Timber rafters - moment plate connection 4

[tasiman]

Hi There
I have a design where the owners require a slanted ceiling (rather than a flat ceiling). The span between walls is 26' (rafter span will be 13'). I have designed the rafters to be 2x8 @ 50", connected at the apex using moment plates either side of the rafters (to prevent sag/creep). By providing a birdsmouth cut to the rafter at the wall end, I am assuming there is no horizontal thrust exerted on the walls, or the horizontal component (in the rafter, due to deflections) is resolved within the moment plate. I have checked the moment plate for this load, the moment due to the udl along the rafter, and the eccentric moment due to a vertical point load at the apex (from hip rafters). I have attached the moment plate design (ignore the M12 bolts - this has been changed to M16).

I do acknowledge that although the moment plates will restrict vertical displacement, there may be long term creep in the timber rafter, which induces horizontal thrust to be resisted by the wall, or the moment plate.

Do you agree with this approach?

All replies appreciated.

Cheers

 

[tasiman]

We are looking at a 30' span for the ridge beam.

 

[CANPRO]

If I absolutely had to make a moment connection with the timber rafters at the peak, I think I would look into using an HSS section to sleeve the timber into and then welding the two sleeves together. The timber would have to be shimmed tight into the HSS. I would image the HSS would have to be fairly long (maybe 3-5 times the depth of the timber?). This however would be expensive and a pain in the ass to build. Hopefully your ridge beam/truss works out.


If you can't support the ridge beam at the gable can you hide a post in an interior wall somewhere and cantilever the beam?

 

[oldestguy]

How about using full span trusses instead of rafters with some depth at the peak but still have a lesser slanted ceiling. Gets rid of all those problems.

 

[Ingenuity]

tasiman,

Welcome to Eng-Tips, and your baptism-by-fire of sorts on this thread! Sometimes we come across as grumpy old men (and women), but we all mean well.

Am I correct in assuming, based upon your 'handle', that you are located in Tasmania, Australia?

If so, you have some great local hardwood species that near replicate 'steel' in their hardness and durability. I especially like Tasmanian Blue Gum for its appearance.

If not, disregard this reply ;-)

 

[JAE]

I'm coming in late here but you will have a very hard time getting the bolts to work with all that moment cranking on each bolt group.

Check out Eng-Tips Forum's Policies here:
faq731-376

 

[dhengr]

Tasiman:
If you insist on using that detail, at least lengthen the two legs on your ridge moment plates considerably, to improve the mechanics of the detail. This will reduce the rafter rotation at the ridge, although it will still be considerable and will cause considerable thrust on the ext. walls. Think in terms of the amount of movement in the bolt holes as the bolts come into bearing on the wood and start to crush the wood until they pick up their full load. Look into the yield modes of bolted/doweled connections in wood. You might do well to look at timber rivets if your wood species is amenable to their use. These joints have much less play in picking up their load, although they probably won’t work in 2x members. Look at selling the owner on tie rods and turnbuckles every forth rafter. Hang a nice light fixture off of them at the center of the room.

 

[MikeHalloran]

The detail drawing shows a solid vertical line at the center of the steel plate. ... i.e., it's shown as two five-sided plates, butted together at the ridge plane, and attached to each other by means not shown. There's a fair chance that you will have to send someone up there to butt weld the plates together, and two other guys to arm the fire extinguishers.

At least the welding heat will harden the rafters a little and perhaps increase their resistance to splitting.
... which looks like a guaranteed failure mode, given the moments and the small number of fasteners.






Mike Halloran
Pembroke Pines, FL, USA

 

[tasiman]

Perhaps the sketch attached earlier is not a good example - it is just indicative. The detail is actually for 4 bolt arrangement on each rafter, with beefier bolts, and the plate will be continuous with no joints. For the plate at least the maths does work, however I see that practically it will have issues. In light of the discussion here and the good and relevant points that have come out of this discussion, I have been able to incorporate the ridge beam at the top by providing an intermediate portal frame to support it. This will also help to tie the top of the walls nicely. I have remodelled the entire roof system to see the overall effects.

Thank you all for your insight and advice. This is really appreciated. Good to see so much interest in the thread. I will, at a later stage research this option, with the objective to perhaps devise (innovate!) a practically sound solution. If you do have any more ideas, please add to the thread. I can truly say this forum is relevant and very much alive, and worth keeping tuned into. Cheers

 

[Jerehmy]

If you use the moment connection someone like me will be writing a report for why the roof is sagging, the wall is pushed out, and all the finishing is cracked. Then they see a faulty design and you get sued.


Use an LvL ridge beam and be done with it, don't reinvent the wheel.

 

[BAretired]

The easiest solution is to use a ridge beam, but if you want to get creative, you could design the roof deck as a folded plate with a tied arch at the gable end.

BA

 

[msquared48]

If you have to, use a steel ridge beam if the PSL or glulam will not work for you.

Mike McCann, PE, SE (WA)

 

[rittz]

I have not seen a residential building that used rafters and ceiling joists for a few years . Light frame trusses are used almost exclusively here in Western Canada , Much easier to install, and cost effective. No thrust to worry about either.

 

[Tomfh]

Good idea using a ridge beam.

Even if you have a rigid cranked apex beam (and I agree with the other poster that you didn't), your walls still get pushed outwards a lot.

 

[XR250]

With a moment connection, your rafter span is effectively 26' (not 13'). I am still wondering how you are going to do that with a 2x8.