CountryPlans Forum

Can the balloon walls be made higher than 10ft ?  No snow load.

Yes, if you are talking about the 20' wide 1 1/2 story plans. The 10' studs can be increased to 12' giving 2' more height in the loft sidewall. This makes the usable space about 4' wider.

If I remember right, John previously pointed out that you can not use 12 foot walls with a cathedral ceiling room. You have to have a loft when you go to 12 feet. Or mabe you can get by with coller ties if thay are in the lower 1/3 of the rafters.

I believe that a critical factor is whether you using a ridge board or a ridge beam supported properly by the gable ends---a ridge beam allows you to build a cathedral ceiling---ridge board design would require collar ties in the lower 1/3 to counter the roof thrust forces.

If the building is 25 ft wide would that changes things appreciably? 

I'm not opposed to a ridge beam, but would prefer to avoid the cost if possible.

The issue Bob pointed out is correct. Unsupported wall height is limited to 10' in cathedral areas and taller walls require an engineering review if they are longer than 10' (such as a 12' high wall in a 14' wide living room). I understand this is to be a separate issue from how you support the roof load. The concern is lateral loading on tall walls.

Thus if you do a full length loft all the walls are braced by the floor system and the lower floor could be 8' or 9' or even 10' tall, but you couldn't have the full 12' tall wall open in the cathedral living room. This complication was one of the main reasons I didn't design the walls with 12' studs to begin with - I wanted the flexibility to have open cathedral spaces where the owner desired.

Here's the earlier link:
http://countryplans.com/smf/index.php?topic=8839.0

A cathedral room without ties at the plate or in the lower third of the height from plate to ridge requires a ridge beam.

At 25' wide and with no other information given you aren't understanding the physics of it yet. What is either tieing or supporting your rafters?.

Table R602.3.1 in the codebook lists allowable stud heights, read the exceptions under the chart carefully.

Sorry, you're right I haven't provided enough info.

No cathedral at all.  Full second story.

I like the 1 1/2 story plans, with a full second floor.  But I think the 2ft knee wall is too short for my tastes. 

If I use 12ft balloon walls, tied together with floor joists the entire length (at ~8ft high) and collar ties at the ridge, would that work with a 25ft wide building?

When you take a 20' wide building designed to have full span joists and make it 25' wide you need to redesign the floor system usually with an interior bearing beam or wall that supports both floors and carries the load to girders and posts at the interior of the foundation.

This is beyond the type of help you can get from this forum and would involve working with someone who can trace the loads and do a structural analysis on your revised design. Depending on how you layout the floors and walls you can place the support posts appropriately and then determine the spans and sizes of the beams.

At the same time you could plan for a taller sidewall at the upper level, a new truss for the wider roof, and a full sized stairway to the now larger upper level (this stair comes in the booklet with the 1-1/2 story plans).

Well, I guess that rules out going 25ft wide  ;)

What I'm trying to decide is whether to go with the "2 story Universal" plan or the "20 X 30 1.5 story"  plan. 

I could live with the 1.5 story if I could make the balloon walls 12 ft high, thereby increasing the knee wall height to ~ 4ft.  If that's not possible the I'll go with the 2 story Universal.  It seems like the 12ft balloon wall is possible with no loft (or with a loft and rafter ties).

You could do the balloon framed 12' walls on the 1-1/2 story and have the extra height in the loft area. But you would need to do the loft for the length of the building - no open cathedral area over the living room. Since the loft is now a larger space, I would suggest trading the steeper cottage stair for the full sized "U" stair.

In many ways this is what the 2-story Universal already has, except the stair is already designed in, the 2nd floor walls are full height, the roof can be trussed and both floors can be platform framed. The Universal is actually simpler construction and probably a bit less costly per usable SF than the 1-1/2 story. But, it doesn't have the "tucked under the roof" feel of the 1-1/2 story and looks taller.

What about...
12' walls with timbers on 4' centers at 8' high and 2x6 car decking on top.
Could you run the 20' long timbers all the way- the full 30 feet length of the building, but hold the car decking back to create the cathedral ceiling?  It would be the timber that was acting as the collar tie, right?  You would be left with at least one timber hanging out in space but the loft would be open to the main floor- somewhat.  Any thoughts?   

Prescriptively (no engineering) rafter ties are required on every pair of rafters in the lower third of roof height, these prevent the roof from splaying. Collar ties are also required in the upper third or across the top at a maximum of 4' on center. These prevent the roof from pulling apart at the ridge in a storm. So both are required, both do different things. Kneewalls do not change the codebook prescriptive requirement of a tie across each rafter pair in the lower third of roof height unless checked by an engineer. I drew it one time and IIRC made it work with a 5' kneewall on a 20'wide with rafter ties at 8'

As the building gets wider the kneewalls can be in the range you desire and the rafter ties will be in the lower third @ 8' off the second floor. There are clearspan floor trusses that can span this width without center support.

With engineering confirmation what you propose will work in many locations SD.

Then there is the structural ridge option.

Stairs and heights start working well in 24' wide buildings IMO... why are we limiting ourselves?

I went with the 12' walls (ballooned framed) with 4x12 timbers set at 8 feet (the bottom of the timber).  It gave me about 3.5 feet upstairs until the roof started.  My cabin is only 18 feet wide.

(https://i275.photobucket.com/albums/jj317/kubes14001/DSC02919.jpg)

Here is an informative piece from an international association of home inspectors:

http://www.nachi.org/collar-rafter-ties.htm

(http://www.nachi.org/images08/framing.jpg)

Note that they support using rafter ties on every other rafter pair not every one. This has been my understanding and what I show in the open beam cathedral ceiling details on my plans. These have been successfully built in many code enforced areas.

Now I'm a little confused-----I built a 24x24 1 1/2 story with a 10x24 loft at the rear with a 10x24 double-decker porch on the front. The roof extends over the decks. There is a 14' open area at the  front with a 4x14 "bridge" crossing to the upper covered deck. My side walls are 12' 2x6 studs with a glulam supporting the loft. This results in a 3' knee wall in the loft area and a 10x19 usable loft area. Roof is 12/12 pitch with a 6x16x36' ridge beam supported at gable ends. Rafters are 2x12x22' which gave me very wide eaves (5' on the run) to dump snow 4" from the cabin walls----wind loads are fairly light and no tornadoes here so I thought I could get away with the big overhangs.  Completely clear cathedral ceiling  with no collar ties-----should I be worried about stresses on the walls? I did run by my plans with a structural engineer friend but no formal analysis was done or required. 

While we can't give specific engineering advice here, having a ridge beam support 1/2 the roof load does away with the need for collar ties. You deep rafters should be fine for the wide overhangs. If you have had a trusted engineer look at the project you are probably on solid ground.

You might ask him about the open area at the front. If I'm visualizing it right you could, if needed, tie the walls together with a tension member through the bridge if there is a concern.

Be very careful with the terms used to describe these parts, try to use the correct terms to avoid confusion.

A collar tie is always required. It is usually under the ridge as in John M's pic, but can be anywhere in the upper third as in John's pic, or can be a metal strap running from one rafter's top edge, over the ridge and nailed along the top of the opposite rafter. A very common failure, especially with improperly installed gable trusses, the gable blows in and the roof zippers open along the ridge. Two roof diaphragms are seen leaving the scene. Garage doors blowing in and lifting the roof from the inside is another way you see this failure. Collar ties are required at least every 4'.  A collar tie does not resist rafter horizontal thrust, it is tieing the rafter tops together, it is too high up to effectively keep the rafter feet from spreading.

A rafter tie is required unless a structural ridge (or ridge beam is the term in the code) is used. The rafter tie is prescriptively required to be in the lower third of height and on every rafter pair. It ties the rafter feet together across the building to prevent them from wanting to slide outward. Sure one can step outside of prescriptive if they know what they are doing, with an engineers approval, and with good reason.

If we go every 4' with the rafter ties, the connection stress doubles, raise the ties and it about doubles again. Raised ties also put an additional bending load into the rafter. That is the reason for the reduced spans or increased depths in the code tables for raised rafter ties. By skipping every other rafter this bending load is also doubled on the tied rafters.

Go higher than the code limit and the stress begins multiplying those numbers fast. Much like the working load vs the ultimate strength on a rope the allowable stresses are not the ultimate stresses wood can withstand. When we step outside of codebook prescriptions without a full understanding of why those things are the way they are, we are into our safety an unknown amount and are using a highly variable material.

The code reference and the inspections reference describe codes that are outdated. The current building code is readily available online and is the minimum standard.

I had this discussion with a local designer less than a month ago. The ties were spaced and in the upper third. I suggested a structural ridge and suggested a size... I didn't know what I was talking about but he humored me since I wasn't going to do it his way. His engineer came up with the same ridge beam I did. The "correct" way to do this is to tie it as per code, use a structural ridge (the supplier will engineer it for free), or have it checked by an engineer.

Well, Don is right! The code wording has changed between the 2003 IRC and the 2006. Rafter ties in the bottom 1/3 of the rafter were allowed at 4' o/c in the 2003 code but in areas where the 2006 code is adopted they must now be on every rafter pair.


(https://i1009.photobucket.com/albums/af219/countryplans/roof_ceiling.jpg) (http://www.awc.org/helpoutreach/ecourses/MAT108/MAT108eCourseV09-2007.pdf) click to go to the PDF article.

If the 2006 code has been adopted in your area you may not be able to use the simple framed cathedral detail I have shown in my plans where a double rafter tie is exposed at 4' o/c and ties the walls together. If you have to tie every rafter pair then you cannot effectively have a cathedral ceiling using standard framing. You will need to go to the much more expensive and difficult job of installing and supporting a ridge beam. Note that raising the rafter tie off the plate and as high as 1/3 the rafter length is allowed and will produce a coffered style ceiling that exposes part of the roof slope.

Please understand that if you have built a roof in a manner that would have been approved in 2003 (or 1973!) it does not mean that this roof will be flying off the walls the next time the wind blows! What is happening is that codes are continually being revised to include ways to build buildings that address more and more unlikely fringe events. This is good and results in continually but marginally safer buildings. Are there older homes in your area? Perhaps you are living in one. Those houses have hundreds, perhaps thousands of details that are in violation of the most current codes. Wood frame construction using sticks and panels attached with wire nails is time proven to be strong, flexible and forgiving and yet it is constantly being improved. Be assured, whatever you are building now it will soon be out of date - that is inevitable.

Thanks John,
One way to look at the cost of the ridgebeam is to deduct the cost of the rafter ties that it replaces. And possibly the cost of an engineer. They can be a pain to lift, however the design is driving the need. We've used machines when available but have also set them using pump jacks, towers of scaffold, gin poles, and the armstrong method.

A ridgebeam makes the kneewall thrust issue disappear as well. If the ridge cannot drop the rafter feet don't try to spread. It is an easy structural solution that satisfies code.

Don:

Your points are well taken. However every time a code revision disallows a traditional framing method (in this case one that has been used for over 100 years) it pushes the home building process further out of reach of an owner builder and the small crew builder, especially those in remote areas.

Traditional small dimension wood framing was what turned every farmer into a builder and made housing affordable for millions of families that settled the West. Going to ridge beams for roof support (as I do with the Victoria cottage) locks the design of the supports and freezes the owner from making free interior layout changes. Such a design has little flexibility and changes will now require redesign and likely re-engineering all the way down to the foundation where new footings will be needed.

Of course there is no one to speak for these owner-builders when codes are being revised and they are a quickly dying out at any rate as building becomes a Pro-only specialized occupation. Who other than Pros could hope to keep up with the constant rule changes?

OK, enough of my self-serving complaining. :P

I would love to know how many and under what conditions roofs actually failed because rafter ties were 48" o/c rather than 24".

Let me add this additional photo, it is a little later on in the process.  I meant to add it first but must have clicked the other photo.  It shows my finished ceiling height of 8 feet.  So I guess I have rafter ties AND collar ties?

(https://i275.photobucket.com/albums/jj317/kubes14001/DSC03341.jpg)

thanks guys, I've learned a lot from this thread.

John_M

While there may be several ways an inspector might find fault with this, let's do a thought experiment.

Do you have a metal framing anchor at the rafter to wall connection? You should and they should hold the rafter against outward forces.

Do you have a metal strap or gusset or wood collar at the top of the rafters to keep them working as a unit along the ridge board? You should.

Then you have your deep rafters on either side tied together and triangulated by the ceiling rafter/tie. This is making the rafters work together a bit like a stiff legged truss.

And all of these elements are further united into a diaphragm with the nails and roof sheathing. These parts all work together to share the loads from side to side and between the rafter sets.

Hello all,

Noob here.  I've been lurking for 6 months or so.  Bought the 1.5 story plans.  No land yet, just evolving plans and ideas, and a monthly deposit into the cabin fund.

First, thanks for the great forum and all the fantastic input.  I've been learning a lot.  Not much for me to post, since I can generally find answers/ideas with a search.

I've been going in circles with wall height and framing for the walls and roof, so this thread piqued my interest.  For the roof framing, I think that the codes and my desired location (NW lower Michigan) have finally overwhelmed my DIY impulse.  I don't want to use a ridge beam, so it's looking like parallel chord trusses are my current design direction (e.g. http://countryplans.com/smf/index.php?topic=8693.0).

Personally, if it wasn't for code, I'd be O.K. with the site-built trusses and 2x12 rafters with a rafter tie every 48".  This is in part because I don't think that 60 psf of snow (~10' worth of powder, ~3' packed powder or ~1' of liquid water) would ever be able to stay on a 12:12 pitch metal roof.  However, for insurance and potential resale implications, I do want to pass code.  Dear God, if I do build this, please don't let me have to sell it........

As part of this exercise, I did some frame analysis for the trusses and balloon framed walls (using this: http://www.ralfmartinhansen.de/twdframe/indexe.htm).  I'm not a structural engineer, and there are a lot of assumptions that have to be made, so I'm not sure if everything I assumed was "best practice" (based on some papers I saw, it even looks like "best practice" isn't all that well defined).  In the end, my takeaway was that IF there was 10' of powder on the roof and IF I balloon framed with a let-in ledger and IF my wall height was 10-12 feet and IF I hadn't used some _really_ good 2x6 lumber for the walls.....I sure wouldn't want to be jumping up and down in the loft.

Please don't take this as anything near a negative remark on the plans....I think they're great and, as I said, would like to build them as is (I'm conservative, so I might beef up the trusses a bit and put the rafter ties right at the top plate).  It's just that with my snow load I don't think they'll pass a thorough engineering review.  I spent some time looking at possible reinforcements at/near the knee wall, and it
definitely could work, but the added time and cost, plus the need for an engineering review were turning me off.  I finally started adding it up:

60 psf snow load --> added bracing and engineering review (plus way too much time analyzing things)
R49 --> fiberglass needs > 12" rafter/truss depth, other insulation options more expensive
24" ties --> not so attractive

So, with all that, I think I'm finally letting go of the do-it-myself thing.  I'm thinking that in the end the trusses might even save money.  Now, can I somehow shove the engineering for the >10' walls I want on to the truss manufacturer?

Sorry 'bout the long and rambling first post.  I hope that at least some of this might help someone in a similar situation......

Best,

--Nolan

Thanks Nolan for your comments.

Engineered trusses are often a better way to go when they are available - engineering done by the supplier, rooftop delivery, etc. The site built truss is for folks who want to or have to build them on-site.

Also, if your analysis came up with deflection that looked out of line in the loft, make sure you were using 30 psf for the loft loads rather than the 40 of the main floor. Also loading is even less below 5' headroom. The proof of that pudding is that none of the built houses have had reports of such an issue.

John,

The area I looked at most was around the wall stud and joist connection.  The rafter thrust from above there has a lever arm to work on and the stud sees a good amount of stress in that area.  An angled brace from the top of the stud back to the joist appears to take care of it, but the brace connections would have to be able to handle a pretty hefty force.

Again, this is all for loading conditions that I wouldn't expect the roof to truly see.  So, with that and the numerous working examples here, I completely agree that the proof is in the pudding.  I'd like some of that pudding!  For my situation, trusses should help with a couple other code challenges, and hopefully reduce the chances for snags with approvals/inspections.  I don't know, maybe those snags are largely a crap shoot based on where you are and who you get.

I hope that none of this has come across as negative or a critique...I have nothing but good things to say about the plans and this site.  I just figured that there might be others running into the same circumstances.  I haven't priced out the trusses yet, so who knows, I might be back begging for ideas quite soon.......

Just a thought:   I talked to a truss company in Albuquerque earlier this week.  He will provide engineered stamped drawings for every roof he builds.   I gave him the dimensions I plan to build, and he said he could provide attic trusses with a dormer that would give me a nice room upstairs.  While we were talking he suggested I let him build trusses for the floor.  He would provide stamped drawings for them as well. 

According to him, I should order the plans from John, draw a box to the floor and roof and say "see roof and floor plan" and submit to CID.   He would essentially provide the materials for a
free-span building.

I'm gonna order the plans and go meet with him. He told me he can show me options on the computer of the various truss styles. 

Sounds like a good option to me.  As for cost, he claims it will be about the same as a stick build, but will save me lots of time.   

This has been my experience in WA state as well.

If you have a good truss company they can do some amazing things with trusses and they do all the engineering for your specific location. It not only eliminates your head scratching but solves the interface with the inspector. Sometimes it even saves your back if they have a boom truck! :D :D :D

Let us know if they will do a truss that includes the floor and sidewall as well. That I haven't yet seen.

What is even more amazing is when you see the trusses, I'll bet they have a lot less lumber in them than you would have used stick building the same elements. :o

When I designed our cabin I got truss quotes. They were marginally more money than all the lumber I used for rafters. IF I had been building in a closer location I would have gone with their trusses. However, the delivery fee was high and did not involve a crane truck, just a roll off. I didn't want to deal with raising the trusses, so I built rafters.

Y'all moved on but darn it, I was ciphering.
I've done attic trusses with a sidewall, there is an inboard kneewall and angled strut. Generally lighter members, higher grade material where needed and better connections than stick built.

We can calculate the tie tension as the rafter tie is raised, lets put some numbers to a thought experiment.
http://windyhilllogworks.com/Calcs/raisedtiethrust.htm
Plug in 96" span (a 16' wide building), 600 lb load, 96" ridge height, 96" to tie, 12 pitch... notice the thrust, 300 lbs. Remember half the load is on the wall half is to the ridge, for every pound vertical in a 12/12 there is a pound of horizontal thrust. That was the tie at the plate, every rafter tied, the codebook prescription. Now raise the tie 1/3 of roof height, plug in 64" for the distance from tie to ridge, tension has climbed to 450 lbs.

Now omit every other tie, so double the load to 1200 lbs. Now put the tie 24" down from the ridge and notice the force, 2400 lbs.

The connections and materials need to be capable of restraining that force. The rafter/tie connection has >25 nails at each end to make this a stiffleg truss. We've got a heavy ton of point load bending the rafter there. That restraint probably really won't happen in this configuration. Load goes to stiffness. That horizontal thrust will be transmitted to the other elements. I suspect primarily the kneewall is the next stiff thing downstream.  The kneewall is a wooden lever with a notch on the tension edge at the pivot point.

Let me try to show how I'm looking at that.
300 lbs of horizontal thrust is pushing against the top of a 2' kneewall. The bending moment at the ledger would be 300lb x 2'= 600 ft lbs...multiply x 12= 7200 inch lbs
Max allowable moment for #2 SPF is going to be in the 1150 psi range
Section modulus required= 7200/1150= 6.26"³
Section modulus of a notched 2x6= 4"³... doesn't check. I think we're deep into our safety's


We can use the composite action of the roof sheathing to transmit some of that force to the endwalls. Our diaphragm is unblocked and has no specified nailing pattern though. I offered this solution to the local designer when we had our discussion. His engineer didn't want to go there without some paid time. The structural ridge makes that entire issue go away.

All the numbers above were at my moderate design snow load and a little building.

I looked in the codebook that rode on the dash of Dad's truck, it's a '68 but is very little changed from it's first printing in '57, when NC adopted it's first standardized statewide code. 3 staple binding, I could fold it and stick it in my back pocket. "Collar beams" and rafter ties are in there, the language is muddier but the intent and concerns cautioned about are the same as now. The code then offered the same remedy as now, if outside of the code one simply needs to show that it has been engineered to safely resist the design loads. Those codes and the present code deferred to the Nat'l Design Specification for Wood Construction in '82 for that engineering standard. This design is not the simple farmers home envisioned by the prescriptive codes, which you can still build pretty much the same as in '57. This design may be fine, the code asks for you to show that it is up to the same minimum life safety standards as that simple farmer's home, it asked the same thing back then. The SBCC codes were the most lenient of the three model codes.  

I have not said I am impressed by the current codes or am even in favor of them.  They are the lesser evil. Sort of like seat belts, irrespective of the laws of man, it's a good idea to wear one. Ultimately I try to build to that standard, the natural laws. When I had my discussion with the designer the plans had already cleared plan review with the county. I don't build for the inspector's approval, I knew that an engineer was required to do it the way the designer wanted to.

If the soffit were a flat horizontal beam, just another thought... ;) ;D

Quote from: MountainDon on July 01, 2010, 08:52:45 PM
When I designed our cabin I got truss quotes. They were marginally more money than all the lumber I used for rafters. IF I had been building in a closer location I would have gone with their trusses. However, the delivery fee was high and did not involve a crane truck, just a roll off. I didn't want to deal with raising the trusses, so I built rafters.

If I was building a single story I would likley just stick build.  I have experience with similar projects.  If we were building 16' wide 1..5 story I might also be more inclined to stick build.  Given our (read: the wife's) desire for a 20' wide plan and a loft, I'm leaning toward trusses. 

Building a small house is like building a puzzle.  Lucky for us that so many of you have showed us (in pictures!) how to put one together.  Don, if you ever see a strange couple staring at you at Lowes or HD, its probably me and my wife.  I've made her look at your build thread a dozen times.  She could pick you out of a crowd in a heartbeat.  Your build is darn near the perfect self-build:  one story, right width, and a roof that is not so steep.

Quote from: John Raabe on July 01, 2010, 08:19:01 PM
This has been my experience in WA state as well.

If you have a good truss company they can do some amazing things with trusses and they do all the engineering for your specific location. It not only eliminates your head scratching but solves the interface with the inspector. Sometimes it even saves your back if they have a boom truck! :D :D :D

Let us know if they will do a truss that includes the floor and sidewall as well. That I haven't yet seen.

What is even more amazing is when you see the trusses, I'll bet they have a lot less lumber in them than you would have used stick building the same elements. :o

I'll ask about the sidewall.  Right now it is roof trusses, and additionally 20' floor trusses with an engineered rim.   The advantage of the floor truss according to them is they could build a "box" within the truss that runs the length of the building to install HVAC duct, as well as the ease of running electric and plumbing.  They claim it is better than TJI's.  I'm still researching them, but I have seen them and they are cool. 

Quote from: Native_NM on July 01, 2010, 10:02:55 PM
Your build is darn near the perfect self-build:  one story, right width, and a roof that is not so steep.

Thanks very much.   My Lowe's is the one in Rio Rancho; HD, the one near Cottonwood.   :)    I don't visit them as much this year as I did the previous two years.

If you'd like a personal visit sometime you will be welcomed and only have to ask.

I do want to follow up a bit on the concerns that Don_P brought up in an earlier post to this thread. While his analysis of the theoretical 16' wide house with the roof supported by high collar ties at 48" o/c does show some problems, I don't think any of the owner built houses on this forum have pushed things that far.

The initial question was posted by John_M with his loft pictures. He was asking about his higher rafter ties. His roof actually has both collar ties and something we might call mid-rafter ties.

(https://i1009.photobucket.com/albums/af219/countryplans/John_M-all.jpg)

The analysis Don_P did was for a house with only a high collar tie and even then only one on every other rafter - 48" o/c. He points out that doing this can end up with the rafters pushing outward on the sidewall and perhaps over-stressing the joint.

First let's go back to the thought experiment I was doing based on the photo of John_M's house with the mid-rafter ties. Those are on every rafter pair as they should be and so are the higher collar ties. Now, as long as this triangulated "stiff-legged truss" holds together, there are no outward forces on the sidewall. Imagine all the rafters and ties are made out of welded heavy steel. This unit will sit rigidly on the top off the walls and all normal load forces will be downward. (There can be horizontal wind forces that would want to push it off the plates so a good connection to the plates is still important.)

So keeping this triangular configuration rigid is very important. If and when it might start to show stress by bending nails or tearing the wood, then there could be forces pushing out on the sidewall.

NOTE: This may be a place where John_M could consider adding reinforcement with extra nails or plates - at the point where the horizontal mid-rafter tie and the rafter connect. The nails there are holding the triangle rigid and it takes more of them to do that when the rafter tie is higher. In my 20' wide 1-1/2 story plans I handle this by making the same rafter tie configuration into a site-built (or manufactured) truss with glued plywood gussets and plenty of nails at the connection points.

In the event of a failure of the triangular roof configuration, Don_P's calculation of these outward forces to the side wall come into play. He does not have benefit of access to the details in the 1-1/2 story plans where the ledger for the loft joists is below the connection between the joist and the stud so that each stud is tied into the floor and the full depth of the stud. This joint is also blocked structurally and for fire control.

Don_P does us all a service by pointing out a worst case scenario where the stiff-legged triangle could fail due to leaving off every other tie and then using insufficient connectors (nails in this case) for the tall but lonely collar ties that remain. This failure could then force the walls to provide backup and, if the wall is weakened by a poorly thought out ledger cut, it could result in bowing out of the walls. Even then, as he points out, we are not actually predicting failure but we are into the safety factors of the calculated wood and connector stress ratings.

You do want to stay below them, but those safety factors are why you can open up old houses and find astonishing structural mistakes, where things are way undersized and connected by little more than a prayer, and yet the place has held together without problems for maybe 50 years.

 ???

Lurker newbie question....I am looking at doing a 20X36 in the future.  (2' added in the back and an additional 4' added to the front area....Haven't even purchased the plans yet, that's how newbie I am.)

I would like to do 12' walls, BUT have a rear loft bedroom, open cathedral area, then a front smaller open loft (6') with a catwalk/bridge between the two (similar to the Victoria model loft). 

NOTE: I also really like rwanders Alaska cabin with the front 10' double deck outside accessed by a catwalk.  I liked it so much, I'm considering extending the roof total of 20X46.  This potential cabin site has a partial lake view, and the second story deck would have a much better lake view than the first....

I will most likely purchase trusses, as there is a factory less than 10 mi. from the cabin site and delivery charges minimal (with crane lift possible for additional fee).  Given that there will be lofts at the front and rear, do you think 12' would be possible considering the structural forces?  It will be located in the middle of Illinois, so there are tornado issues to contend with!

So many options that look wonderful have been presented on this forum!  It is truly amazing to see all the knowledge available in one site.

Quote from: John Raabe on July 02, 2010, 07:24:32 PM
I do want to follow up a bit on the concerns that Don_P brought up in an earlier post to this thread. While his analysis of the theoretical 16' wide house with the roof supported by high collar ties at 48" o/c does show some problems, I don't think any of the owner built houses on this forum have pushed things that far.

Well not to hijack but now John has me wondering if I did it wrong.  My walls are 14' balloon framed with a 9' lower ceiling height and 8' upper ceiling height.  Did I mess up and go too far?  I have no cathedral or open ceilings.

(http://www.ouramericanadventure.com/coppermine/albums/house%20feb3/DSC_6785.jpg)

Edit; hmm, I wasn't even sure I was going to post this but I guess I should have been disconnected while typing, I must have hit the wrong key.

There are several ways to avoid doubt. One is to build to the prescriptive code. Another way is to have a truss plant engineer trusses for a roof, another is to use a structural ridge beam, the supplier of the beam will engineer if for free. Another is to hire a structural engineer to check the design. The photo is a case that I would run by an engineer, all you will get from any of us is an unqualified opinion. I hope the walls were fireblocked, that is crucial in balloon framing. Open stud bays floor to floor are chimneys that have led to many fatalities.

If you go to the wall chapter of the IRC there is a wall height table R602.3.1 that runs to 24', do pay attention to the footnotes. Where it says design required, it is this use and the design required is by an engineer. The two pages prior to that figure are very helpful for general knowledge.

Phalynx:

You have a similar situation to the photos from John_M's project and the suggestions are the same.

However, you have a midspan rafter connection (rather than a rafter tie in the lower 1/3) and it looks like only 4 nails there. I don't see the gussets and nailing that I show in the site built truss plans or the connectors at the wall plate. I don't know what the connection is at the top of the rafters. A metal strap will help if you don't have the gussets I show or a collar tie. See Example (http://www.countryplans.com/mcelroy.html).

It looks like you have the loft joists nailed off to the studs, sitting on the ledger and I assume you have the fireblocking as shown. If so, that should be fine.

I think you do need to get more rigidity into your rafter triangulation and lock it down better. An engineer could model the forces and provide guidance as could an experienced local roof framer.

Having been around plenty of local framers...
The link I posted above models the forces in raised ties, the math is basic, it came from an engineer, I simply put it into a javascript to make modelling easier. Once you know the forces you are trying to resist, you can take the output and compare it to the desired connections. One place you can do that is the AWC connections calc, awc.org and determine the number and type of fasteners needed. Generally a dozen nails is better than one bolt, the many nails distribute the load over a greater area, a good thing in a material like wood. They would give plenty of warning while failing where a single connection would not. Also it's not a good idea to mix connectors thinking one will help the other. Typically a bolt takes load immediately where a nail slips a bit and then takes load. In that scenario they are not helping one another they fail in succession.

What I'm trying to do here is make you aware of the forces on that stormy day. It is real easy to see something standing strong on a clear day and not think about the day down the road when it really needs to work. Our buildings will be here for a few generations, they will probably see a bad day or two, think about how they will handle that.

Couponvanalek,
I've attended a good lecture by Brian Readling of the APA on building to survive tornados. He goes in after the storm and tries to see what happened and how to avoid failure in the future. I've seen his powerpoint from the lecture online, I'll try to dig it back up. Basically you need to look at the building from top to bottom. It needs to be on a continuous foundation to absorb and distribute the forces from the structure above without overturning or sliding. The walls need to be adequately braced, the roof needs to be well tied and stable. If you are serious an engineer should review the plan, the weak link is what fails. More eyes on the plan are more apt to spot those weaknesses. It might not be fun sometimes to have someone poking at your plan but the intent is to make it a building you can run to rather than running from.

It is always wise to keep in mind that all we will ever get is opinions - here or anywhere else. The actual decisions will be left where I think they should be - with the owner and builder. Building codes are nothing more than opinions written down. Engineering calculations are opinions based on models of very rare environmental occurrences that attempt to simulate reality. These opinions are always changing and subject to the personality of the engineer, inspector or whoever is in negotiation with the builder. If you think there is a single rock-solid reality here just go ahead and get 10 different engineers to review your plans. I'll bet the cost of implementation will vary by a factor of 4x to 10x and all will be able to satisfy the inspector.

Should every house be built to withstand an engineering model of a sever tornado, earthquake or hurricane? How about a direct meter hit or a 100' tall tsunami? There will always be an even worst possibility that some very cautious people will expect to be protected from.

And there will be a code standards expert that is working to get that written into building codes.

Does everyone need to drive a Hummer or a tank to feel safe going to the grocery store? No, but some do as sales of SUV's will attest. And, lest we forget, cars (and houses) are safer than they were 50 years ago.

My advice: Build to local code whether or not you will have inspections. That will get you all the current airbags and safety features. Then, if you face a question or specific option you want to do, study what kinds of houses have survived your local weather conditions and build like that only better.

Quote from: John Raabe on July 24, 2010, 09:48:34 AM
My advice: Study what kinds of houses have survived your local weather conditions and build to that, or a little better.

Good advice....I will try to find the video presentation that Don P recommends regarding building for tornado areas.  We live/plan to build in an area where these are a real possibility. (But at least we don't have the earthquakes we used to have in CA!)  Personally, I'm the cautious kind that would rather overengineer the framing/foundation/roofing....

I build and try to show methods recognized as being safe based on our experience and knowledge at this time, nothing wildly overblown, simply up to or beyond minimum accepted standards. I am not requiring anyone do anything, its just information.

Rational design (not my term) is the process of determining loads and showing adequate resistance without stressing any parts beyond accepted allowable limits. The methods and tables in the building codes are the easy way to do that, but are not the only way. A designer can also have an engineer review the plans and show that they have met this standard.  

Notice the photos here are of real events, no tsunamis or falling meteors, just a bad day at the ranch. Take a look at the percentage of occurance chart. If you build above an F2 how rare the chances of a failure are. That isn't a real big hill to climb and if the F5 is not a direct hit, you'll probably survive it too.
http://www.apawood.org/FSD/CharlotteWSF_LateralMarch08.pdf

Thanks Don: Great resource with valuable pictures.

I know we agree about wanting to help the forum members to build strong safe houses.

When planning a house you want to know what types of design forces you might face at your building site. Call up the building department or a local engineer and they will know. The maps on page 15 are a start. The wind map HERE (http://www.fema.gov/plan/prevent/saferoom/tsfs02_wind_zones.shtm) is a bit more readable.

These are for three second bursts of wind such as found in a tornado and not sustained winds. The lowest wind on that chart is 130 mph and that is considerably higher than the design wind speed for most of the white areas shown there. Thus these maps don't say much about the chances of a given building ever experiencing such winds. That's what you would really like to know. Still, if you are in the Red zone you can expect to have a higher chance of wind damage than if you are in a white zone. Act accordingly.

The force diagrams in the booklet are very good at showing how the shear walls and anchors into the foundation work to stiffen a building. The diagrams about tying the parts of the building into a fully connected load path are also helpful in understanding the relationship between the foundation, floor, walls and roof.

If you are in the "dark" zones on either of these maps on page 15, you would be wise to spend a bit more money on an engineer to make sure your plans are reinforced to meet local wind and earthquake conditions. The design loads and recommended solutions are site specific for your local loads. Standard stock plans (mine or those from anyone else) are NOT designed for the worst case conditions of a sever earthquake, hurricane or tornado hit. This type of customization needs to be done locally or you will be paying for strength where you don't need it.

Even if you aren't in such a high danger zone some of the simpler things such as good hardware connections, plenty of nails and a concrete perimeter foundation tied into 4' reinforced corners will add extra safety and strength to any house. As I have mentioned in many places on this site the beam and pier foundation may not work well in all locations and should be engineered where forces of earthquake and hurricane force winds are to be planned for.

Another good read on high wind construction;
http://www.weatherpredict.com/pdf-downloads/HurricaneCharleyDamageSurvey.pdf
I am in that special wind region at the upper end of the Blue Ridge in VA on the map John posted. We kick in at 4000', gorges, and ridges. I worked on one job last summer and we were several feet below the limit, the inspector let me know we didn't have to use the high wind rules if we didn't want to. I didn't figure anyone had consulted the wind on the precise location of that line, we were looking 2 counties away from up there.  In some wind events around here the damage has been the neighbor's flying utility structure. We called our place windyhill after watching the wind guage blow apart. We are technically not in the SWR here at the house, 1200' below it and protected. Prior to those designations, we lost the first roof on our 8x12 camping shed, it went down the pasture a couple of hundred feet. The joys of being young and optimistic  :D.

Edit; more links during today's siesta;
Supposedly some videos on hurricane construction here... sorry I'm too slow to review these;
http://www.hurricaneconstruction.net/?q=node/58

Simpson's high wind page;
http://www.safestronghome.com/highwind/

On structural ridges and balloon frame walls;
http://www.structural101.com/Structural-Ridge-Beam.html
http://www.smandf.com/25-ConventionallyFramedRoofs-TheGabledRoofProblems&Solutions.pdf
http://www.eng-tips.com/viewthread.cfm?qid=162638&page=1
http://www.eng-tips.com/viewthread.cfm?qid=140191&page=1

A couple of days ago I was working on my build when a storm rolled in. It started raining and soon after, hailing. The temperature immediately dropped about 15 degrees.

Nothing unusual so far, it is the season for that here, but then from a different direction a very warm, tropical feeling breeze started blowing. I noticed  the sky was a bluish green. You don't need to know anything about weather to know that conditions like these are not good. The hair on the back of my neck stood up.

It was pretty stormy for awhile, but nothing really came of it.

I doubt there are too many people, faced with a situation like that who would still say that they overbuilt, or say that they are still glad they cut certain corners.

I agree with Don. Tornados are very rare in my area, but you should build with nearly the worst case scenario in mind. I say nearly because in a worst case scenario you had better be in a deep cave!  [shocked]