Replies

1. 
   DarinS | Jun 28, 2010 12:11pm | #1

   Deflection

   There are steel construction manuals out there put out by the AISC.  It lists a TON of properties about different steel sections, so you may be able to read it directly out of the book.  The deflection, though, is going to depend on how the C channel is oriented.  The force would obviously would be more if it is oriented in a "V" shape as opposed to en "L" shape.  Where it is loaded also plays a factor as it will deflect more in the middle than the ends, if both ends are held in place.  If you need further help on this, feel free to contact me.  The main questions are how is the C channel held in place, are both or only one end held, where is the load applied.  From there, the deflection can be calculated pretty easily.

   Hope this helps you out at least a little, and once again, if you need further help on this, feel free to contact me.

   Darin

2. 
   PSeverin | Jun 28, 2010 12:11pm | #2

   Structural Engineering

   Your question, as a previous question about the holding power of a fastener is more complicated than it seems.  I think you are headed into the realm of structural engineering.  You might check and see if you have a friend who is one.  Then you will get into the complications.  Which channel are you using, a real steel channel like a C3x5 or C12x 20.5 or a piece of unistrut?    When you say shear force are you looking for the ultimate yield strength that will cause the head to pop off as you keep screwing in or the shear strength across the screw perpendicular to the screw axis?

   It seems like such a simple question and the answers are out there, but probably buried under complications most of us get lost in.  So I'd find a structural engineer.  If you have a school that trains them you might start there.  Or you might see if you called a structural engineer listed in the phone book if he could get you what you want as a casual inquiry.  If all else fails and you are really desperate, hire one...pay for an hour of their time.  It might be worth it to you.

   Peter

3. 
   DonStephan | Jun 28, 2010 11:11pm | #3

   I 'm hoping to investigate on a gross, onceptual level a king sized bed support system.  All in walnut, approx 3" square posts, headboard, footboard, and approx 7" x 1 3/4" side rails.  Attached to each side rail are three angles, each made by bending in the middle a 12" x 1 1/2' X 1/4" steel bar.  Each angle is surface mounted to the side rail with two #10 x 1 1/2" wood screws.  Running across the bed between the three pairs of angle are three  76" x 2" x 1" steel C channel, with the 2" flat middle surface facing the ceiling.  To keep the concept simple, the C channels are simply resting on the angles..  The mattress, box springs, and two occupants are considered to weigh a total of 1000 pounds.

   Using data and a formula in Understanding Wood, page 123 and thereabouts, the force needed to pull a single such wood screw from walnut is about 400 pounds (using measuirements from a screw on hand).  In this bed concept, the withdrawal force would not really be along the axis of the screws, but this is just a conceptual look.

   The mechanics of load distribution in this concept are complicated, and right now there isn't need for sophisticated analysis.  I'm wondering what force in pounds would be required to break or shear off a pair of these wood screws, and how many pounds, applied in the middle of the 76" long C channel, would cause 1/8" of deflection.  1/8" of sag may or may not be everyone's success/failure indicator, but it seems suitable for a conceptual look.

   I agree completely that a detailed analysis would require the services of a mechanical engineer.

1. 
   RalphBarker | Jun 29, 2010 12:43am | #4

   interesting exercize

   Interesting mental exercize, Don. It seems to me that the force on the screws won't be straight down, but at some unknown angle. If we assume that both the C-channel and the supporting steel angles don't bend or flex, the downward load force would start to compress the wood the screws go into, mostly at the surface near the steel support angle. At that point, I think the screws would start to pull out, rather than being sheared off. And, there's the potential that the corner of the support will act as a fulcrum, exerting force against the head f the screw along the axis of the screw, as well. Thus, the top screw might have to work harder than the bottom one. That could be minimized by bolting the C-channel to the support angles, though, I'd think.

   FWIW, my instinct is that #10 screws might not be up to the job. I'd lean toward three #12s in each of the supports, two near the top, and one in the middle. Note, however, that I'm not a mechanical engineer.

2. 
   PSeverin | Jun 29, 2010 04:04pm | #5

   A call to a Structural Engineer

   The answer is about 144 lbs per #10 screw in shear, fastening a 11 ga (close to 1/8")  metal angle to a mixed maple wood support.  There may be a safety factor of 3 to 4 probably built into those numbers meaning tested failure occurs at multiples of the value listed.

   The information came from a call to a structural engineer.  He said the information is found in a NDS (National Design Specifications) for Wood Constructions.  He thought that the manual is like a lot of these code books, available if you pay money for access.  Codes for connections through plate anchors like Simpson Strong Ties is found in another of these lovely books.  He said there are a lot of factors in determining this number and they consider failure by shear or crushing the wood fiber and use the weakest failure situation.  We used maple because the book doesn't consider walnut and oak is stronger, but we know that.

   Most loads are distributed across multiple points.  A  1,000 lb load in the  center of one of your channel slats would be carried as a 500 lb load to each side rail.  Remember that you are putting a box spring on top of the slats which further distributes the loads to numerous points on the side rail.   So 1,000 lbs would have to be carried by 7 screws.   That would seem to be an entirely reasonable expectation.

   Your interest in deflection of the channel slats deflection seems somewhat arbitrary.  I would say go and test a king size mattress in a store, see if the deflection is noticeable and see what they are using.  Unless you are putting a sheet of glass under the box springs, the limit on deflection unless it is really excessive, seems not that important.

   And lastly structural engineers  are the ones who size beams, columns and connections to make sure that structures don't fall down.  Mechanical engineers are the ones who work out plumbing, heating, ventilation and air conditioning systems in a building, a difficult task on its own.  Like chisels and planes there are many varieties of engineers, each to his purpose.

   Peter

1. 
   RalphBarker | Jun 29, 2010 04:47pm | #6

   titles

   Peter, thanks for the correction on structural vs. mechanical. I used the wrong title. When I worked on Stonehenge, we weren't as specialized. ;-)

3. 
   sapwood | Jun 30, 2010 01:25am | #7

   I'm an engineer and I've got all the answers

   Don, 

   I am both an engineer and a furniture builder. The engineer part of me says he's gonna go get a beer and sit this one out. The furniture maker part of me (partly informed by the other) says to divide your 1000 pounds by 12 screws. This gets you 80 some odd pounds. Attach a strap of something to the leg of your bench using one #10 screw. Stand on it. Jump on it. Does it break? No... it won't. So the screw is OK. To be sure (and to employ the age old adage of engineers: "When in doubt... make it stout.") add a screw or bump them up to #12's. 

   Now consider the deflection issue. There isn't any. A box spring is designed to be supported by the edges. Problem solved. But, if you want to do the same sort of analysis that I do... get a length of the channel.... pile some weight onto it while it is being supported on its ends by a couple of bricks (or use beer bottles cuz by now the engineer has emptied a couple anyway), see if the channel deflects beyond what you consider to be "too much*". If it does, get a bigger channel or sell the bed to smaller people. 

   * too much = some amount that gives a center sag that causes both occupants of the bed to roll to the middle. Note, "too much" is inversely proportional to their ages and/or the length of the relationship.

   This isn't rocket science and while I like a good engineering analysis as much as anyone... strict structural analysis isn't the way I derive the sizes of furniture components. Engineering process has huge safety factors built in.. it simply isn't appropriate. The methodology is, the actual number crunching isn't.