Login | Register

Previous in Forum: Electro Magnetic (EM) sensor for cable stayed bridge   Next in Forum: How to Calculate the Load for a Small Bridge
Close

Comments Format:






Close

Subscribe to Discussion:

CR4 allows you to "subscribe" to a discussion
so that you can be notified of new comments to
the discussion via email.

Close

Rating Vote:







23 comments
Power-User
United States - Member - New Member

Join Date: May 2008
Location: Arizona
Posts: 225
Good Answers: 1

Collar Tie Design

05/17/2009 10:42 AM

My collar tie design software limits the dimensions of my collar ties to 1.5 x 7.25". My client would like to use 2x8 R.S.

The rafters are 2x8, 16" oc, and 10.6ft horizontally from cl of ridge beam to inside of wall. Roof slope is 6/12. The collar ties are 48" oc. The software thinks the collar ties are on every rafter. From the peak of the ridge vertically downward to the cl of the collar ties is 3.1 ft. Another 2.87 ft vertically downward gets you to the elevation of the top of the walls.

There are no joists. The collar ties (& the sheathing) are the only thing holding the structure together. The software says it won't work with the 1.5x7.25 Ties. The Moment is the controlling factor. I'm thinking if I could limit the forces on the Ties to tension only by using a single bolt at each end, thus forming a pinned, or hinged joint, I could design for tension only, and analyze the collar ties the client wants to use. But how would you make sure the joint was allowed to rotate? Put a washer between the Tie and the rafter, with instructions to snug up, but not over-tighten?

Send to a friend Digg this Add to del.icio.us
Interested in this topic? By joining CR4 you can "subscribe" to
this discussion and receive notification when new comments are added.

Comments rated to be "almost" Good Answers:

Check out these comments that don't yet have enough votes to be "official" good answers and, if you agree with them, rate them!
Power-User
United States - Member - Member Engineering Fields - Mechanical Engineering - Member

Join Date: Mar 2009
Location: Sarasota, FL
Posts: 119
Good Answers: 1
#1

Re: Collar Tie Design

05/17/2009 6:46 PM

I'm not even sure what You are building from Your description...

My first thought was a roof for a porch, or pole barn, but why would You want connections that rotate. Wouldn't it be difficult to make a fold away temporary use structure with 2 X 8 lumber?

More dimensions are needed to determine what kind of structure You are designing, pictures also do wonders.

__________________
"Don't let those telling You, it can't be Done, stop you from Doing IT", Albert Einstien...
Score 1 for Off Topic
Power-User
United States - Member - New Member

Join Date: May 2008
Location: Arizona
Posts: 225
Good Answers: 1
#4
In reply to #1

Re: Collar Tie Design

05/18/2009 7:39 AM

Yes, it is a porch roof.

Because classical mechanics views the assembly as an inderterminate one. Eliminating the moment, by providing a pinned connection, simplifies the solution.

Guru
Popular Science - Cosmology - New Member

Join Date: Feb 2009
Location: Halcottsville, NY
Posts: 623
Good Answers: 16
#2

Re: Collar Tie Design

05/17/2009 8:01 PM

You will need to put your collar ties at 32" oc., with glued and screwed 3/4" plywood gussets. Because of this you will be best with 'nominal' sized lumber, butted at an angle and gusseted both sides. This will be in tension. The alternate is to add the collar ties after the rafters are in place, lapped and glued, using 16D common nails (screws do not have enough sheer strength). Use the roof pitch to cut the ends of the ties for maximum surface and nailing. 2X6 nominal lumber will be sufficient. You must support the ridge beam until the ties are in place and the glue is set, otherwise the roof will collapse. If the ceiling is to be finished, you might as well use 2x4s at each rafter, same method.

Good luck

__________________
De gustibus non est dispudandum.
Power-User
United States - Member - New Member

Join Date: May 2008
Location: Arizona
Posts: 225
Good Answers: 1
#5
In reply to #2

Re: Collar Tie Design

05/18/2009 7:44 AM

Good morning Tippycanoe.

I believe the client would sooner opt for the more rustic look of steel gussets. Other than my drawing up a custom design of say 3/16" plate, do you have any comment?

There will be no ceiling nor joists.

Guru
Popular Science - Cosmology - New Member

Join Date: Feb 2009
Location: Halcottsville, NY
Posts: 623
Good Answers: 16
#15
In reply to #5

Re: Collar Tie Design

05/18/2009 9:56 PM

3/16" (?) I've used 1/4" plate gussets with rough sawn on a significantly bigger roof. They are unobtrusive, but look 'authentic', and within context when you do see them. Use square headed bolts, washers, and nuts if you can get them. Distress the edges, apply stove black, and you have 'the look'. Yours should be sandwitched between the rafter and tie.

Regards, Carl

__________________
De gustibus non est dispudandum.
Power-User
United States - Member - New Member

Join Date: May 2008
Location: Arizona
Posts: 225
Good Answers: 1
#18
In reply to #15

Re: Collar Tie Design

05/19/2009 8:03 AM

Good morning Carl.

Excellent suggestions. However, I could not get the assembly to calc out unless I went for 4x8 rafters at the collar ties. Since this produces surfaces of the rafter and collar tie in different planes, I cannot use gussets.

I will recommend the use of black square headed bolts and nuts tho.

Guru
Popular Science - Cosmology - New Member

Join Date: Feb 2009
Location: Halcottsville, NY
Posts: 623
Good Answers: 16
#19
In reply to #18

Re: Collar Tie Design

05/19/2009 8:57 AM

Hello FH,

Yeah, sometimes you have to 'trick' a cad program. I save what I have, copy, and rework the calcs so the program 'thinks' contact is made. Print a copy of the new calcs, so the client and building inspector know what you've done.

Actually, If you're sure the moisture content of you're lumber is below 14%, the joints could be epoxied, eliminating the gusset. Faying surfaces need to be freshened with coarse sandpaper. The joint then becomes stronger than the wood.

Stay with it. It's part of the fun!

__________________
De gustibus non est dispudandum.
Guru

Join Date: Nov 2007
Location: Sherwood Park, Alberta, Canada
Posts: 886
Good Answers: 40
#3

Re: Collar Tie Design

05/17/2009 9:39 PM

flyinghigh,

Can you provide a cross section of the rafters and collar ties. If you have ceiling joists, please indicate.

I have found that the term "collar tie" as used by framers is frequently incorrect. In many instances, the collar tie is a strut (taking compression). Give us a sketch if you don't mind.

__________________
Bruce
Power-User
United States - Member - New Member

Join Date: May 2008
Location: Arizona
Posts: 225
Good Answers: 1
#6
In reply to #3

Re: Collar Tie Design

05/18/2009 8:04 AM

Good morning ba/ael.

I have not yet figured out how to make a drawing on CR4. A collar tie is a miniature joist, usually placed in addition to joists, and aligned horizontally connecting 2 rafters, just under the ridge beam.

So if you look at Cyberfool's sketch, with the rafter notched for the wall plate, and in your mind add a horizontal collar tie halfway vertically between the ridge beam and the top plate, without joists, you'll have the design the customer wants.

In this design, without joists, the collar ties are taking all the load, preventing the rafters from splaying out flat; therefore they are in tension, and some moment.

Power-User
United States - Member - Member Engineering Fields - Mechanical Engineering - Member

Join Date: Mar 2009
Location: Sarasota, FL
Posts: 119
Good Answers: 1
#13
In reply to #6

Re: Collar Tie Design

05/18/2009 9:12 PM

Quote: I have not yet figured out how to make a drawing on CR4.

I used my "Paint" program on my Computer to make the drawing, to insert it, click on the thing that looks like a small Camera in the message editing window, and search for where you put the file.

Quote: So if you look at Cyberfool's sketch, with the rafter notched for the wall plate, and in your mind add a horizontal collar tie halfway vertically between the ridge beam and the top plate, without joists, you'll have the design the customer wants.

Like this kind of bracket?

I just wish CR4 didn't shrink my drawings, oh sure they allow you to resize them up, but when you make them bigger, they get fuzzy.

__________________
"Don't let those telling You, it can't be Done, stop you from Doing IT", Albert Einstien...
Power-User
United States - Member - New Member

Join Date: May 2008
Location: Arizona
Posts: 225
Good Answers: 1
#17
In reply to #13

Re: Collar Tie Design

05/19/2009 7:51 AM

Good morning Cyberfool. Thx. for the tip. I will try that the next time.

Guru

Join Date: Nov 2007
Location: Sherwood Park, Alberta, Canada
Posts: 886
Good Answers: 40
#7

Re: Collar Tie Design

05/18/2009 10:55 AM

The moment is the controlling factor. Not the moment in the collar tie, but the moment in the tied rafter. Say the uniform roof load is w psf and the overall span of roof is L. The simple span moment is wL2/8. If you put a collar tie every foot, the tension in the tie would be wL2/(8*3.1). Since you want to use a collar tie every 4', the tension in the collar tie will be 4wL2/(8*3.1) = wL2/6.2.

Assuming w as 15 psf dead load plus 25 psf live load = 40 psf and L as 2*10.6 = 21.2, the tension in the tie will be 40(21.2)2/6.2 = 2900# (a horizontal force).

The bending moment in each tied rafter is Pab/L = 2900*3.1*2.9/6.0 = 4345'#.

The section modulus of the rafter is 1.5*7.252/6 = 13.14 in3. The maximum fiber stress is 4345*12/13.14 = 3970 psi. That is too much, so the program is telling you the rafter is overstressed in bending.

__________________
Bruce
Score 1 for Good Answer
Power-User
United States - Member - New Member

Join Date: May 2008
Location: Arizona
Posts: 225
Good Answers: 1
#8
In reply to #7

Re: Collar Tie Design

05/18/2009 6:06 PM

Thx Bruce, GA.

I followed everything you detailed. I just want to confirm that in your 3rd paragraph, about the Moment in the tied rafter, the ratio 2.9/6 is the fraction of the maximum moment that occurs at the midpoint of the 20.2' L, based on the vertical distance from the wall top plate to the point at which the collar tie is exerting its force, iow, 2.9 ft up of a total distance of 6 ft. Correct?

Thx again.

Guru

Join Date: Nov 2007
Location: Sherwood Park, Alberta, Canada
Posts: 886
Good Answers: 40
#9
In reply to #8

Re: Collar Tie Design

05/18/2009 7:45 PM

flying high,

Consider a beam 6' long with a load P acting 2.9' from one end (and hence 3.1' from the other). The moment is Pab/L or P*2.9*3.1/6.0. In this case, the rafter is longer than 6', in fact it has a length of about 12.2'. Because the force is horizontal, we don't have to consider the sloping length of rafter, just the vertical length. Neglecting the 16" width of load on the tied rafter, the moment in the rafter varies from 0 at the top of wall to maximum at the tie, then back to 0 at the ridge line.

__________________
Bruce
Power-User
United States - Member - New Member

Join Date: May 2008
Location: Arizona
Posts: 225
Good Answers: 1
#10
In reply to #9

Re: Collar Tie Design

05/18/2009 7:56 PM

Got it. Thx for your expertise.

Guru

Join Date: Nov 2007
Location: Sherwood Park, Alberta, Canada
Posts: 886
Good Answers: 40
#11
In reply to #10

Re: Collar Tie Design

05/18/2009 8:16 PM

flyinghigh,

You're welcome. How are things in Arizona? I just heard on the TV that we are going to get 12 cm of snow tomorrow. What a ripoff...and just when everything was greening up.

__________________
Bruce
Power-User
United States - Member - Member Engineering Fields - Mechanical Engineering - Member

Join Date: Mar 2009
Location: Sarasota, FL
Posts: 119
Good Answers: 1
#12
In reply to #11

Re: Collar Tie Design

05/18/2009 8:49 PM

It could be worse, You could be getting 12 inches.

__________________
"Don't let those telling You, it can't be Done, stop you from Doing IT", Albert Einstien...
Guru

Join Date: Nov 2007
Location: Sherwood Park, Alberta, Canada
Posts: 886
Good Answers: 40
#14
In reply to #12

Re: Collar Tie Design

05/18/2009 9:12 PM

Can't argue with that.

__________________
Bruce
Power-User
United States - Member - New Member

Join Date: May 2008
Location: Arizona
Posts: 225
Good Answers: 1
#16
In reply to #11

Re: Collar Tie Design

05/19/2009 7:47 AM

Good morning Bruce.

It's starting to warm up, temps in the 80's up here in the mountains of N. Az, but in the valley (Phx) its over 100ยบ F. I'm jealous of your snow.

May I pick your brain a little more? The software now says the collar tie tension is 1970#, and that each end connection needs 13 ea. 16d nails.

My reference book says ea. nail will hold 107#, for a total of 1390#, < 1970#, :.NG.

In investigating a possible bolted connection, my book says for single shear, bolt "length" 3.5", tension parallel to grain, I would need a 1" dia. bolt.

Then computing the length of the shear plane (Lh) required using Vc=n*fh*t*Lh, where Vc = 1970#, n = 1 bolt, fh = safe shear stress ║ to grain (95 psi), t = thk of collar, I get a length for a single bolt longer than is possible.

Therefore, I equated the area of the 1" dia. bolt (0.79 si) to 2 ea. 3/4" dia bolts (2 * 0.44 = 0.88 si > 0.79 si). Using the above figures and t = 2" (he's using R.S. [Rough Sawn]), produces a Lh of 5.2", or 2.6" per bolt; which would be achievable in the amount of space provided.

In my low-experience this still seems like over-kill, but numbers don't lie, do they? <g>.

Off Topic (Score 5)
Guru

Join Date: Nov 2007
Location: Sherwood Park, Alberta, Canada
Posts: 886
Good Answers: 40
#20
In reply to #16

Re: Collar Tie Design

05/19/2009 11:21 AM

The capacity of nails varies with the species of wood and the size of nail used. If a nail is clinched, it holds much more than one which is not clinched, so I am not sure how you arrived at 107# per nail.

A 1" dia. bolt will not come close to resisting 1970#. Comparing 3/4" bolts on the basis of bolt area has no meaning at all. The capacity of a bolt is determined by the bearing stress on the wood, not by the shear stress in the steel. The bearing stress in the collar tie is parallel to the grain, but in the rafter it is at an angle to the grain, so you must take the worst case, namely the slope of the grain in the rafter. Also, you must observe the various end and edge distances for both pieces of wood. To do that, you need to draw the overlapping joint to scale and plan where the bolts are positioned.

I don't believe you can transfer that much load using bolts in single shear. I think you will need to align the tie with the rafter and use steel side plates with bolts in double shear. Also, make sure you check the bending stress in the rafter at the tie point. That tie force still sounds pretty high.

__________________
Bruce
Off Topic (Score 5)
Power-User
United States - Member - New Member

Join Date: May 2008
Location: Arizona
Posts: 225
Good Answers: 1
#21
In reply to #20

Re: Collar Tie Design

05/19/2009 3:08 PM

The software calculated a 1970# tensile load on the collar tie. I checked this figure using your method, and it checks out closely.

The software also specified 13 each 16d nails. In my Architectural Engineering Design book for DF it shows 107#/nail, or for DF #1, * 1.2, for a figure of 128#. Even with this figure, for 13 nails, the total load capacity is 1670#< 1970# :. NG.

Since the collar is at least 1.5" thk, and maybe closer to 2" (since it is R.S.), and the rafter is 3.5, for a total of 5", clinching is out of the question.

In the same book for bolted wooden connections, it prescribes a method of calculating the length of shear plane (Lh) required. For a single bolt, this length was longer than that available. Using two bolts however, allowed enough room for the two shorter Lh's.

In the process above I did make a drawing to scale, to make sure it was feasible. If your computer can decipher AC13, I could attempt to send it.

Off Topic (Score 5)
Guru

Join Date: Nov 2007
Location: Sherwood Park, Alberta, Canada
Posts: 886
Good Answers: 40
#22
In reply to #21

Re: Collar Tie Design

05/19/2009 3:55 PM

I didn't realize your rafters were 3.5" thick. I thought they were 1.5" thick. Clinching is out of the question. Would there be any merit in using a two ply collar tie, one ply each side of the rafter? That way the bolts would be in double shear instead of single shear. Don't know. It's hard to say without a picture.

Sorry, I don't know what AC13 is but I'm pretty sure my computer can't decipher it.

__________________
Bruce
Off Topic (Score 4)
Guru
Popular Science - Cosmology - New Member

Join Date: Feb 2009
Location: Halcottsville, NY
Posts: 623
Good Answers: 16
#23
In reply to #22

Re: Collar Tie Design

05/20/2009 1:33 AM

GA ba/ael, I missed that myself. Not at all off topic. The double ties with some construction adhesive would spread the nailing schedual, so that it wouldn't look like the shoemakers kid did it.

__________________
De gustibus non est dispudandum.
Off Topic (Score 5)
23 comments
Interested in this topic? By joining CR4 you can "subscribe" to
this discussion and receive notification when new comments are added.

Comments rated to be "almost" Good Answers:

Check out these comments that don't yet have enough votes to be "official" good answers and, if you agree with them, rate them!
Copy to Clipboard

Users who posted comments:

ba/ael (7), Cyberfool (3), flyinghigh (9), Tippycanoe (4)

Previous in Forum: Electro Magnetic (EM) sensor for cable stayed bridge   Next in Forum: How to Calculate the Load for a Small Bridge
You might be interested in: Cable Ties, Slewing Rings and Turntable Bearings, Lockbolts