Timber frame garage

[cbecker]

Here is my sketch-up of a 3 bay 1.5 story garage I have been working on designing for quite some time.  Finally got it to a point where I think it is ready for some feedback on it and if there are any major issues that jump out.  It is not completed with all of the joinery yet.

Plans are to put it on a concrete foundation that is 12" above grade. Also plans for a second floor in the future potentially finished as living space. Roof will be metal.
Future plans to add a shed roof off the back so there will be only temporary 2x purlins across the back wall to hold board and batten siding. I was planing on putting 2x purlins on the outside of the entire frame so that I can add 1.5" foam before putting on the siding so that I don't cover up any of the exposed beams inside. 

Also have a few questions:

1)I have it drawn with 4x8 sill plates.  Is it better to go with post bases anchored to the foundation or to anchor the sill plates down to the foundation and then have stub tenons on the posts? 

2)How do you determine how many knee braces are required?  I know they are required in opposite pairs.  I didn't put them on the interior queen posts to purlin plates to maximize headroom but if they are needed I can add them.

3)With purlins on the outside are girts necessary in between the posts as well?


Any feedback would be greatly appreciated or if more info is needed I will do my best to supply it.

Thanks in advance

Chris Becker

[Don P]

Hmm, downloaded 13 times and no comments.  Is it that ugly? :D
I'm stuck at the 2016 version of skp and couldn't open it. Can you save as an earlier version or post a pic?

[cbecker]

I hope it isn't that ugly?? Not looking for engineer/load analysis just looking for feedback on general frame design.  Will try to save as an older version and post it.

Chris

[cbecker]

Here it is saved as 2016 with some updates in joinery. Still not complete but added detail and corrected braces

[Dave Shepard]

Can you post a screenshot? I don't do sketchup. All my drawings are in my head.

[Don P]

Well, how about that, it's rather handsome 
Dave, you may have been the first post of the new year.
I got it downloaded, this is a screenshot;


 
Structurally it would be more efficient with H frames under the purlins but that would make the end door about useless just size the ties appropriately for the point loads and any floor loads if they are carrying a loft.

1) either would work. A treated sill gets the end grain of the post up off the masonry and provides a good nailer for siding, studs, etc. I'd rather see a connector from post to sill than mortises in the sill, they are almost always rotted out in old work. Not old school just observation.

2)Stiffness attracts load. In a situation where the  frame is being racked by wind or ground motion the longest, stiffest, diagonals forming triangles in the siding will attract the lateral load. If the siding is installed with this in mind it is potentially a far longer hypotenuse of a much larger triangle than the perimeter braces. There is also the opportunity to distribute the load over many more connections. The roof plane is no different. We don't use knee braces in modern construction. The sheathing provides better bracing. In post frame buildings you'll often see steel X's of strap or adjustable rods in sided bays, those are their brace bays.

3) No they are not required. Reading the above with respect to using the siding or sheathing for bracing. If you use lightweight girts they are only bracing the structure if the walls and girts are stiff enough not to buckle out of plane when the building attempts to rack.

4) Double check the roof purlin dimensions

[cbecker]

Thanks Don for the feedback!

Now I am going over my load calcs and have come up with a few questions

Attached is a photo of some of the calcs.  Just want to make sure I am on the right track

1) When calculating the load on the roof purlins do you count the rafters that fall directly over the queen posts as part of the load for that section or is it not counted because the load is transferred directly through the post

2)When using the beam calculator for the tie beams how do you account for the floor loading and point load from the queen post.  Is it correct to run two separate calcs? One with the point load and then one with the combined point load and floor load in the uniformly loaded beam calculator

[Don P]

These are similar in a way.
The calcs I've made are just a few of many that are for different situations. I picked a few that I use often. An engineer figures out the bending moments along the beam for each loading condition and location and adds them. At each location or more importantly at the locations where the bending moments are going to be highest. (Moment; the force and desire of a beam to rotate around its supports in response to a load or loads, measured in ft-lbs). Where a combination of the loads produces the highest bending stress, that needs to be quantified and compared to the allowable stress for the proposed beam. 
Whew  :D

1) There is a "cheat sheet" the table of concentrated load equivalents. So to answer your question in the laziest possible way, I looked at the cheat sheet first. The bending moment produced by 3 equally spaced rafters point loading the beam looks like it would be the same as if we looked at the beam as uniformly loaded along its entire length. 



 
There's the tributary area to the beam as I see it, ~12'x9' or 180sf
180 sf x 45psf=8100 lbs on the beam. For the trib area to the queenpost shift the blue left or right to center it over a post... in other words that is also the queen post load.

I've lost this twice today, I'll post this part  :D

[cbecker]

Ok I think that all makes sense.  So by moving the queen-post/roof purlins closer to the peak it would lessen the tributary area and in turn decrease the load on those beams/posts?

[Don P]

In 2 ways. It would reduce the roof area supported by the purlin and increase the area supported on the exterior wall. It would also decrease the floor area above.

Have you looked at a single ridge beam?

[cbecker]

I originally was against a single ridge beam to keep open space in the middle but now that I have learned more it may be a better way to go that way.

That would also increase the number of rafters needed correct?

Also I grew up on a farm with two 1800's barns one with canted queen posts and one with vertical so played into my design direction as well

[Don P]

The rafter span goes to something like 12', doesn't mean you need more but they probably need to be deeper.

[cbecker]

Also been working on practicing brace layout and cutting



 



 

 




  


And started eyeing up some timbers for best layout and yield


 



 

[cbecker]

Here is a version with a structural ridge

I am not sure how to capture a screenshot of it otherwise I would post that as well


Also a question about layout of the timber above.  Is it better to put the wane up or down?  I am looking to use it for the east end tie beam.

[Don P]

For a screenshot, I rotate and zoom on the whole thing or the area of interest then File>Export>2D Graphic and save to my desktop or the file for that project. I usually email those quick screenshots to clients, subs, inspectors as we are talking back and forth.

Grading, that is tough and requires that you step outside of your immediate need for that timber, its best and highest use might be as a 5' brace. This is where I certainly understand the call for an independent grader, be dispassionate. In that phase I am Eeyore "Looks like we're going back to the woods, oh well".

What you have that the grader doesn't is a knowledge of where the timber is going and how it is loaded. This beam is spanning less than it's mates in the clear span of the building so you can figure out what is required in that place. There will be less load on it and more support. We're blinking power, posting..

[cbecker]

 I am trying a screen shot.  

Is this the correct tributary area supported by the ridge in this bay?  Which is about 12'x 12'


>

[Don P]

Good deal, you got it, and yes around 144 sf x 45 psf= 6480 lbs uniformly distributed.

[wkf94025]

 



I lost my previous shop in the CZU Fire in August 2020.  Which led to the acquisition of a mill, and countless other obsessions.  Doug Fir was milled and baked last summer, and planing and joinery starting later this month.  15' x 29' with long edge facing the ocean.  Debating building my own doors and windows, and learning a lot about Lift and Slide this month.  Using redwood for 8x8 porch posts, and 1x8 V-rustic vertical siding.  Poco a poco...

[cbecker]

Ok I think I have refined my design enough to have it checked over by an engineer.  I am located in eastern pa.  I don't need a sealed set of prints just want to make sure that I didn't overlook anything in the design.



   

 

[Don P]

Well, the monitor blew last week, this old machine is almost ready to be replaced. Couldn't open the file but the screenshot looks fine. I'm assuming you checked the ties for floor load. I think you are about ready to turn it in.

[cbecker]

Here it is saved as Sketchup 2016 version

[Don P]

I'm glad to see joinery drawn in, but it brings another round I think.

I've switched to x-ray to look at these;

Start with this housed birdsmouthed rafter foot. As the rafter shrinks, the level cut will support the rafter, the housed heel will lift off its bearing seat, dangling that lower 1/4 of the rafter. The stress will concentrate at that inside corner and try its best to split out of that corner.

The housing is doing damage to the beam, It's not the gross size of the beam you need to check, it is the net section. I'd like to hear other's thoughts, mine is to lift the rafter and reduce the notch and housing. The rafter is hanging from a ridge beam, no thrust, I'd run a modern birdsmouth and lag em down to an unnotched beam. The rafter is then full section bearing the load from above. Brace mortise, read the grading rules "knot or hole", keep the mortise in the right range to not cost a grade drop.


 

Same thing, stress riser at the re-entrant corner. Put a beveled cant strip on top of the beam and set a bridle jointed rafter couple on top of the cant strip, no notch, full bearing, solid roof plane connection across the ridge. Make the rafter couple drape over the ridge with its feet resting on the lower plates.


 

Net section check. Because of those drop in notches the beam is evaluated as a rectangle of full depth but only the width of the wood beyond the notch. In this case those ties need to be checked as a 5x14 (there's a cleanup needed at both ends of those joists before turning in). If that isn't working beamwise either a tusk tenon or set the joists on ties with blocking in between.


 

That's first quick look, I've spent more time writing than looking at the moment.

[cbecker]

My mistake.  Both ends of the loft floor joists are suppose to have the same joint. The reduced size dropped into a housing.  I will fix that later and upload the updated model
Thanks for the input

[Don P]

Here's what was in my mind.
The tie is deep enough to do the "best" tenon for a joist to beam. This is leaving a 2" strip of undamaged wood along the compression edge, the upper flange of an imaginary I beam. The mortises are right along the line of the neutral axis, it is in the web zone between the compression flange of wood above and the tension flange below, there isn't much needed in the area of the web in an I beam. It is designed to resist shear. The neutral axis does have the highest horizontal shear, right at the supports, look at the net shear area right around the pocket closet to the post and check against max shear for the beam from the calc.

I'm showing a single 10" washer head structural screw in double shear through the 4" long tenon, 2" from the end. I'm mentioning checks as I see them go by.



 

With the joists installed, switched on x-ray. It goes straight in on the bottom, unnotched, a soffit tenon. On the top I drew a healthy diminished haunch or tusk blending it back out of the neutral axis towards the high compression zone on the upper flange. The shape of the slope of the haunch in the tie is also pretty much exactly following what the stress diagram looks like. But in this case our bottom flange is 6.5" tall, sweet.




With this type of notching the beam is undamaged from a beam design standpoint, where the one yesterday had effectively reduced the beam thickness by 3". The strength loss is huge when you do drop in connections.

I'm done playing with your model for the night, also modified and lifted the rafters up onto the ridge and plates. I didn't draw in a bridle joint over the peak. I was playing around a little bit to show you some tricks with skp as well. Pull out the default menu on the right sidebar of the screen and scroll to "layers" if it isn't there. I set up a few layers, start switching them on and off. I wanted the rafters of of my way while working on the joists, etc. Anyway, if it seems like a step forward, modify on this one, if its backing up hit delete.

[Ljohnsaw]

I set up a few layers, start switching them on and off.

Layers are great.  I think I have about 20 on my cabin plans.