*
Andrew:
Maybe I’m just being dumb, but it looks to me like the rafter you are bolting to that notched joist is going to do a pretty good job of reinforcing the notch. Also, if you’re going to double up the rafters, how about one on either side of the notched joists. Who’s gonna see the notch then?
Any engineers out there?
Replies
*
Andrew:
Maybe I'm just being dumb, but it looks to me like the rafter you are bolting to that notched joist is going to do a pretty good job of reinforcing the notch. Also, if you're going to double up the rafters, how about one on either side of the notched joists. Who's gonna see the notch then?
Any engineers out there?
*
Yep, I thought about that. And yes I can certainly "hide" the defect. I'm still grumpy that these carpenters 60 years ago reduced these beautiful old growth 2x10's to 2x4's at the plates.
*Andrew,I have to agree with Steve. It looks like your bolt pattern will satisfactorily (sp?) reinforce the end of the joist.If the only increase in load on the josts is tension, I think what you have is going to be fine.I'm assuming that the joist spacing and size is sufficient to handle the additional ridge beam loading without pulling out the bolts.
*
Andrew
I may be wrong off the top of my head, but I think if you were to consult a good span table or better yet consult you local building code you would find that notching the joist as your drawing indicates does little to affect the structural integrity of the joist. I haven't had occasion to notch a joist in this manner recently so my memory may be serving me wrong (At this moment i don't have access to MY code book, or else I would look it up myself.) I sure you can notch at least a third of the joist legally here in Vancouver. Five inches is extreme, mind you, but concidering the joists aren't carrying a live load this shouldn't be a structural problem that way.
My concern, as you note, is in the tension. The ceiling joists primary function in this case is more of that as a collar tie in which case the joist to rafter connection is your main concern. Your proposed sollution appears to address that nicely.
I am curious though about the ridge beam. Does it end over a gable, or does it 'die' into a hipsed roof? Or perhaps more succinctly, is the ridge supported at either end, that is a post that follows down to a load bearing beam or continues until it bears on a foundation? And is your new ridge beams sized and rated for it's required span? If so, the need for your joist/collar ties is greatly reduced. You didn't mention why the ridge needs replacing, ie. is the old ridge sagging? An addition? Attic space?
Whatever the case, you look to be on track and good luck from here on in.
Helpfully (?) yours, cybergreg
*
I think the notching rules are these:
No notches within center 1/3 of span.
Notches outside center thind of span not to excede 1/6 of joist depth.
Notching directly over bearing supports not to excede 1/4 joist depth.
There is almost no moment stress on the ends of the joists but this is where the shear stress is the highest. see the attached picture for what I mean. (I can't take credit for the picture)
*Yo Andrew - You helped me out with my little "2X4 problem", so I thought I'd like to try to help you out with yours......(-: I'm not actually an engineer, just a P.E. (That's "Pretend Engineer") So I can't really give you a formal answer. But I'd be willing to give it a shot. Could you give us a bit more info? What span are we talking about out to out of walls? What's the pitch of the roof? What spacing are we talking about? What's the snow (live)load in your area? What species and grade are the rafters? Is there going to be floor loading on the 2X10 joists ? Are the 2X10 joists continuous from one plate to the other, or do they lap on an interior wall? Why double up the 2X6 rafters? Might be cheaper and easier to use a single 2X8 or 2X10 rafter. Sorry to ask so many questions. I just want to get a clearer picture of what's going on.
*
I'm not quite getting the picture here. It looks from the scaled drawing that the rafter location is moving; are you lowering the rafters?
And Andrew, if you are really installing a "Ridge Beam" (a true Ridge Beam) then there is NO increased load on the rafters.
But, your probably not installing a true ridge beam. Let us know more detail about that apparatus.
*
OK, the larger picture:
Shed dormer project on 60 y.o. Cape. Existing roof is 11/12 with a 1x ridge board, 2 doghouse dormers in front. Plan: Install 32' ridge beam -- two 12" LVL's spliced -- supported by four trusses spaced roughly every 9' and made up of just two opposing rafters and the floor joist. The truss is a simple triangle, just like the current roof construction, but now it supports a ridge beam. An old JLC article called this a "king truss." The rafter span is about 12', and the vertical drop is about 10'. Now, this places the rafters in compression and the floor joist in tension, resisting the outward thrust of the rafters. More numbers: the floor joist span is 12', lapped (I assume) at the center bearing wall. I'll see to it that they are adequately connected each other to resist the tension. As for floor loading, these will be sleeping rooms and we could easily get by with 2x8's. There is some plaster cracking in the ceiling below consistent with joist sag, but who doesn't have that?
I decided to double the truss rafters because at 2x6 they're a bit flimsy, plus this will cut the shear stress at the bolts in half (1/2 on each side of the joist). The lower rafter end does need to be lowered a couple of inches because I want it to bear on the plate directly and because I need good wood contact for adequate bolting -- especially given the deep notches! The trusses towards the center of the beam span will be concealed in partition walls, essentially invisible.
Finally, remove remaining rafters along back of house and hang shed dormer rafters at 3/12, supported equally by ridge beam and by direct bearing on rear wall. Some numbers: local code requires roof loading calcs of 30 + 10 lbs/sf (major overkill); this works out to about an average 4000 lb point load on each of the four trusses. A structural engineer has already reviewed and stamped my plan.
I know this all sounds a bit complex, but it's not so bad in reality, and this approach solved several problems I had getting the result I wanted in a cramped Cape. There are no posts or load-bearing walls to be added.
Did I mention we're expecting a baby in December? This project is meant to add anothe bedroom upstairs and a usable bathroom. The deadline, my wife informs me, is non-negotiable, heavy penalties for running late. :) And the law firm I interviewed with tells me I'm in final four for two positions, out of a "zillion" interviews, so the future is looking more and more muddled.
What exactly is a P.E.? A structural engineer?
*Andrew - I did something very similar to what you're describing a few months ago. Sounds like you've got everything under control, plus lots of good suggestions, but here's another. In addition or in place of the bolts you could use a nailing plate (one of those Simpsons with about 50 holes in it), and nail it to the inside face of the rafter with half of it sticking towards the interior, so you could then nail it off to the floor joist a few inches further in from the end, where there is a little more meat. The only disadvantage is it might make the rafter not butt tight to the floor joist.
*
Andrew, I feel sorry for you about the deadline. Makes me feel a little better about this mess.
*
Andrew,
A P.E. is a professional engineer. After getting a BSME, an engineer can spend five years as an EIT,(engineer in training) and can take a licensing exam. In most states, only P.E.'s and architects can sign off on building plans.
I have one disagreement with your reasons for doubling your rafters. Using one rafter on each side of your bolts will double the shear strength of the bolts. But your week point is not the bolts. You will tear the joist before your bolts shear. No matter how many shear points your bolts have, your joists will only have one.
*True -- but my motive was primarily to reduce the required bolt count from 5 to 3, so I could use a better bolt pattern and avoid rendering Swiss cheese of the slender 2x6. The tension on the bolts should be at worst case a couple thousand pounds. The bolt strength estimates assume bolts installed in SPF lumber. I'm sure three 1" bolts are ultimately a lot stronger than this!I think I may well scab on a foot or so of 2x10 to increase the grip on the joist and to space the extra rafter out a bit more from its twin. I think a 16d nail is good for a few hundred pounds in shear? If the rafters meet only one joist width apart at the ridge, the metal hangers will overlap anyway.Thanks for the feedback. I enjoy mulling over these details as much as or more than solving the problem. The search for the "elegant" solution.
*
Andrew - have you considered the shear load spreading device that looks like a huge oversized washer with teeth poking out both flat sides? They are sandwiched between two members, ie. rafter and joist. I read (never used one) that they reduce the need for multple bolts and reduce drastically the potential for bolt tearout because of the area they grab in both members. ???? Ralph
*
Andrew - You certainly do have an unusual situation ! I've never heard of anything quite like it.
A few things come to mind. You mentioned that the 2X6s are "A bit flimsey". I'd have to agree. I'd personally like to see a larger rafter size. If you want to save on bolts, maybe you could move up to the next larger size (diameter) and still only put 3 bolts in. A larger rafter would also give you more depth for insulation.
You mentioned something about "32' LVLs spliced". Do you mean putting 2 LVLs 32' long together ? Or are you buying shorter lengths and splicing them ? I don't like to see spliced beams. You can definitely get LVLs in 32' lengths.
Your "King truss" will have to support a lot of weight. Has the engineer addressed the rafter to ceiling joist connection on these? Or is that the detail you're showing us?
I'm not totally sure about the value of 16D nails in SPF, but I think it's about 100 pounds, not several hundred.
The thing Ralph is suggesting is probably a "split ring". Someone still makes these things - I believe it's a company called Cleveland. (They make hangers and stuff like Simpson) Last time I checked into one of those, the tool to make the ring in the lumber was about $400. I don't have any info about the design values of the rings, or how much they cost.
Major pressure with the little one on the way. Is this your first?
*In old timber frame construction, they used to use a cylinder of steel about 2" in diameter and maybe 1-1/2" long. The wall was about 1/8" thick.They would cut circular grove around the bolt hole and this cylinder would "lock" the two pieces of lumber together. Essentially creating a much larger bearing area and therefore stronger joints.
*Child #2. We can always live downstairs, heck we were in a 700sf one-bedroom when we had the first. Before he was born people kept politely asking, "Do you have the baby's room ready?" Uh, yeah, considering it's our room with a bassinet.The splice is end-to-end and in the middle of a particularly short 8' span (the trusses are not spaced exactly equally). So the engineer wasn't terribly worried. Yes, I can get 32' LVL's, but can I lift them? No. Can I maneuver one through the thicket of trees and wires and houses here (our neighbors are about 12' away)? No. You get the idea -- the real world v. the engineer's world. Plus my lumberyard has these sizes in stock.You're right, the shear strength on a 16d is less. I wonder how much it varies -- the angle it is driven relative to the load direction must be very important... Anyway, I will use 2x8's for the trusses now that I've seen the notching mess. This will make placing the bolts more comfortable and all rafter material will be 2x8 too. I was fairly confident of the existing lumber because it's the old style stuff that's larger and 4x the density of modern SPF.On the truss bearing weight: there are ceiling joists, but they are not part of the load-bearing calcs. I doubt their contribution will be much at 8 ft on a 11-ft. roof anyway. Nor are the partition or knee walls counted. In reality, of course, they will contribute significantly. In fact, the two center trusses are incorporated into partition walls, and maybe I'll slap on a couple sheets of plywood. Plus we almost never get much snow here, and local code sees a 3:12 roof as steep enough to meet only 20+10 (30 lbs/sf), etc. etc.On the rafter-to-beam connection: a strong metal hanger is assumed and I will try for the best possible attachment. There will be about 12" of wood-to-wood contact at the rafter-ridge junction, so room for lots of nails. Suggestions welcome! I could run a bolted LVL gusset from rafter to rafter and under the beam but, sigh, that's even more work... Houses appear to either (1) overengineered or (2) falling apart. Never just right. What's happened to elegance? Yeah, I know, look what Frank Lloyd Wright did.See, I'm basically trying for what you truss guys provide -- a floor plan devoid of load bearing considerations!P.S. I've seen split-ring connectors mentioned, but had no idea what they were, so I ignored it!
*
I had to sketch out on paper what you were doing to fully understand it all. It looks like you have it all undercontrol, especially if the PE looked it over. That's a real interesting solution, one I've never seen. In essence you have a ridge "beam," not "board," going to be supported at midspans by the trusses rather than end bearing locations (at the side walls)as they usually are.
I'd be real careful about the beam and the splices. Get the PE to give you the nailing schedule on that, it could be intense.
I was thinking that if you can get a truss up to this attic space can't you bring the full length LVL's in the same way?
One final note, you mentioned in one of your posts that your ceilings are showing cracks at the center span of the room below. I'm not sure of your attic space wall layout, but, this might be a situation where you could install a "strongback" ('stiffback') atop the joists (if I'm understanding your situation correctly). A strongback could stiffen the ceiling joists near the center of the wide room below and help reduce possibility of further cracking in certain spot locations.
Best of luck w/ the new baby! One is fun, Two is work; you won't find a better job to work at than "working with" those two. Hoping all goes well.
Phil
*
As I should've done earlier, here's the CAD of the site-built truss. I hacked it a bit to make it smaller, so I hope it travels legibly.
The splice will probably be a piece of LVL or 2x12 with bolts through it, quite strong. Also, it is sort of in a cantilever position -- thrust up by the long spans at the ends of the house, pushed down by the weight bearing on the center span -- so that the net moment shouldn't be much. You know how it is -- if it just doesn't look right when you put it together, you add some more.
The floor I'll look at ... currently we have 1x oak directly over the joists as a second floor, so I'll probably lift it up and poke around. Might be easiest to drop a few more 2x10's in there. If I really had a lot of time, it'd be nice to hang an entirely separate set of joists to support the upper floor to avoid any extra load on the plaster below and to isolate the "pitter-patter" of little feet (or the sound of grown-ups below who actually like to stay up past 8!). Now that I think about it (uh-oh) it would be great to physically separate the two planes. I could run TJI's to get a nice flat floor, shim them up a bit to clear the tops of the 2x10's. How would I keep them from rolling on the top plate? Hmmm.
*
As the attached sketch to scale indicates, the builders notched the 2x10 joists, um, a bit deep (nearly 5") to install a ledger to support the rafters above the top plate. Now I'd like to double up and bolt the 2x6 rafters to the joists so they can support a greater load imposed by a ridge beam. The load ON the joist is not going to increase with the remodel -- the rafters will carry it to the plates while placing the joists in tension -- but I don't think the notch will impress the inspector much.
Any thoughts on an attractive way to do this? Obviously I can sister or replace the joists, but that's overkill IMHO. I considered scabbing something on but don't know how to go about it.