I'm toying with the idea of installing a vehicle lift in my garage. I've seen the low profile lifts, but want one that allows me to stand up while under the vehicle. Therefore, I'm trying to modify an existing ceiling to incoporate a raised, boxed area for the vehicle when the lift is fully raised. The remainder of the ceiling would remain at it's current height. I only need to raise it 2 or 3 feet. The rafter system is an older stick built system not the modern truss system. Any ideas are greatly appreciated especially from those who've done this already. Thanks.

 

what's the width of your garage?
what's the slope of the roof?
you may be able to install a collar beam on each rafter 3 feet above the ceiling jost that is there now and build a raised ceiling section
by doing that you are actually build a truss, just use plywood gusset plates on each side to strengthen the joints

 

Be careful, the spreading load on a roof is tremendous. This is the load trying to knock the walls down or spread them apart letting the roof collapse. On my 5 pitch roof it is 811#per lineal foot on both ends of the "A" framing when material, snow, and wind loads are taken into account.

 

brownfoot,
My garage is 16' wide with a 6 / 12 pitch roof. Installing collar beams at the desired new ceiling height is kinda what I'm thinking. I'm just not certain how to tie it in with the existing ceiling joist while not creating added stress on the lower portion of the rafter where it meets the sill plate on top of the wall. Plywood gusset plates are a good idea, Thanks.
Torque1st,
Is there any formula or calculation I can use to determine the force exerted against the walls based on the distance the collar ties / joists are from the upper sill plate, and the added stress on the portion of the rafter between the collar tie and upper sill plate? Snow weight is definately a consideration. I've seen times where we only got 2" of snow one year then 18" the next. Thanks.

 

An Architect or Structural Engineer can do the calculations. You can find some formulas etc in various handbooks but it takes knowledge, experience, and that nasty thing called mathematics to use them. Any time you make structural changes like that an architect should be consulted. A permit is required by most cities and codes governing agencies for any structural changes. The permit process insures that changes have been properly checked by an architect or structural engineer (which I am not).

Most carpenters can size a simple floor joist or simple rafter. A carpenter can build something that will stand up on it's own but guaranteeing it to stand or be safe under a snow or wind load takes an actual Engineer etc. Actual structural design takes an Architect or Engineer trained in structural engineering.

Remember, even Architects and Engineers make mistakes tho. Watching "Engineering Disasters" is always good for an eye opening.

 

I actually did this. My building was 34 feet wide and was originally constructed using site-built trusses on a 4 in 12 roof. The 4-in-12 pitch and the struts of the trusses made it difficult to get as big of an opening as I wanted. The portion of the truss needing to be cut out is one of the most critical parts -- the bottom chord resisting the "rafter spreading force".

I used a freeware finite element analysis program called "Grape" to calculate the loading on the truss members and to investigate approaches to transfer load across the gap. The approach I used involved using structural steel beams to pick up the tension load from each side of the 4 trusses I cut, and additional steel beams perpendicular to these two to connect (and cancel out) the tension load. The perpendicular beams (running parallel to the trusses) were bolted to the ends of the main beams.

Using 40psf roof loading the tension load was something like 2800 pounds per truss, so a fairly hefty steel beam is needed to span the 10 feet.

Another tricky part is the connection of the steel to the wood members (the same issue you face in attaching your collar beams, only worse). I used Simpson PHD5 hold-downs and their SDS3 screws to tie the cut tension chords to the steel beams.

I got it all in place, bolted together, and demolished the shoring walls. I was pleased when the ceiling did not sag even 1/16 of an inch. The building has survived one full winter although the wind and snow load so far has not even come close to the 40 psf design loading.

I'm not sure I'd do it again. It was a lot of work, and it still won't let me put vans and SUVs at the full height of the lift, due to the restricted depth of the opening. (It works OK for cars and pickup trucks). In retrospect, it might have been easier to cut out this section of roof and re-frame and re-roof the whole thing, and the end result would have been better. But it was kind of fun to take on as an engineering project. The one engineering mistake that I caught prior to completing the project was failing to consider lateral-torsional buckling of the steel beams. I had sized them adequately for the moment load but the unbraced length for out-of-plane bending was too long. I ended up welding on some additional steel to the compression flange to stiffen the beams in this axis.

If you are wondering about codes and permits, my building is considered "agricultural" and is permit exempt. I'm not a structural engineer by training. I do have a shelf full of books on the subject and I think I know what I'm doing (and my limitations). I do not have any employees, and if my building falls down it is only going to squash me and my projects.

 

fefarms,
Thanks, that's great advice. I like projects like this. I built a 16 x 16 shed myself, however this sounds like more then I should tackle. It's looking more like I'll get some custom engineered trusses and just redo that portion of the roof. Thanks again for the insight.

 

I remember looking at the Grape program but at $190 I could not afford it at the moment. My total tension load was 1622plf on a 24'wide roof. It does take a hefty beam and ties. I have codes so everything has to go thru a structural engineer.

 

You can download and run the Grape program without paying the registration fee. As in $0.00 or "free". When the program starts and pops up box requesting the registration code, just click "OK". When it then complains and asks "abort, retry", just click "abort". At the next popup, click OK again. You are then in the main screen of the program.

The authors of the program advertise that it can be run without registering it on their website, so I don't see any licensing issues with just using it this way.

It takes a bit of practice to set up a good model of a truss. The first thing you learn is that you must constrain the nodes in the Z-axis (out of plane) direction, or the solutions are unstable. (This is just like a real truss). The next step is to understand the difference between beam and truss modeling elements. An idealized truss is loaded only at the joints, but real roof trusses have a distributed load all along the "rafter". You can't model the resulting moment load and total stress on the "rafters" using truss elements. But if you use beam elements you need to use "release" nodes to model pin joints at the ends.

Be careful with truss members subject to compressive loading. Consider the possibility of Euler buckling, particularly if they are also subject to externally applied moment loads (as the "rafters" are). Garage size "W" trusses are usually OK given the ratio of width to unbraced length.

 

Thanks for the tips, I will try the program out for fun. I hate those limited time trial programs, it always seems I get back to actually using them about 15 minutes b4 they expire $#*%^@!

Can we say "wet noodle"!!! ROFLMAO, I can just imagine what would happen without that Z-axis restraint (chuckling so hard I can't type). Of course anyone that has handled trusses out on a construction site knows just how flimsy they are before they are secured. I have seen a few "accidents"...

I worked my own calculations but had them checked. The numbers came out the same within rounding errors.

I am still looking for a good CAD program for Linux. I use an old $35 off the shelf W-98 IntelliCAD on windows XP and while it has a few bugs I can work with it. I used ACAD for years. A good cheap, fee, or very low cost CAD program for Linux is the only thing holding me back from using Linux. After using a full blown CAD system it is impossible to go back to those limited "Etch-a-Sketch" CAD programs.