Purlin placement question

[meddins]

I'm working on purlin placement in a roof design and I'm treating their placement and load bearing calculations as I would a floor joist.

The question is, can I treat the principal and common rafter connections at the peak (no ridge beam) as carrying half of the roof load between that connection and the first purlin below the peak?

Or do I need to figure on that first purlin below the peak carrying all the load between it and the peak? Thanks,

Miles

[routestep]

If you just look at the forces and moments acting on the common or princple rafter then I would say yes.

However, when you analyze the purlin, all the force that the common rafter carries is passed to the purlin as a point force.

The weight of the common rafter and sheathing, snow load and wind load are the inputs.

[Jim_Rogers]

I'd have to see the frame design to comment.

I don't understand how you can have principal and common rafters in the same roof system, with purlins.

Usually if you have purlins then you have a principal roof system no common rafters. If you have a common rafter system then you usually don't have principals and purlins.

However there are always exceptions to everything.

Jim Rogers

[meddins]

Routestep, in the purlin sizing calculation I'm treating the purlin as a uniformly loaded beam, taking rafter weight and live/dead loads into consideration. I guess my question is more accurately not about the purlin but about the rafter connection at the peak - that is, can the rafter connection at the peak be considered to carry roof load if there is no ridge beam there? In reality it does carry roof load, but for engineering purposes can that connection be considered load bearing? I'm going to need an engineer's stamp on this design so I'm trying to look ahead and make sure I've got all the bases covered.

Jim, I'm working on a sketchup model now and I'll try and post it in the next couple of days so you can see what I'm talking about. The design has big principal rafters connected by a couple of fairly large purlins, with small (3x3) common rafters on top of the purlins. A lot going on I know but it's going for the "old world" look.

[Jim_Rogers]

I have seen something like that before....

[routestep]

I don't think so. What would the free standing peak pass the load to?

The rafter and its mate on the other side would push against each other. The horizontally and the vertical forces would go down the length of the rafter axially until it got to the purlin. The purlin would feel a horizontal thrust and a vertical weight. The purlin on the other side would feel the exact same thrust and vertical force. I think that is how you have to consider the peak.

If the common rafter length is small at the peak down to that upper purlin, then you could consider the purlin to have a distributive load across it. (It will also have a small thrust see above). But if the peak rafter is large a better approximation would be to treat just the rafter's own weight as a point load where it meets the purlin and everything else as distributive load on the purlin. You decide what large is 10 percent? 20 percent? etc.

[Thehardway]

Meddins,

The roof framing you are talking about has some historical precedence, however, it was not usually used in full bent style construction but rather in traditional continuous plate and sill frames and often with the anchor beam style frame found in dutch barns.

Purlin plates mounted on two principle posts used to support principle rafters with an the anchor beam between them controlling outward thrust. Sometimes purlin plates were placed mid span by using canted struts which could transfer load to posts.

The principle rafter joint at the peak was typically a tongue and fork joint.

This arrangement then supported smaller intermediate common rafters over the purlin plates or principle purlins  You will find some of the roof joinery details here  http://www.tfguild.org/joinery/part5.pdf  could be helpful.

A modified example of this can be seen in the Wickwire Barn featured on page 33 of Ted Benson's "The Timberframe Home: design, construction and finishing"

Alternatives to this would be to use a true truss design such as kingpost truss or queenpost truss with purlins and common rafters over them. Outward thrust is thereby contained by the truss design.

[meddins]

Thanks for the replys. I think they've answered my question.

So the uppermost purlin in this design carries the load up to the peak, and also half the load between it and the lower purlin...

[Jim_Rogers]

I think I'd say yes.

You may have too much joinery at one point in those principal rafters. What that does is it takes away too much wood and makes the rafter weak at that point.

It really depends on the size of the tenons of the collar beam and purlins....

Jim Rogers

[BlueDove]

Hi meddins

I am drafting a similar roof layout and was wondering what type of joinery you are doing to connect the purlin to the principal rafters. The structure I'd like to build will resemble the hammer beam on page 24 of Tedd Benson "Building The Timber Frame House -- Revival of a Forgotten Craft" (1980 edition) Ed Levin build it with principal rafters , purlins and secondary rafters.

[meddins]

Hi meddins

I am drafting a similar roof layout and was wondering what type of joinery you are doing to connect the purlin to the principal rafters. The structure I'd like to build will resemble the hammer beam on page 24 of Tedd Benson "Building The Timber Frame House -- Revival of a Forgotten Craft" (1980 edition) Ed Levin build it with principal rafters , purlins and secondary rafters.

Of course a dovetail or simple pocket drop-in purlin would be the easiest route. But my roof design has the top plane of the purlin below the top plane of the principal rafter, so a spline or tusk tenon into the center of the principal rafter would be called for. Crazy..I know.