I am remodeling an old house which has a screened second story porch. I would like to remove the 2×4 rafter collar ties and replace with cables to open the room and ceiling. Room is approx 14 by 15 feet. Roof is 12/12 with 2×4 rafters on 2 ft centers set on a 4×6 which is equivalent to the upper plate. No snow load as in the desert! Ridge is a 1×6 which I plan to strengthen and add strong tie connectors at ridge as well as rafter on 4×6 plate. Goal is to have visually open so can see a stained glass window which will be set into the gable end. Rafters are about 12 feet long and have true 2×4 cut showing no obvious deflection,.
How best to substitute cable for existing 2×4 collar ties??? Thanks
Replies
Is this a shed style roof or a gabled roof over the porch?
If the former, you may not need any rafter ties, depending how the rafters are connected to the main roof
Welcome to the
Taunton University of Knowledge FHB Campus at Breaktime.
where ...
Excellence is its own reward!
Sorry, forgot the detail that is a gable roof.
Don't discount consideration of thin rod vs cable ties ... may look cleaner. Just thinkin out loud. Always consider many options, the one you choose will be the best fit.
OK, then you do need something.
One option is to place a structural ridge beam, which can be set in under the existing.
The cable ties will work too. I've done that on a couple.
I see it more often on larger structures, like old churches and factory type buildings.
But you really should have an engineer to spec it all for you.
Welcome to the
Taunton University of Knowledge FHB Campus at Breaktime.
where ...
Excellence is its own reward!
Thanks, How did you attach cables to rafter and tensioner to make them asthetically pleasing?
I let the engineer spec that one.Incidentally, yesteday i was inspecting a house with all high cathedral ceilings. Each room had one or more rod ties with turnbuckle connection in the middle. The way they were attached to the rafter system was hidden to me behind the ceiling finish surface
Welcome to the Taunton University of Knowledge FHB Campus at Breaktime. where ... Excellence is its own reward!
Thanks, incidentally, I have already used hurricane ties for the existing rafters to the 4x6 top plate and plan to sister 2x6 next to the rafters for insulation space as well as for the increased structural strength.
We recently did a system similar to what you're looking for in a cathedral family room. In this case, wall/rafter spread was mainly handled with a large ridge beam, but we also had steel tension rods with center turnbuckles as supplementary support.
The steel rods were threaded for the turnbuckle and had 1/4" steel plate welded to the ends for attachment to the rafters. Each steel plate had (4) 3/8" holes drilled in them. The plates were set at the geometric center of the rafter and 3/8" X 3 1/2" lag screws with washers completed the installation.
View ImageView Image
Diesel,I'm going to have to do something like this soon too. Just curious, why did you feel the structural ridge was not enough?Steve
It had nothing to do with how I felt..... just followed the architect and engineer's specs. I'd say they ran the calcs and felt the that dormer wall needed more support against out thrust. That ridge is a triple 16" LVL spanning 26' with no place for an intermediate post... so tension rods it was.View Image
it should be i pretty simple job to do, cables are much stronger in tension than wood so just a tiny cable will do the job of a 2 x 6 etc.
the thing i would be most worried about is the attachment point. you want to make sure it doesn't come loose there. i would use a belt and suspenders. remember the rafters are usually tied to the joists or collar ties as well as the wall. make sure of the rafters connection at the wall also.
but once you have that figured out i would just make up the cables to the points at either end and then tighten the cables with a come-a-long untill you are relieving the tension on the joists. clamp down the cable at that point and then remove joist.
to do that you use the crosby clamps or cable clamps to make the "eye" on the ends of the cable. you have to make the cables long, and a temporary "eye" on the long end to be able to hook the come-a-long and apply tension. once you have tension you tighten cable clamps close to attachment point and cutt off the excess.
"cables are much stronger in tension than wood"
By weight? By size?
Wood is very strong in tension. Arguably stronger than in compression. It never really gets used effectively that way because of the difficulties in attaching the ends.
By the way, this is just a niggling aside. Your post had excellent advice.
cables are stronger in tension than wood.
think about it like this, from a rigging perspective.
how much (what size of) cable would you need with a crane to pick up a load of 10 tons?
now on a straight vertical pull and not even worrying about the attachment points just the strength of the wood itself, how much (what size of) wood would you need?
i have the charts for the steel cable, but i am just guessing about the wood but i think it would have to be at least 6x6 to even come close to 10 tons.
so i think by both weight and size the cable is stronger, or just about any other way you could mention it, i think the statement steel is stronger than wood is true.
thanks for the niggle, i don't mind.
In the small town museum near me, which is mostly filled with moldy looking stuffed cougars and rusty logging equipment, there is a picture of a huge Douglas fir. The caption goes on about "these magnificent trees" (which the early inhabitants immediately logged) and then states that pound for pound Douglas fir is stronger than steel. When I read it I started ranting on to my wife: "How can they say that? In shear? In compression? Anyway, a steel beam that weighed what that tree does would probably be a tenth of it's size..." They also repeat this canard every time they publish a picture of a good sized tree in our local paper.
That having been said, I have never seen a wood member fail in tension. The values must be pretty high.
i don't know what to say about that, but as a rigger i have a good practical knowledge of rigging and the forces tension exerts in picking loads.
rope is very strong, manilla made from hemp especially. wire rope the strongest. how can wire rope be stronger than fiber rope, but a piece of wood with its imperfections and made up of similar fibers be stronger than a wire rope?
that is not logical, it just doesn't make sense. i don't believe it is correct that a piece of wood is stronger than a piece of cable.
i don't know what to say about that, but as a rigger i have a good practical knowledge of rigging and the forces tension exerts in picking loads.
rope is very strong, manilla made from hemp especially. wire rope the strongest. how can wire rope be stronger than fiber rope, but a piece of wood with its imperfections and made up of similar fibers be stronger than a wire rope?
that is not logical, it just doesn't make sense. i don't believe it is correct that a piece of wood is stronger than a piece of cable.
Because you are viewing strength as a function of diameter. To be equal "pound for pound" fiber rope would have a cross-sectional area greater than ten times that of wire rope. A fair comparison would be 3/8 wire rope versus 1 1/8" fiber rope. Both (according to my quick numbers) should have a pound per foot roughly equal.
thats all well and good measuring "pound for pound" if it is true, and i don't want to have to look up or calculate the weights.
i would much rather just shoot from the hip, i still say steel is stronger than wood. you are not convincing me otherwise, also in this application we are discussing, i think 3/8 wire rope is plenty strong to replace the tension lost by removing a 2x8 joist.
so for our application, the practical matter we are discussing, the wire rope is much "smaller" than the wood, and again just shooting from the hip here, lighter than the 2x8.
you must be an engineer, quick to calculate, but with no bearing on the application. you guys do have your uses though, don't get mad.
Nope, I'm a carpenter. Not my fault I went to college.
Done a ton of rigging too, used to work for a rigging outfit
The whole premise of this argument was wether the claim that pound for pound wood is stronger than steel is in anyway valid. Which is appears to arguably be, or close to. That does not in any way make wood or other natural fibers a better building material for skyscrapers, boilers, dams, planes, or rigging
Edited 6/28/2008 2:07 pm ET by McMark
Now you are reminding me of those astounding pictures of bamboo scaffold structures in Asia!
Looked it up
1 1/4" manilla .42#/ft SWL = 1.2tons
1/2" wire rope .42#/ft SWL = 3.2tons
So at least as far as wire rope goes, steel is much stronger pound for pound than natural fiber.
But the strength of wood could be much different that manilla fiber
The caption goes on about "these magnificent trees" (which the early inhabitants immediately logged) and then states that pound for pound Douglas fir is stronger than steel.
Pretty un-scientific, and taken from Wikipedia.
Tensile strentgth for A36 steel 400Mpa, density 7.8g/cm*3
"pine wood" 40Mpa, density .8g/cm*3
Tensile strength as a function of density
Steel (400Mpa/ 7.8) = 51 (units uncertain)
Wood (40/.8) = 50
I do not know if this is valid scientifically, and only applies to tension
Edited 6/28/2008 1:32 pm ET by McMark
I think they can get away with it by saying "pound for pound," bla, bla, bla. Doug fir may be stronger for a given weight, though I would venture to say in tension and compression, but would wonder about shear. For example, a spider web, pound for pound is an order of magnitude stronger than steel, but I won't be using any cobwebs for collar ties any time soon.
There are too many variables involved to make a blanket statement that Douglas Fir is stronger than "steel" by weight. Wood is orthotropic, which means it has different properties depending on the direction of the grain relative to the direction of measurement of a property such as compressive or tensile strength. Steel is a generic term, and there are thousands of types of steel, plus different heat treatments that yield different properties for each one.
Steel often also has a grain direction, depending on how it is made, but properties are much more uniform and predictable for steel than for wood.
Wood works better in compression than in tension, mostly because we can utilize the entire end surface of the wood to carry the load, such as in a column. In tension, only the longitudinal cross sectional area of the nails or bolts transfer load to the wood. If you look at a Simpson Strong Tie specification, you will see that the load ratings of the brackets are only valid if all of the required fasteners are installed. Likewise, notice how the steel plates that sandwich built-up trusses have hundreds of fingers that grab the wood in an attempt to distribute the load through the most area.
Another drawback of using wood in tension is that it can split at fasteners and lose all of its load carrying capability.
In most residential construction, wood can be used in tension and compression, because the loads are light and there are lots of pieces for redundancy. That's why collar ties and trusses are ok to be made of wood.
If you replace lots of wooden collar ties (which were safe due to redundancy and load sharing) with fewer steel ones, be very careful with the end connections. The steel ultimately connects back to wood rafters, and that wood needs to take the entire tensile load of the steel tie. The best connection would be to make the steel bracket loop around the back of the rafter, so fasteners would only be needed to prevent the bracket from slipping up the rafter. Alternatively, spread the load out over several bolts.
The folks at Simpson Strong Tie have an architectural series of brackets, so the guy who asked the original question should check into that and might find what he needs. http://www.strongtie.com/products/apg/index.html
Paul Formisano, P.E.
Thanks, that's a pretty nice precis of the situation. This whole relative strength of wood-steel discussion has very little useful application, it's just interesting. Because the connections dictate the useful strength of wood in tension, it really doesn't matter how high the values are. There is no way to practically approach them. My understanding is that most tests of wood in tension produce only approximate numbers because of this.
Edited 7/1/2008 12:55 am ET by fingersandtoes
Needless to say it would have to be approved by the building inspector, which means it would have to be speced by an engineer. Certainly it could be done. The biggest issue, besides the strength of the cable, would be the connections. You couldn't just drill throug the joists because you would only be holding half the joist, from the hole drilled to the inside edge. You would have to use a bracket of some sort, probably something that would wrap around the joist. Connecting the cable to the bracket might also be an issue, would probably have to be welded or maybe connected with a hub, like a giant guitar string. Whatever the case, it would have to be a strong enough connection to withstand maximun strain, i.e. during heavy winds.
I don't know a lot about these things, just pointing out some issues that I see. But it would look nice and it certainly could be done.
I was thinking something like this...
View Image
--------------------------------------------------------
Cheap Tools at MyToolbox.net
See some of my work at TedsCarpentry.com
Mine have been eyebolt through top plates with heavy washers.
Welcome to the Taunton University of Knowledge FHB Campus at Breaktime. where ... Excellence is its own reward!
I'd have reservations about securing the top plate but not the rafter itself. What if a heavy wind grabs that roof and pulls it away from the top plate? I realize you're way beyond me in these matters, but i can't help but wonder about this. Even without collar ties, the rafters are tied to the ceiling joists. Maybe if there were Simson ties holding the rafters down, then I'd feel safer. --------------------------------------------------------
Cheap Tools at MyToolbox.netSee some of my work at TedsCarpentry.com
Good point. That is why I recommended the engineer. Think I did anyways. The PE wil spec all the attachment points and methods for the live loads in the area.
Welcome to the Taunton University of Knowledge FHB Campus at Breaktime. where ... Excellence is its own reward!
I thought we were the engineers.--------------------------------------------------------
Cheap Tools at MyToolbox.netSee some of my work at TedsCarpentry.com
Instead of cable, think about using 1/2' emt conduit. At $2 for 10', it's a pretty cheap material, and easy to work with. Use a vice to crush the ends and drill holes in the new flat surface. Screw those into the rafters just over or under your existing, then knock out the 2x4 that currently exists.
Rebuilding my home in Cypress, CA
Also a CRX fanatic!
Save the Whales! and Guns!
Was that an attempt at humour or are you trying to hurt somebody?
Welcome to the Taunton University of Knowledge FHB Campus at Breaktime. where ... Excellence is its own reward!
Well, I used 1/2" conduit for a cannon barrel when I was young and (more) stupid. I was damn lucky it never split or shattered. Yeah, I would question its use as collar tie material--for one thing, flatttening the ends will tend, I would guess, to "work harden" it and make it brittle, as well as making an area where stress will concentrate and make it likely to fail there. My lawn mower handles always break right where the tube is flattened at the point of attachment to the bracket. Break a handle on a lawn mower, it's inconvenient (and the tubing for that is much larger and stronger than 1/2" EMT); break the conduit holding up you roof, and it's more than inconvenient!
Was going to post this only to your email, but figured maybe the whole forum could benefit from my sage advice, since I did manage somehow to live through my mispent youth into a mispent middle age. Don't mean to be picking a fight or anything. (Though this is a fairly important thing--as you said, someone following questionable advice on structural matters could get themselves hurt, or worse.)
Not only is the metal conduit brittle, but the point of attachemtn is critical in a system like this too. A couple misc screws thrown in helter skelter won't pass muster.
Welcome to the Taunton University of Knowledge FHB Campus at Breaktime. where ... Excellence is its own reward!
Just for the sake of being a PITA (those wonderful flatbreads that you put gyro meat on), I'll defend my idea a little more here.
In this case, we are not talking about a very serious roofing system here. These loads you are envisioning for this roof aren't happening. No snow, though there might be wind. At 12:12, the loads on the rafters are pretty evenly distributed at the ends. The only thing these collar ties are going to do is keep the two halves from seperating at the ridge. Half of the roof is already supported by the house itself - it's not the cap on the structure. The original builders used 2x4 because that's what they were already using and had it on hand.
Replacing the wood with solid steel or tightend cables just seems like overkill - fine for coastal blizzard conditions with a slate roof with 2x10 rafters 16" oc, but too much for a light 2x4 roof with no loads other than a breeze.Rebuilding my home in Cypress, CA
Also a CRX fanatic!
Save the Whales! and Guns!
I would bet that the OP meant to say "rafter ties" instead of "collar ties".
Jon Blakemore RappahannockINC.com Fredericksburg, VA
I agree Jon, the OP is referring to a 12 in 12 gable roof that he wants to "open up" to the rafters. Sounds to me like he wants/needs something to substitute for the work that the CJ's (or wall ties) are doing seeing as he doesn't have (or intend to install) a structural ridge to combat the out-thrust at the walls. And EMT conduit ain't gonna do it. Even if it would.... try explaining that one to the inspector! LOL...View Image
"The only thing these collar ties are going to do is keep the two halves from seperating at the ridge. "You mis-understand the situation.now I were to tell you that the only thing the wheels on your CRX is doing is holding the frame from touching the ground, would I be right? Nope - they need to roll and transfer power also.You don't see me giving too much mechanical advice tho.
Welcome to the Taunton University of Knowledge FHB Campus at Breaktime. where ... Excellence is its own reward!
NO - SHEESH!
Thats what the cinderblocks do. Gets it high enough I can mow under it.Rebuilding my home in Cypress, CA
Also a CRX fanatic!
Save the Whales! and Guns!
You don't let your wife have any exercise?
Welcome to the Taunton University of Knowledge FHB Campus at Breaktime. where ... Excellence is its own reward!
The last thing that will happen without ties is the rafters separating at the ridge. He's got to worry about the walls bowing out.
Another thing that puzzles me about your post: If half the roof is already held up by the house, what holds up the other half?
Edited 7/1/2008 4:09 am ET by fingersandtoes
The way he has described his porch is this:
14'x15' walls
12:12 roofline
2x4 rafters, 24"o.c.
4x6 top plate
one side attached to existing structure (side of house)
The other side he wants to place a stained glass window... presumably where a gable vent would be in a conventional roof. That's why he wants to get rid of the "Rafter Collar Ties" as he puts it. BTW, any attempt to make these out as "Rafter Ties" going from top plate to top plate has no basis in what the original poster entered.
By my reading, it's likely that 4x6 top plate goes all the way around. Thats a pretty good start for holding one side of the wall from bowing out, the house keeps the other half of the wall from bowing out.Rebuilding my home in Cypress, CA
Also a CRX fanatic!
Save the Whales! and Guns!
Ah, I understand what you meant about the house holding the roof now. Although with only the ends secured I don't think the 4"x6" top plate will present enough resistance to stop the wall bowing in the middle.
The distinction between collar and rafter ties has as much to do with function as location, and in this case I'm pretty sure they were acting as rafter ties in that they were there to resist outward thrust. However it does bring up another point. Our code allows rafter spans to take into account the presence of solid collar ties. This benefit would disappear if cables were used.
"BTW, any attempt to make these out as "Rafter Ties" going from top plate to top plate has no basis in what the original poster entered."Not true at all.
using the wrong terminology and calling a rafter tie a collar tie is one of the most common errors I see here.The fact that he wants to open up the space for view describes where these members are! There are low enough to spoil his view so they are in the lower third, making them structural rafter ties.That and the fact that the ridge is a non structural 1x6 ridge board offering no support whatsoever means that there is an absolute need for the ties and that brittle conduit would not begin to do the job.
Welcome to the Taunton University of Knowledge FHB Campus at Breaktime. where ... Excellence is its own reward!
Paul, I think you're missing the part where he talks about his "1x6 ridge board"...... this thing is a gable.... and gables need to have something to combat the out-thrust at the plates. I believe his "rafter collar ties" are doing that job now and it is this function that needs to be replaced with another means.
I maybe be wrong... but when he talks about "one side being attached to the house" I think he is talking about the porch itself and not the roof. Otherwise there'd be no need for the ridge or the walls ties for that matter.View Image
brian, that twisted rod and turnbuckle in the photo, was that the actual product speced ?
we use that for our formwork systems.
is that going to stay as a finished product ?
carpenter in transition
Hi Tim... somehow missed your post.
The stuff in my pics consists of solid 1" round steel stock, threaded on one end, with a 6"x6"x1/4" steel plate welded to the other. Two of these pieces are connected with the turnbuckle to become one "tension rod". They were made up by the GC's steel shop and were then inspected (the welds mainly) by the structural engineer before install.
Yes, they will stay. They will not be very noticeable though. There will be two decorative arches in the room (three sided hollow) which will render them virtually invisible from the floor of the finished room once they're painted to blend in with the ceiling. It's a little over 22' floor to ridge in that room.
EDITED to answer the second half of your question!
View Image
Edited 7/5/2008 11:45 pm ET by dieselpig
Edited 7/5/2008 11:47 pm ET by dieselpig
I like the look of them. Wouldn't bother me at all if they were visible.
I think if they were to be visible, I'd want a few more of them... like maybe four spaced out equally of course. That's a pretty big room... about 22' wide by 30' long and 22' to the ridge...the pair of tension rods look a little lost up there. I think the arches will look awesome too though. I'm definitely going to go back to that one and take some pics as the finish comes along. We just finished framing that house about 10 days ago.View Image
I'm thinking that like Diesels's job, rod and turnbuckles will be the best. Solid rod doesn't stretch, like wire rope could, and looks better
i think the thinner cable and absense of turnbuckle looks better. if you have ever installed safety cables or gut lines or guy lines it is very easy to tension them, then after they are in tension you clamp them off close to the anchor points and trim excess for a very clean installation, plenty strong.
I've seen it done with rods also. Rods with a turnbuckle in the middle. Just for chuckles I guess.
The rods might be cheaper too.
Heysmith,
Can you clear up this confusion for us?
Are you referring to removing "collar ties" which are generally in the upper third portion of the roof- they typically have the function of keeping the rafters together at the ridge in case you have a high wind or some other force that wants to open the roof like a sardine can.
Or, are you referring to "rafter ties" which predominately tie the rafters ends together in the lower third portion of the roof. The rafter ties withstand the lateral thrust that weight on the rafters will tend to exert.
Jon Blakemore
RappahannockINC.com Fredericksburg, VA
Thanks to everyone who has replied thus far. Quite a discussion! The existing horizontal 2x4 pieces go between the rafters at about 1/2 the vertical height so not really upper 1/3 and definitely not lower 1/3 as side wall height is only 4 and 1/2 feet. Gable end is 2x4 framing. The 4x6 which functions as a top wall plate is on the sides and these have been attached to the original house. The object is to open up for and replace the current 2x4 horizontal pieces which connect rafters. I am thinking of having a metal plate with attachment for aircraft cable made with turnbuckle in center of room. Will attach the cable to the plate bolted to the rafter (probably 3 1/2 x 8 inch long with at least 8 5/16 bolts through the rafter). This plate will be hidden from view and cable ties also hidden under 5/8 drywall. Forgot to mention that I am planning to sister new 2x6 rafters glued and nailed to each rafter and the plate attachment bolts will go through both. Of course, hurricane ties at the 4x6s. Hope this helps clear the air. Thanks
Sorry, the air is not clear.Can you post some pics? Maybe some are following you, but I cannot determine exactly what is going on.Is my drawing even close?
Jon Blakemore RappahannockINC.com Fredericksburg, VA
Drawing looks right. As to pics, my wonderful wife who knows how to post pics is visiting relitives and I will have to wait. Would rather fish here in Alaska than learn how to post pics now!! The new reinforcing for the existing 1x6 ridge will be a boxed beam of 2x8s which will be visible. As to the location of the existing 2x4 horizontal ties at the mid position, guess they are OK since been in place for over 70 years and no sign of wall spreading or rafter sagging.Happy July 4th.
Alaska?I thought you lived in a desert?
Jon Blakemore RappahannockINC.com Fredericksburg, VA
Desert is to dry out and warm the bones.
I think you are on the right path there. I would be comfortable with that.The fact that you do not have a structural ridge makes the rafter ties an absolute necessity. What you have are not in the ideal location but they are functioning as rafter ties since there is nothing else holding the structure together. The problem as they are at that location is twofold. One is that they don't function as efficiently but the 4x6 helps contribute.The other weakness in that location is that they stress the rafters at that point, so it is good that you are doubling them up.sounds like a good plan you have.
Welcome to the Taunton University of Knowledge FHB Campus at Breaktime. where ... Excellence is its own reward!
I plan to put up a new ridge just under the original as well. Thought I had mentioned this but many postings ago. Again, appreciate all the input.
You did, but that got lost in the scuffle. Since it wasn't detailed as to exactly what size and how supported, it probably didn't get too much attention. If you have it engineered for the span and loads, and fasten the rafters to it, you could do away with the ties altogether.
Welcome to the Taunton University of Knowledge FHB Campus at Breaktime. where ... Excellence is its own reward!