheels (fancy casters, with the trim extending almost to the carpet, to conceal the casters). Thus, the "shell" on the left, with identical dimensions for the sides and the top.

I have the following questions:

a. Has anyone built something like the shell in the image?

b. My concern is that, without substantial support, even a 2-inch top on the shell will likely sag. I have considered moving the modesty panel about 6 inches toward the front, to help with this issue.

3. If you have built something like this, did you incorporate metal framing to minimize the risk of sagging of the top?


I look forward to your input. Thanks!


Al

Hi Al. Welcome to the madhouse!

Have you considerred making a "pencil drawer" at the top of each pedastle as well as the traditional central one? This would, in effect, make the top a "bridge" and would be strong enough to avoid sagging, even if you STAND on it.

As for the casters/skirt question, why not mount simple casters on a cammed system so that they can be retracted out of sight when not needed? Easy and cheap solution that won't compromise the esthetics of the desk.

I love walnut. If it were up to ME, the whole desk would be made of it, but the oak and walnut combo should be very attractive. Have you though about doing walnut drawer faces trimmed in the oak like a raised panel? You could carry this treatment over to the modesty panel and even the sides and fronts of the pedastles.

Just to put in a plug for my other hobby, this would look WILD with an inset leather top!

Keep us informed as to your progress on this one. It sounds interesting.

Paul
the little guy

I have not built anything like the shell version.
While moving the modesty panel inward certainly won't hurt, it's effect on the open edge of the desktop will be minimal. The vast majority of the structural issue (which is mostly deflection not truly structural) runs parallel to the panel and therefore for the panel to really stiffen the open edge it would have to be quite close (probably < 12” away). So I wouldn't move the panel if it compromises any other part of the design. The deflection issue should be solved at the edge where it will be most acute and most noticeable.

I think what I would do is incorporate some sort of beam (ie something that is deeper than it is wide) at the front edge. This could be done in wood or in metal. Using metal would allow a much small cross-section than wood just by the differences between the inherent strength of materials. Using the proper distance, I would test a series of cross-sections of aluminum bar stock that is commonly found at any of the big box stores. Once I determined what the cross-section will be I would design the top such that I could bury (and epoxy in) most of the metal beam in a slot (groove), this will give the the beam the lateral rigidity it must have in order to work properly. I would allow the metal to project below the underside of the top enough so both ends could housed in slots (pockets) on either ends of the carcasses side panels. I would then design some sort of edge detail that would hide the exposed part of the beam.

Couple of other notes:

It should be emphasized that people of very sensitive to even the most minor deflection and the user of the desk will be sitting at exactly the spot where the deflection will be the greatest, therefore the desktop needs to be quite rigid or you may end up with a dissatisfied customer.

It is possible to buildup a series of thinner pieces of metal to develop the amount of cross-section you need to provide the proper stiffness. You might find it cheaper to buy several thinner pieces than one thicker one, given they all will be 'locked' into a groove they will end up acting as one.

If I understand Tiny, he is suggesting you build a torsion box with drawers. This could work as a solution as well and have the additional benefit of providing a pencil drawer (which after all is a typical feature of a desk). However, this could be problematic depending on the desired depth of cabinet drawers.

I think you will definitely have to do something about sag. I think I would try to design a slightly arched rail under the edge. I think that would help with the proportions as well. Seeing the thin edge of the top looks a little strange to me given the overall bulk of the piece. Another good option was the torsion box top, the trick with the torsion box is to keep it thin enough for the eye to believe that it is solid wood.

Steve

One way to "hide" the torsion box wpi;d be to use "copy slides" on both sides. The framing enclosing the slides would make up the bottom of the torsion box over the pedastles and the pencil drawer in the centre would complete it. Just need some good joinery getweem the slide frames and the drawer enclosure to make it stiff enough.

Paul
the little guy

WOW! My reaction at the terrific input here!!! I will see if I can make justice to all the wonderful feedback:




Thanks!

The previous pieces I have made for this client are quarter-sawn white oak and walnut legs. You are right, I have been working to find a way to incorporate walnut in this design. I am making a mock-up, and will share with everyone once ready.





I am afraid you ARE right - and that is my biggest fear.


I have been toying with the idea of placing an apron at the front, maybe two inches thick (please read below).


Terrific! I had NOT considered the beam concept, but that is one worth exploring. I imagine the top must be thicker, sufficiently so to accept this beam.


I will visit the BORG in the next few days, to see what is available. We also have several fabricators that could make me just about anything. I could incorporate multiple layers, given a good top design.





You are right, Steve. I am definitely considering a thicker top, to allow me to bury some kind of structural/support component(s). Also exploring the feasibility of a torsion box, without getting it too thick. The client hates anything veneer, so that eliminates some neat options, such as thin ply layer with great-looking quarter-sawn white oak veneer.






Paul,
The concept of "copy slides" is entirely new to me, but will look into it. I think a little sketch would be helpful, as my mind is not grasping this one. I am definitely open to the idea, once I have a good grasp of it.



My thanks to all, for such wonderful feedback!

By way of background, my client already has in his office this stand-up desk I recently completed for him - walnut legs, and quarter-sawn white oak (his office is currently undergoing renovation), as I mentioned earlier - it is 72" wide, 24 inches deep, and 44 inches tall at the back, while the inclined top slopes toward the front at 7.6° - the joinery on the carcase is all drawbored mortise & tenon:




Also, I hand-cut the half-blind dovetails in the four drawers:




I will keep working on the concepts you all have mentioned, and will post back when something clicks. Thanks again!


Al

I am having a short break at work, the desk I use
is similar to the "shell" version and I don't notice any
sagging.

The top is veneered (BTW veneer is great,
clients that think it is low cost furniture should check
what was done (still is) in France since the XVIII, but I digress)
I guess the substrate is MDF or particle board, about 1 inch thick
but I agree, solid wood should be more prone to sagging.
There is the "sagulator"' http://www.woodbin.com/calcs/sagulator.htm
but I think it applies to wood under constraints...

Do you plan to have drawers attached to the top or just
the side cabinets?
Maybe some battens disguised in the attachment to
the legs would stiff the solid top ?

-Rondo

A copy slide is that little slide out shelf secretaries use to give them more room on their desks for stuff they want to copy on their typewriters. (boy, is THAT ever dating me!)
Some desks had one on one side only, the better ones had two, one on each side. They are simply a sliding shelf about 3/4 thick by about 10 inches wide and 18 inches long (exposed length) with a stop to keep it from coming all the way out and falling on the floor.

Paul
the little guy

Definitely educate the client on veneer and explain that veneer allows you to do things structurally that there is really no other way to do. I think the rail under the top is the best way to go if the customer won't tolerate any "veneered" solutions. I think embedding the metal in a solid wood top would not work so well since metal that thin would flex as much as the wood. The only metal solution I think would work might be square steel tubing.

If you client can tolerate plywood, I saw an ad for some honeycomb cored panels that are supposed to be super sag resistant. It kind of a torsion box in a sheet form. I will try to find a link on this stuff if I get a chance.

Steve

Maybe one single narrow drawer, centered on the top, and mostly to hold pens and pencils.






Ahh! A "pull-out shelf" is what I have been calling those all my life, Paul - I have them in the desk in the shop, an old thing we have had for years. NOW I know. Thanks for this suggestion, as I had considered this early on, but forgot about it as I struggled with the new options. Between these, one on either side, and the drawer in the center, it gets me close to the torsion box concept.







Steve,

Yes, in this case, veneered anything is out. His very old desk was veneered, and it got banged up during a move. Never fixed, the gash looks awful...

Resin-impregnated honeycomb is a wonderful material to make panels, but I have never used it to make such a large top. The company that sells it is VacuPress (at least the honeycomb I am familiar with):



I will be using this in a future project. In fact, will probably have 2-3 vacuum bags going at once...


My thanks again - terrific input!!!


Al
Sandal Woods - Fine Woodworking

Just to be clear the 'metal beam' solution should be installed as a beam (ie the cross-section deeper than it is wide, this could be as small as 1/8"x1", but you would need to test it). As with any beam it must be restrained laterally in order for it to resist a lateral load. A beam would have to be loaded perfectly in line with its center of gravity in order not to have any lateral loading, therefore there is always some lateral loading and you must plan to restrain the beam so it can resist it (thus the embedding). If you have doubts about how strong a very small section beam is get out your yard stick and try break it over your knee on the edge (don't worry about actually breaking the yard stick but do be concerned about hurting your knee ). Now since any metal (except lead and maybe wrought iron) exceeds the tensile and compressive strength of wood by several factors a metal beam of the same proportions of a yard stick will exceed it's strength by many factors. Square tubing would work as well but it won't be as efficient since the top and bottom of the tube provide very little strength (bad cross-section) and are therefore just hanging around so the tube can be a tube and not a beam. An I-beam is the most efficient shape but in this case it would be hard to manufacture There are several ways to skin this cat, and a metal beam is certainly one of them.

hmmm we are maybe adding too much stuff to the desktop
to avoid sagging and the design is going astray.
having one drawer in the center steals leg space,
and having the pullout shelves will force you to have
an apron to support it. The advantage is that you can
dissimulate any strengthening scheme behind the apron.

but I was wondering about the sagging problem, it could be that
we are over stressing it (i.e. it is not a real problem.)
I think you should ask someone with experience in restoration
to check whether desk tops sagged during the years or not.
Maybe RaDioAcTivE could give us a clue!


if you bang solid wood, it would also look bad!!

Thanks for the vote of confidence Ronaldo but about all I have to offer is what's already been said. The top will without a doubt sag without some other design enhancement. I'm in the "just add a rail across the front" camp if one wanted to keep the design simple. Pencil drawers if it works for the ultimate need.

The only thing I'm sure about is that the top will sag at some point with that design. Even with some of the already mentioned design changes it'll probably end up sagging at some point to a minor extent. Tis the nature of the beast that we work with.

Good work guys.........THANKS!!!!!

Neil

As an architect I always over design the structure... I always want to be on the 'over design the structure' side of the equation when developing a design, particularly when the use is unpredictable. One must remember that this is a desk and it is highly probable that it will in fact also be used as a seat (or leaned on) at times... and therefore needs to be engineered to support such loads, the issue of deflection is not simply an issue of sagging (which typically is caused by gravity and or dead weight loading, ie like the weight of books on a shelf)

I've been giving this one some thought, and here's an off the shelf idea that might just solve the problem...
Put an "L" shaped bracket under each end of the top and a cleat under the upper edge of the top of the two pedastles that will interlock with the bracket. This way, the top will be catilevered over the inner edges of the pedastles. That wouls support even MY weight without noticable sagging!
To assemble. just lay the top in place with the pedastles a bit wide, then slide them towards each other untill the brackets lock the top down. This would also keep the top from tilting if someone gets up too fast and hits their legs on the underside of the top, which otherwise might dump the stuff on it onto the floor.

Paul
the little thoughtful guy

I have a faint souvenir of a "factor of safety" in the course
on mechanics of materials, but my remark was that we were
making lots of suggestions that were maybe going away from
the problem or desk.
I agree a desk should resist someone seating on top
but the thickness that would resist that would not sag
on its own weight (save the user always leave something
on top of the desk, i.e. a secretary who knows! )

Agreed. I thought that was what we were talking about and then I got the impression that maybe not everyone was... or maybe it's just me being confused

Glad everything is OK, then , because for a while I might have felt a little the same way... But the discussion always brought me back on track. ? Great stuff!!!


Al

You can go for the modification and the adjustments according to your requirements and choice. So try as per your likings and demands.