Wood does not shrink in length to any degree that you could notice. Wood shrinks radially and tangentially - width and depth. The rafter shrinks along those two shrinkage planes. Since the plumb cut is on an angle with the top restrained, some of the shrinkage has the effect of shortening the lower part so the bottom of the plumb cut opens up a bit. This is normal and expected.
I don't think this is the case in the OP's situation since they've only noticed the gaps in the area where they removed the collar ties.
That is not true. A cathedral ceiling supported by a ridge beam does not historically have any different connections at the ridge and it does not have collar ties, yet it doesn't become an airplane wing.
That is true.
That is not exactly true.
This statement indicates you do not understand the forces a rafter experiences.
That is not true. I could use a 2x16 rafter and that does exactly nothing to the connections or loads at either end.
You do not know what you're talking about. Could you diagram the forces on the rafters? Do a little moment diagram for a section through the rafter? Do you engineer your nailed connections and determine what shear load a connection will experience when you decide how many and what size nails to use? You're lack of understanding is misleading others. This is not a good thing.
I suspect that in the OP's case you may be right, but without seeing it firsthand and determining exactly what is happening and why, I wouldn't bet the farm on it.
Installing a post under a ridge, essentially creating a structural ridge, can put a very large load on that post, and that load must be carried down to the foundation. I would strongly suggest not installing a post without a pro designing a complete solution.
I would attempt to put all back the way it was. The biggest problem would be closing the gap at the ridge board and rafter plumb cuts.
If the rafters have not lifted at the exterior wall, there may be a way with lateral leverage and weight on the roof to close the gap and reapply collar ties.
It's not the nailing, it's the approach that would appear to need some review. The unsupervised DIY'ers have already shown their competence on this job by causing the problem.
Amen, Glenn. Every statement you made jives with what I know about collar ties. I am not an engineer, but I think some people should do a bit more research before spouting off as experts.
Most collar ties are 1x material nailed on with 8d nails. I really can't envision any circumstance where they could prevent the spread of the side walls. It is my understanding that a collar tie is to stiffen and prevent uplift on the leeward rafter. Does installing a stiff back (purlin?) braced off to top plates take the place of collar ties? I would appreciate Bob Morrison's clarification of the issues.
There seems to be a general misunderstanding about the difference between "rafter ties" and "collar ties". I wanted to verify my information so I googled. For an interesting discussion of the same thing go here:
If I was able to get a passing grade on stresses on a multi-span bridge years ago in engineering school, I really believe I do understand stresses on a simple roof but I bow to your "superior knowledge". (laughing and bye)_
Apparently it's time for a refresher course. You're letting your assumptions run away with your diagnosis. Let's let the facts speak for themselves, and not your assumptions. Fair enough?
Collar ties are not necessary to tie the walls together _if_ there is a suitably connected rafter/ceiling joist system to take the load. There is separation at the ridge. Differental movement in structural members is by definition a failure. Neither of us have seen the place and have no idea of how the place is framed, whether there are other existing structural issues or not, right? So we're starting at the same place. Let's summarize the information to date.
The OP has done a substantial amount of work to the house, including foundation work and replacing the roofing, and he's been living in the house since February. I think it's fair to say he is familiar with the house. His descriptions have been clear, and it is also clear that he is not clueless when it comes to construction. The OP is observant and is asking a question about the gaps that have appeared at the bottom of the plumb cuts only on the rafters where the collar ties were removed.
I won't use the word if in any of this because that requires an assumption. There are gaps at the rafters where the collar ties were removed and not at the rafters with the remaining collar ties. The OP did not mention whether the gaps were on only one side of the ridge or both. No assumptions, let's deal with both situations.
Gap at only one side of the ridge: The OP did a fair amount of foundation work, so there were obviously some issues there, and we can't conclusively eliminate the foundation as the proximate cause of the gaps. A gap at the bottom of the plumb cut on only one side requires either the ridge to drop in elevation or for the foundation to move in one of two ways. 1). the foundation must rise on the side of the gap (possible due to frost heaving or the foundation work, but an extremely remote chance); 2). the foundation must drop on the opposite side of the gap (possible for the same reasons as above). You would have to make some assumptions to calculate how much differential settlement would have to occur in the foundation for a
1/4" gap to open up at the plumb cuts, but, no assumptions > no calculation. Suffice it to say the foundation would need to settle a lot to open up a gap at the that size. Foundation failures show up on lower floors before they show up on upper floors due to distribution of the stresses through wall and floor diaphragms. The OP would have noticed a whole host of problem areas, such as doors and windows binding, wall and ceiling cracks, etc.
Gap at both sides of the ridge: The ridge has to settle. Again, only two possible scenarios: 1). the rafters have to bow _up_ as the ridge settles. Hmmmm....okay, that's not a possible scenario.
Which leaves us with only two possibilities: 1). the walls the rafters are bearing upon have spread apart. 2). the OP isn't observant, or not able to make a coherent statement. I've seen nothing to support that conclusion.
To DanG: I checked out that link and it doesn't really differ from this one. There is a lot of misinformation being bandied about. Let me put it another way that should make it easier to digest.
In a correctly framed house there is a lot of redundancy and a lot of dual/triple purpose structural members. Floor joists tie the walls together, act as a diaphragm (even if not designed that way) to distribute racking and shear forces, and also provide support for the floor. Compromising any one of those functions can be problematic. In the OP's situation, there has been a localized failure due to the removal of the collar ties. Regardless of the designer's primary reason for their installation (stabilizing rafters, preventing uplift, tying walls together, etc.) their removal has caused the walls to spread apart. The evidence certainly seems to point to the collar ties tying the rafters together as an important function, whether designed that way or not.
As far as the size of the collar ties, a 1x4 with a sufficient number of nails in the correct pattern (no splitting) can take a very substantial load - wood is good in tension. Which also begs the question, why would someone use a 1x4 in axial compression to tie rafters together? Uplift is certainly one explanation, but as I posted earlier, the nailed rafter/ridge connection, and sheathing/rafter/ridge connections would perform that function in almost all areas except hurricane-prone areas. In my area I am currently required to strap over the top of the ridge and tied the rafter pairs together. That obviates the need for collar ties to prevent uplift right there, but it does nothing to prevent the rafters from splaying.
In that other thread in that other forum, someone pointed to code as not requiring a sufficient number of nails in the collar ties to handle the load required to hold the walls together. Since we have no idea of the construction of the OP's collar ties, house framing, whether or not it was over or under built, and it's unlikely we can ask the original designer their intentions for the collar ties. We can't say why they were installed. All we have is that they were removed and gaps opened up.
It's also not a suitable defense to say you followed code for your nailed connections if you're an engineer, architect or simply interested in quality construction. Code is the minimum acceptable codified construction and does not address the requirements of a particular structure. Loads and connections must be designed. I hope everyone understands that something built to code and no more is not quality construction.
I would suspect some structural damage. Take cross measurement in the attic and height at all four corners. IMHO if the differences are less than 1/4", I wouldn't worry about it. I would replace collar ties also. Cheap fix would be to pull rafters closer and then nail collar ties.
HomeOwnersHub website is not affiliated with any of the manufacturers or service providers discussed here.
All logos and trade names are the property of their respective owners.