rafter ties

[paulmars]

Im planning on installing rafter ties in my 1952 cinder block home. It has no rafter ties at all. Every time Im in the attic and look around, I wonder what holds up my roof. The ceiling joists run parallel to the roof ridge. Im also going to install collar ties. Currently there are only two. The roof does not have much sag and the walls have no apparent bow. However, Im adding hurricane straps, supports for the gable ends, and extra fasteners to hold the roof planking to the rafters. Vertical rafter supports and then a new roof cover too. So, I decided to add the rafter and collar ties too.

In my research, I have discovered the idea of using cable instead of wood for the rafter ties. Cable is stronger and will allow me to remove some of the pressure on the side walls, using a turn buckle I can pull up the roof slightly. I have a string pulled straight across the ridge inside the attic. I do not plan on trying to take out all the sag (3'), just a little to remove some existing pressure on the side walls. Cable will be easier to install too.

I have two unresolved questions concerning using cable. One is what size. I have found several formulas on the net on figuring side thrust, but each i use gives different results. Its most likely because I am using the formulas incorrectly. I have spent many hours trying to figure this out. Now i just want answers. Can someone help?

Also, many sites discuss the number of nails/screws/bolts that are needed to fasten wood rafter ties. Using enough fasteners to hold it secure. Now, Im not sure how many that i need to connect the cable to the rafters. One would be easy, more requires a special designed fastener.

Another reason for not using wood rafter ties is that I would need to cut each rafter into three pieces to get them thur the attic access hole. I have a bad back and cable is so much lighter too.

This site formula gave me thrust of 1250 http://www.timbertoolbox.com/Calcs/RafterThrust.htm

this one gave me 18000 http://www.timbertoolbox.com/Calcs/raisedtiethrust.htm

I have used other sites too. I just don't understand all this math.

So, I want to know how strong a cable to buy and if a single attachment point to the rafters is adequate.

14 rafters (not including the gable rafters), 24" oc 2x6

4/12 pitch

room measurement eve to eve 205"

shingle roof

tks much,

pa

[John Kogel]

I am not an engineer and this is no substitute for an engineered design.

The attachment could be simply a hole in the rafter with cable threaded thru it and secured back to itself with at least two cable clamps. The hole should be as close as possible to the outer edge, the sheathed side. If the wood is soft you may need to install brackets instead held with several thru-bolts. At the other end you could bolt the turnbuckle but it will pull sideways, unless you add a horseshoe-shaped strap like a yoke. Or use a short loop of cable and clamps to attach the turnbuckle.

For cable strength, 'airline' cable, the type you might use on an electric winch, could be good, but it will cut in to the wood, unless you lined the hole with a piece of pipe. 'Strawline' is heavier but will not dig in. Too hard to work with.

Cutting cable up in the attic with an angle cutter could start a fire up there. Cut the lengths and rig up the turnbuckles before you go up in the attic.

Check ceiling edges before pulling the walls in. The walls might not want to move and you could splinter a rafter trying. In other words, don't even start if it doesn't give easy. Jacking the ridge with screwjacks would help.

Conventional wood rafter ties can be installed. all you need is a small hole in one of the gable ends. Even a roof vent can be a big enough hole to bring 2X6's in. Wood is less likely to raise an eyebrow in the future and it's easier to work with. Jack up the ridge, nail in the ties, thru-bolt them or tie them to the rafters with plywood gussets on each side and lots of galvanized nails.

[mjr6550]

Your thrust per rafter is 765 pounds. The International Residential Code calls for 7 16d common nails per rafter. When designing with cable you would typically use a factor of safety of 10, so the cable should be rated for an ultimate load of 7650 pounds. The connections are the problem.

[paulmars]

Your thrust per rafter is 765 pounds. The International Residential Code calls for 7 16d common nails per rafter. When designing with cable you would typically use a factor of safety of 10, so the cable should be rated for an ultimate load of 7650 pounds. The connections are the problem.

what is wrong with my math? I never got a value that low. So how many cables would that be? Every rafter?

[kurt]

I like Kogel's idea best. Precut all ties, cut a hole in the darn roof, slide in all the material, then cover the hole with a roof vent....or get nice and feather in shingles and patch like people used to know how to do.

Working it with cable would be more work, more money...imho.

[Erby]

Precut and shove them up through the soffit!

Or, with all the work you're doing, you'll have the roof all torn up anyway. Cut a hole in the roof sheathing just above ceiling level. Precut and shove them through, patch the hole and put your new roof cover on.

[mjr6550]

Your thrust per rafter is 765 pounds. The International Residential Code calls for 7 16d common nails per rafter. When designing with cable you would typically use a factor of safety of 10, so the cable should be rated for an ultimate load of 7650 pounds. The connections are the problem.

what is wrong with my math? I never got a value that low. So how many cables would that be? Every rafter?

I did not look at the formulas you used. I used tables to get the numbers. If you want to use a formula it is:

1/4 times total roof load times the inverse of the roof pitch

In this case: .25x[2feetx205/12x(10+20)]x12/4=769 pounds

The 10 is a typical dead load (so engineers use 12, but for a low slope roof 10 is fine). The 20 is the minimum roof live load. In most parts of the country this would be a snow load of 30 psf or more, but in FL I assume no snow load. 20 psf is the minimum live load per code. The 2 is for 24 inches o.c.

This assumes a tie at every pair of rafters. With relatively low pitch roofs the forces are pretty high. Its hard enough to deal with that force. If you put toes every 4 feet the load would be doubled.

[paulmars]

...769 pounds

so cable rated at 1k every rafter would do?

[Jim Baird]

...math aside, here is a WAG as to what kept roof up over time. (How many block walls have you seen with a bow?) Maybe top course was a lintel one with steel laid in and poured continuous? Would that be stiff enough to resist rafter thrust? May be you don't need no stinkin cables. If ridge does not bow, why bother?

[paulmars]

...How many block walls have you seen with a bow?) ... If ridge does not bow, why bother?

None, but I've only seen 3 wood buildings with sag and bow. All three garages. When a hurricane strikes, they will blow away. I plan on being here and intact after the storm. My ridge bow is about 3".

[mlparham]

Im planning on installing rafter ties in my 1952 cinder block home. It has no rafter ties at all. Every time Im in the attic and look around, I wonder what holds up my roof. The ceiling joists run parallel to the roof ridge. Im also going to install collar ties. Currently there are only two. The roof does not have much sag and the walls have no apparent bow. However, Im adding hurricane straps, supports for the gable ends, and extra fasteners to hold the roof planking to the rafters. Vertical rafter supports and then a new roof cover too. So, I decided to add the rafter and collar ties too.

In my research, I have discovered the idea of using cable instead of wood for the rafter ties. Cable is stronger and will allow me to remove some of the pressure on the side walls, using a turn buckle I can pull up the roof slightly. I have a string pulled 0straight across the ridge inside the attic. I do not plan on trying to take out all the sag (3'), just a little to remove some existing pressure on the side walls. Cable will be easier to install too.

I have two unresolved questions concerning using cable. One is what size. I have found several formulas on the net on figuring side thrust, but each i use gives different results. Its most likely because I am using the formulas incorrectly. I have spent many hours trying to figure this out. Now i just want answers. Can someone help?

Also, many sites discuss the number of nails/screws/bolts that are needed to fasten wood rafter ties. Using enough fasteners to hold it secure. Now, Im not sure how many that i need to connect the cable to the rafters. One would be easy, more requires a special designed fastener.

Another reason for not using wood rafter ties is that I would need to cut each rafter into three pieces to get them thur the attic access hole. I have a bad back and cable is so much lighter too.

This site formula gave me thrust of 1250 http://www.timbertoolbox.com/Calcs/RafterThrust.htm

this one gave me 18000 http://www.timbertoolbox.com/Calcs/raisedtiethrust.htm

I have used other sites too. I just don't understand all this math.

So, I want to know how strong a cable to buy and if a single attachment point to the rafters is adequate.

14 rafters (not including the gable rafters), 24" oc 2x6

4/12 pitch

room measurement eve to eve 205"

shingle roof

tks much,

pa

You are wasting your time. Install rafter ties. Even better hire an engineer to determine if any repairs are even needed

[mjr6550]

...769 pounds

so cable rated at 1k every rafter would do?

Yes, if it is rating for working load. If rated for ultimate load then about 8000 pounds.

[mjr6550]

...math aside, here is a WAG as to what kept roof up over time. (How many block walls have you seen with a bow?) Maybe top course was a lintel one with steel laid in and poured continuous? Would that be stiff enough to resist rafter thrust? May be you don't need no stinkin cables. If ridge does not bow, why bother?

I see plenty of block walls that bow. With 3 inches of ridge sag the walls had to have bowed, or the rafters shifted at the top of the wall.

However, I have seen very few roofs that have the code required nailing at ceiling joists or ties to rafters. Roofs seldom see the full design load and with the roof sheathing in place there is some diaphragm action that is not accounted for in calculations. The load sharing can make a big difference. Lower pitch roofs are less forgiving and tend to sag more, but probably never see much load in Florida.

[mjr6550]

...769 pounds

so cable rated at 1k every rafter would do?

Your bigger problem is how to attach the cables.

[paulmars]

Yes, if it is rating for working load. If rated for ultimate load then about 8000 pounds.

8000 lbs per cable per rafter why?

what is ultimate load?

[paulmars]

Your bigger problem is how to attach the cables.

I agree!

[mjr6550]

Yes, if it is rating for working load. If rated for ultimate load then about 8000 pounds.

8000 lbs per cable per rafter why?

what is ultimate load?

The ultimate load is the load where the cable breaks. You typically use a factor of safety of 10 when using cables. The actual maximum load you should place on the cable would typically be called working load.

[Steven Hockstein]

Im planning on installing rafter ties in my 1952 cinder block home. It has no rafter ties at all. Every time Im in the attic and look around, I wonder what holds up my roof. The ceiling joists run parallel to the roof ridge. Im also going to install collar ties. Currently there are only two. The roof does not have much sag and the walls have no apparent bow. However, Im adding hurricane straps, supports for the gable ends, and extra fasteners to hold the roof planking to the rafters. Vertical rafter supports and then a new roof cover too. So, I decided to add the rafter and collar ties too.

In my research, I have discovered the idea of using cable instead of wood for the rafter ties. Cable is stronger and will allow me to remove some of the pressure on the side walls, using a turn buckle I can pull up the roof slightly. I have a string pulled straight across the ridge inside the attic. I do not plan on trying to take out all the sag (3'), just a little to remove some existing pressure on the side walls. Cable will be easier to install too.

I have two unresolved questions concerning using cable. One is what size. I have found several formulas on the net on figuring side thrust, but each i use gives different results. Its most likely because I am using the formulas incorrectly. I have spent many hours trying to figure this out. Now i just want answers. Can someone help?

Also, many sites discuss the number of nails/screws/bolts that are needed to fasten wood rafter ties. Using enough fasteners to hold it secure. Now, Im not sure how many that i need to connect the cable to the rafters. One would be easy, more requires a special designed fastener.

Another reason for not using wood rafter ties is that I would need to cut each rafter into three pieces to get them thur the attic access hole. I have a bad back and cable is so much lighter too.

This site formula gave me thrust of 1250 http://www.timbertoolbox.com/Calcs/RafterThrust.htm

this one gave me 18000 http://www.timbertoolbox.com/Calcs/raisedtiethrust.htm

I have used other sites too. I just don't understand all this math.

So, I want to know how strong a cable to buy and if a single attachment point to the rafters is adequate.

14 rafters (not including the gable rafters), 24" oc 2x6

4/12 pitch

room measurement eve to eve 205"

shingle roof

tks much,

pa

The house is still standing since 1952 without much sag and the walls are not bowed. Personally, I would spend the money on a vacation.

[Jim43]

Wood rafter ties will carry no load (will be useless- do NO GOOD) UNLESS you have some means to actually load the ties in tension by first jacking up the ridge 4 inches before you attach the rafter ties to the rafters, or otherwise move the bottom of the rafters toward the center of the house 1/2 inch before attaching the rafter ties. (Do NOT do EITHER one because you will destroy the house trying to improve it.)

1. Aircraft cable with steel/ stainless steel turnbuckles is a good idea (I plan to do this to my garage with a 26' unsupported ridge and 2" sag with a similar pitch; my WOOD walls are bowed 1" on one side and 3/4" on other side, measured by stretching a string from end to end).

2. Heavier cable (2000 to 8000 pound working load) than you thought you needed is necessary because the lighter (5/32" or 3/16") cable will stretch (experience) so much, your turnbuckles will run out of travel before you do any good.

3. Take some careful measurements BEFORE you start, so you know EXACTLY the distance between the walls, and exactly the distance between known points on the rafters (distance from lower end of rafter on one side of house to lower end on other side- drive thin screws or nails perpendicular into side of rafter so measurements will be precise and "repeatable"- WRITE the measurement ON each rafter). Just a wild GUESS, but this distance MAY decrease 1/4" when cable is fully tensioned. Cable when tensioned may sound like a very low bass fiddle when thumped; but NOT like a guitar. Remember that you have more than one cable, so 500 pounds of tension on each cable times 20 rafters would be 10,000 pounds of inward force which sounds like a lot- maybe too much, or much too much. Re-calculate tension per cable assuming just the weight of 16" or 24" linear length of roof and shingles PER RAFTER. Too little tension would be much better than too much tension. Maybe you only need a cable every other rafter.

4. For attaching the cable to the rafters, you might consider drilling a perpendicular hole approximately 1 inch in diameter (barely big enough for your iron) thru the rafter (as close to the outer end of the rafter and as close to the roof as is practical) and inserting a 5 inch long piece of iron (1/2 inch steel water pipe is cheap). Form a loop in the end of the cable with a cable clamp, and loop the cable up and outward over the pipe on BOTH sides of the rafter. You will need to protect the bottom of rafter where the cable passes under the rafter- I suggest the metal plates from the electrical department at Home Depot/ Lowes that are used to protect electrical wiring in walls. $.40 each. Ask in the store. http://www.homedepot.com/p/Halex-1-1-2- ... /100151742

4. Get GOOD (steel/ stainless steel) turnbuckles that will carry twice the load needed/ cable about 10 times the load). Tighten the turnbuckles to take ALL the side thrust off the walls (this will be an estimate on your part; the walls MAY move SLIGHTLY toward the center of the house when you relieve the outward side thrust load, which is OK, but do not PULL the walls together. If you tighten more than just taking the side load off the walls, you run the risk of cracking the walls, and reducing the strength of the walls. Just relieve the outward thrust, and no more. This MAY remove perhaps 1/4 of the ridge sag. If you remove all the ridge sag, you will damage the outside walls doing that. What you are trying to do is relieve all, OR MOST, of the side load on the walls, so it MAINTAINS the current position. Wood walls you could pull back a little; block walls you will damage if you try to pull them back.

5. Do this over time: relieve 40% side load initially, 10% more each week for 6 weeks.

In a year, you could check your MEASUREMENTS, and give the turnbuckles one or two more turns as needed to maintain the planned distance.

6. The goal is to relieve the pressure, so it does not get worse; NOT to pull the house back together. Good Luck, and tell us how it came out.

7. this link is too much unrelated information about unevenly loaded masonry basement walls, but may explain why you do not want inward, OR outward pressure, on your block walls- the load should be only straight down.

http://ezinearticles.com/?Are-Your-Base ... id=1519592 .

Jim43

[mjr6550]

Jim43-I'm generally not in favor of cables because they stretch. However, once you tension they cables they carry load. If you use collar ties they also carry load if the walls try to move outward. Jacking up the ridge 6 inches will pull the walls inward.

[Chad Fabry]

I wonder how on Earth I managed to cable so many buildings and push the ridges back in place with all the crude tools I've been using. I had no idea pulling two walls together required so much effort and thought.

[paulmars]

I wonder how on Earth I managed to cable so many buildings and push the ridges back in place with all the crude tools I've been using. I had no idea pulling two walls together required so much effort and thought.

hey Chad, u gunna visit florida anytime soon?

p

[Les]

Jim43,

I find much of what you are saying of not much value. your ezine cite is by a less than credible person. As a retired engineer you should have a basic knowledge of loads and forces that are applicable to residential structures and the materials used therein.

[Jim43]

MJR6550,

Wood walls would pull in. Cinder block walls will be damaged if completely PULLED in.

There is no way to know from behind our computers how firmly the roof is attached to the CINDER BLOCK walls, or what kind of cracks are already in the mortar joints; but the roof may be very securely attached. Even with ONLY a friction "attachment", the CINDER BLOCK walls will be damaged if you try to completely pull in the walls that took 5o years to get where there are. The house was built improperly (ceiling joist direction versus ridge direction). There are too many unknowns to RISK anything but taking the outward pressure off the walls, which will allow the walls to straighten up somewhat, and is a very good thing to do (and the house will last another 50 years). Keep It Simple, and safe.

(I did not take the time to do the calculations for his house. For my 2 story garage I will just pull in the WOOD/ Vinyl walls until they are straight and won't need to know if I am using 1000 pounds or 2000 pounds each of only 4 cables across the knotty pine paneled great room. My real problem is one end of my 26 foot long unsupported "ridge" beam is "supported"/built over a 6' doorwall with NO header over the doorwall. The doorwall/garage needs a 18 foot long, double 2x12 header over the doorwall, to span the garage door that is below on the first floor of the garage. Much more involved than just pulling in 2 walls with cables.) Jim43

[mjr6550]

MJR6550,

Wood walls would pull in. Cinder block walls will be damaged if completely PULLED in.

There is no way to know from behind our computers how firmly the roof is attached to the CINDER BLOCK walls, or what kind of cracks are already in the mortar joints; but the roof may be very securely attached. Even with ONLY a friction "attachment", the CINDER BLOCK walls will be damaged if you try to completely pull in the walls that took 5o years to get where there are. The house was built improperly (ceiling joist direction versus ridge direction). There are too many unknowns to RISK anything but taking the outward pressure off the walls, which will allow the walls to straighten up somewhat, and is a very good thing to do (and the house will last another 50 years). Keep It Simple, and safe.

(I did not take the time to do the calculations for his house. For my 2 story garage I will just pull in the WOOD/ Vinyl walls until they are straight and won't need to know if I am using 1000 pounds or 2000 pounds each of only 4 cables across the knotty pine paneled great room. My real problem is one end of my 26 foot long unsupported "ridge" beam is "supported"/built over a 6' doorwall with NO header over the doorwall. The doorwall/garage needs a 18 foot long, double 2x12 header over the doorwall, to span the garage door that is below on the first floor of the garage. Much more involved than just pulling in 2 walls with cables.) Jim43

I'm not sure what you are talking about. I did not sugest pulling in the walls. I did point out that raising the ridge, which you suggested, would result in the walls moving in.