Ridge beam and Design?

[addicted]

Hey Guys
I've been looking around for a short time on information about ridge beams. My question is... How much does a ridge beam reduce the outward thrust at the rafter plate?

Thanks in advance.
Rusty
beware that answers will be followed by more questions.

[beenthere]

Theoretically, I would say all of it.

If the ridge beam deflects under load, then there would be outward thrust.

[addicted]

So.............does that mean you need a larger ridge beam as the pitch decreases? If you go from a 10/12 to a 6/12 does the amount of force on the ridge increase?
Rusty

[beenthere]

Let's hear more details of the question.

[Jim_Rogers]

We need to first clarify what the first posted means by a ridge beam.

A ridge beam that connects rafters together at the top doesn't do much other than connect them together, unless it is supported by a post. Then it is a "supported ridge" or a "structural ridge".

Think of it like this.

Let's say you lean a ladder up against a house, on an angle. The ladder is sitting on a puddle of water, but the puddle is frozen so it's ice. Very slippery. At nearly any angle the ladder is going to slide out away from the house as you climb the ladder. Faster when it's at a low angle.

Now, let's hook the top of the ladder to house.

It can't slide out very much at all, even at a low angle, unless the point where it is hooked to the house is lowered.

A ridge beam would/could be easily lowered with a lot of load, such as snow.

A structural ridge or supported ridge, can't be lowered unless the foundation or some joint between the ridge and the ground/foundation fails, and the ridge beam is lowered.

My point is we need to understand is this just a ridge beam? or a supported/structural ridge beam?

Jim Rogers

[addicted]

Thanks guys
You're right... I should have been more specific.  I'm in the beginning stages of designing a frame and was thinking of using a structural ridge beam, supported by posts, on which the rafters will be attached on top thus reducing the outward thrust at the rafter sill plate. (high post cape) So with what I've learned from most of the books mentioned on this sight and a few TF courses, if you have the rafters unsupported at the ridge, the outward thrust at the rafter sill plate increases dramatically as you lower the pitch of the roof. IE. 12/12 to an 8/12 pitch. So if you put in a supported ridge beam it will remove all outward thrust. But let's say that your local saw mill can't cut a post long enough to reach all the way up to your ridge beam at a 12/12 pitch, hence you reduce the length of the post, which reduces the pitch of the roof.

Here's the question. If you reduce the pitch of the roof do you have to change the depth of your supported ridge beam. My thinking is the lower the pitch of the roof the more force you will have at the supported ridge. Is this correct?
Rusty

[Jim_Rogers]

If your sawmill can't make a post long enough to go from the foundation up to the ridge beam, then change the design by adding a scarf into the post to make two beam into one.... vertical scarfs can be done very nicely and they are strong, if done correctly.

The depth of the ridge beam should be calculated to support the load that is on it. That is a vertical load of holding up half the rafter span, on each side of the ridge.

Here is a cross section to show what I mean:



Keep asking questions.....

Jim Rogers

[witterbound]

Seems like one would typically have a tie beam that ties the two ouside posts together, and then you'd have a king post sitting on top of the tie beam running to the ridge.  If you needed, you could add a third post under the king post. 

[icolquhoun]

just to rule out the obvious, but if the sawmill can't cut a long enough post, what about the sills, plates, and tie beams, which in most traditional capes are all over 20' ??? I understand these can be scarfed
typical new england cape was 24widex36deep or thereabouts
not that I approve of this frame, but it's a supported ridge using a 2 piece post:

(http:timberframedesign.net/files/ridge_beam.jpg)
{Off forum pictures are not allowed, please place them into your gallery and post a link from there, moderator}

[addicted]

The limitation on the saw was just hypothetical. I was just trying to imagine the forces on a supported ridge beam at different pitches. But the idea of having a post on top of a tie beam brings up more questions. The post on top of the tie beam would now be a king post right? So aren't king posts in tension? Or does the load from the supported ridge beam nulify the tension and put it in compression?
Rusty

[Jim_Rogers]

The post on top of the tie beam would now be a king post right? So aren't king posts in tension?

If the post on top of a tie beam is hung between two rafters then yes it is a king post and it is in tension.
But we are talking about supporting a ridge with a post above a tie beam.
The rafters will not "hang" the post.
The rafters would look something like this:



The rafters would be secured to the ridge beam with some long timber lock screws or something like that.
The gap between the rafters is because the ridge beam will shrink and if they touch before it shrinks then the will not be supported when it does.

Your design has to be right to convert the outward thrust of the rafter to a vertical load on the frame. (very important)

Jim Rogers

[routestep]

If you lower the pitch of the roof, the ridge beam would have to support more load and hence you would increase its depth to keep it from sagging. If you had a flat roof in the extreme, the ridge beam would carry half the weight of the roof and the plate would support the other half. There would be no outward thrust.

If the ridge beam is well supported by posts down its length (it doesn't sag) then the rafters should not put any outward thrust on the plate. Or at least I don't see how it could. If the ridge sags you could get a lot of thrust force, but the movement out might not be too great.

If the rafter is bolted to the ridge and plate and it (the rafter) sags there could be an inward thrust (or the bolts might tear out in an extreme case).

[addicted]

Thanks guys
So now I've been looking for a way to calculate that load on the supported ridge beam, but all the books that I have don't include the pitch. If the load on the beam is greater with a lower pitch then wouldn't the pitch be included in the calculation? And if the pitch must be included calculating the load on the ridge beam, then shouldn't it also be included in calculating the load on the rafters? This is starting to make my head hurt. So what is the standard way to calculate the load on a supported ridge beam and rafters if the pitch is directly related to the load?
Rusty

[meddins]

... If the load on the beam is greater with a lower pitch then wouldn't the pitch be included in the calculation? And if the pitch must be included calculating the load on the ridge beam, then shouldn't it also be included in calculating the load on the rafters? This is starting to make my head hurt. So what is the standard way to calculate the load on a supported ridge beam and rafters if the pitch is directly related to the load?

My understanding of this situation...

I don't think the load on the ridge beam is increased by the lower pitch. The load is simply the weight of the structure the ridge beam supports plus forces like wind.

However, a lower roof pitch does increase the rafter's horizontal thrust at the wall plate/post, which is different from saying it increases the load..

Controlling or eliminating that horizontal thrust is a function of truss design, that is, inserting the right members at the right spot in the truss to transfer loads vertically down and away from the wall plate/post.

My understanding is that when you calculate load for the ridge beam, you can draw your roof in plan view and treat the ridge beam as a uniformly loaded beam (assuming rafters are spaced at regular intervals along it.)

Hopefully Jim or someone will correct this if I'm wrong. I'm going round and round with roof design too at the moment..

[Jim_Rogers]

The post on top of the tie beam would now be a king post right? .....
Rusty

As every timber in a timber frame is given a name based on it's location, the name of the post that supports a ridge beam may not be called a king post. I'll have to do some research on that post's name and see if I can find out what it is called.

If a post supports a common purlin, it is called a purlin post. When a post is on the first floor and supports the middle of a continuous tie beam, I've heard it called a prick post, but I don't know how or why it's called that.

The post that holds up a ridge beam, I think is called a crown post, but I'll have to look into one of my books to see if that is correct. This post would not touch the rafters, where a king post is hung between the rafters and holds up the tie beam.

Jim Rogers

[addicted]

Thanks guys. I guess I'll start sizing the rafters and supported ridge beam based on horizontal distance and 50% shared load between the rafter sill and the ridge beam.  This is what I get for over thinking things.
However this does not mean that the questions will stop.

Enjoy the snow
Rusty

[Jim_Rogers]

According to my research:

"Crown post ends at a collar or collar purlin, king post goes to ridge beam or apex of roof."

Source: Jack Sobon.

[Thehardway]

Jim,

I agree with you and Jack on the crownpost/kingpost however it seems as though the confusion here on the term "king post" stems form the fact that a post historically was only referred to as "kingpost" when used in a truss. That truss may have had a "ridge beam" or "ridge plate" or "Ridge Piece" which it supported but the thrust was directed through the truss rafters/top chords to the wall plates, not straight down to the ground through the post.

We have crossed the thin line here between timber frame construction which never really used a central post to support a "ridge beam", and modern "post and beam" construction which commonly uses tall posts and transfers loads directly to the ground and would use typically steel plate joinery.  Therefore this member is an orphan doesn't have a proper name and the joinery at the apex is questionable.

I believe the traditional way of framing this structure would probably have been purlin to a purlin plate.  It could also have been done with crown posts.  I suspect the structural "ridge beam" with central post was not used because the number of intersections and size of the joinery that would have been required in the apex at the post connection would have compromised its integrity, therefore, it evolved later in modern post and beam when steel plate connections and glu-lams became the norm.   This is not to say it was never done but that there are probably not many examples of it that are left standing for documentation and perhaps there is a reason for this?
What do you think?  Is there any good reason to depart from traditional design parameters here?

[addicted]

Thehardway
Thanks for your input on this ridge beam subject.  Being very new to timber framing, I was not aware of the short comings of this type of design. Would you mind expanding on the weakness in the joinery at the apex and any other areas in this design that might pose a problem?
Thanks
Rusty

[Left Coast Chris]

One more thought to consider:  The International Building Code and the Uniform Building Code allow for snow load reduction for steeper pitches of the roof.   So...... for the steeper pitches the ridge beam size could be reduced depending on your local snow load.   Also keep in mind that the International Building Code requires more load combinations to be considered so if you are trying to meet code you should have an Engineer determine the load and size the beam(wind, rain, live, snow are now combined).

Here is another tricky thing to consider:  for dead load only, the steeper pitched roof has the same tributary area for load analysis as the flat roof but the rafters, sheathing length and roofing material need to be longer to span from plate to ridge.  Since they are longer they have more dead load than the flat scenairo.   Engineers adjust for a little more dead load for the steeper pitches.    More common though is for steeper roofs to be used in heavy snow load conditions so snow controls more than the small increase in dead load and the reduction in snow load for slope is often significant and almost always more than offsets the increase in dead load so you can reduce the size of the ridge beam due to the steeper pitch.

[Thehardway]

Addicted,

Placing too many intersecting joints at one location removes too much wood from the members and weakens the overall strength of the joint. We use the term relish. Spreading the joints out some distance apart is preferable as it leave more relish. A tongue and fork joint was the preferable joint used for the apex of principle rafters rather than a mitered connection.  It gave a substantial amount of wood for each rafter to bear on so that shearing action was prevented and provided for plenty of relish.

Here I am am paring the tongue and cleaning the fork to prepare rafters for assembly.



Here is the test fit. It will later be bored pegged when finally assembled.



A collar beam will be fitted fitted to additionally support the rafters.



Finished pairs



Thrust at the eave was controlled with an anchor beams or continuous tie beam. Collar ties, purlin plates, and canted or raked struts occasionally referred to as "prick posts" were used as well to support principle rafters at intermediate points dependent on spans and loads.  Most buildings were designed with posts at regular intervals.  Long spans were not as important as was economy, ease of assembly and structural soundness.

Here is an example of a traditional frame design By the New England Barn Company http://www.newenglandbarn.com/dutch-anchorbeam-barn.php using rafters and purlin plates and posts.  Notice that the joints are all staggered and No two joints fall in the same location on a single member.



Here is a picture of a barn using purlin post and plate construction. notice that the common rafters are actually spliced at the plate.  This was common when rafters were not available that were long enough. You may also notice a mid-span scarf joint in the plate.  This was done to avoid excess joinery over the post which would cause weakness. I would rather see a continuous plate without the joint but never-the-less this is an example of a building which is over 200 years old and still sound.



Most long spans were found in storehouses or in churches and cathedrals. When the span moved past the point where a tie beam was practical, a truss was used.  The Kingpost and Queenpost truss were most prevalent with Kingpost being the simplest and Queenpost the most efficient.
Here is an example of the joinery for the apex of a kingpost to rafter intersection. Notice the tenons are designed to prevent the kingpost from traveling downward by means of a wedging and ratchet effect.  The kingpost does not support with upwardly, rather it pulls the rafters together with its weight and by supporting the tie beam below it. Plenty of relish is needed in the kingpost top.



Most kingpost trusses that failed, failed because of additional joinery at this point.  It removed too much wood.  In reality there is no need for a ridge beam here. Common rafters can be birds mouthed and pegged into purlins and plates.

Desire for longer spans and lack of longer timbers led to the development of the Hammer beam truss but this required shortened posts at the walls or some form of abutment to control the huge amounts of outward thrust they generate.

Later on the Post and Beam construction style popularized tall posts supporting longitudinal beams. Most of it omitted Mortise and Tenon joinery and used steel gusset plates and brackets which were thru bolted or they used concealed steel fasteners to make the joint strong without removing so much wood. These two distinctly different types of construction have now merged and the terms are almost used interchangeably.  This has been promulgated by the use of Internet search engines to the point where just about every timberframing company uses the term post and beam interchangeably with timber frame and vice versa.  It has led to some unorthodox structural arrangements which require a lot of unnecessary engineering and confusion as to what supports what and how it is connected together.  I predict it will cause problems in the future when people try to add, modify or maintain a structure as this is when it becomes important to understand what supports what, what is load bearing and what is not etc.

One rule of timberframing is to try and use the simplest joint possible to do the job well.  If you can avoid complex intersections you will serve yourself well not only in longevity and structural soundness but in effort, assembly and less waste.

That said, with enough money, effort, and time, anything you want to see can be accomplished.

What you are doing with a supported ridge beam can be done and is not wrong, it just makes life more difficult than it need be and I would avoid it unless you have a particular reason for doing it, which you may have that I am not aware of.

Hope this helps more than it confuses.  Keep asking questions.  We all learn together, here amongst friends.

[addicted]

Thehardway
Thanks for shedding some light on this design. My goal was to make a frame that used the least amount of wood for the most space, all the while shooting for a cross between a high post cape and a colonial.  I was trying to stay away from collar ties because I thought they would be in tension, then I went to collar tie with queen posts then came the ridge beam idea which I thought would solve all the problems but it seems to have sprung up several more.
Any way I slice it, I can't find a way to have a usable second floor without having more than two joints in one area. If I can upload the pic from sketchup I will, but I have to figure out how to shrink the file first.
Thanks again
Rusty

[Jim_Rogers]

but I have to figure out how to shrink the file first.

java uploader does that automatically for you....

[addicted]

The last time I was able to access the page to upload a picture it said file length too big. Now when I click the upload link I get an Internet Explorer error, which closes everything.
Is there something I should be doing differently?
Rusty

[Jim_Rogers]

Not sure, there is a post here some where that tells all the ways to upload pictures. I don't know where it is for sure, but someone usually post a link to it for us.

Or you can do a search for the instructions.

Every time I use the java uploader it works great for me.....

Jim

[addicted]

Thanks Jim
It looks like jpegs load just fine but the png file from sketchup won't. I'll keep trying

Rusty

[Jim_Rogers]

Maybe you should convert the png to a jpeg first.....

Or can you save as a jpeg from sketchup?

[addicted]

How would one convert a png to jpeg? And or how does one save a sketch up file as a jpeg?
Thanks

[Jim_Rogers]

How would one convert a png to jpeg? And or how does one save a sketch up file as a jpeg?
Thanks

When you export, a 2d graphic, use the pull down blue arrow on the right end of the box that says: Export type: and select jpeg.......

hope that helps....

[addicted]

 



That looks like it did the trick Jim Thanks

This is version one.  I put different sized and spaced rafters on either side to see what the wife thinks. I've also decided to delete the joist that joins the center post and either put a smaller one on either side or space the whole group accordingly. Probably the later. I'm sorry there is no joinery detail but I've just started using sketchup about a week or two and have quite a ways to go.
If anyone sees problems or future headaches please feel free to speak up. Don't worry about being gentle , I've got thick skin. 
Rusty

[Thehardway]

What are your estimated dimensions?

[meddins]

Rusty, in your sketchup model there doesn't appear to be a collar tie or tie beam. As I see it, the weight of half your roof is being transferred directly to the wall plate/post connection. All of that horizontal thrust is being transferred to some 1.5" or 2" tenons at the top of the posts?

I think you would be better off incorporating tried and true historical precedents into your frame design.

On page 29 of Richard Harris' Discovering Timber-Framed Buildings there is an example of an historical frame that achieves what your trying to do. I would highly recommend that book along with Cecil Hewettt's English Historic Carpentry. Of course you can't go wrong with Jack Sobon's and Steve Chappell's books either...I'd say don't try to re-invent the wheel but see if you can make your design work within an existing, tried and true model that is the product of centuries of trial and error...

Free advice...for what it's worth..

p.s. I like your general frame concept and dimensions...not too different from a frame design I'm working on at the moment..

[witterbound]

How tall are you second floor knee walls?  It looks to me like they're not that much shorter than your first floor walls.  If you make them 8' you could put a tie between them, and put a king post on top to support the ridge which would be overhead and out of the way, and reduce the pitch of the roof, and have any entirely open second floor.  I guess if you've got a 1000 sq ft footprint, and you add 3' to a wall, you're adding 3000 cubic feet to heat and cool.  If your house is two stories, you're adding  23 % to heat and cool (if we assume you originally had an 8' first floor + 5' second floor) = 13,000 cubic feet and 3,000/13,000 =   23%.  But you're also getting more usable space, so you could theoretcially reduce the footprint.

[addicted]

TheHardway
The dimensions are..... Footprint 30' x 53'
Posts  10"x10"x15' outer      10"x10"x27' inner
Tiebeams 10"x14"
Raftersill 10"x14"
Supported ridge beam 10"x14"
Rafters 6"x10"
bents on either end  are roughly 14' and the center two bents are roughly 12'


Meddins
Thanks for the book suggestions. I will be searching for Harris'  book soon and I already have most of Sobon's and Chappell's books. I'm really not trying to reinvent the wheel, since my experience is very limited. In one of the TF courses I took, the idea of a supported ridge beam was mentioned as a way to eliminate outward thrust.  I then saw a frame raised locally by a big TF company that used the same idea. That is what started my questions of this design and if it works. Although the the frame I saw raised locally used splines on all the post/beam joinery and resulted in a lot of short beams. I read in the Timber Frame Joinery and Design book about the pros and cons of using these splines but haven't seen anything talking about the use of a supported ridge beam. Though Ted Benson does have a design in the back of that book that has what appears to be a supported ridge beam, it also looks as if there are mechanical fastening plates. Not sure.
Does the picture you're talking about in Harris' book look anything like the one on pg14 of Chappells' book? That may be a better idea. I was trying to avoid using purlins because I heard they were a pain to raise but a scarfed 53' ridge beam is starting to sound a lot worse.

Left Coast Chris
Thanks for the advice. knowing that the pitch could reduce the snow load would make a big difference. I will have an engineer look at the plan but I would like to have most of the obvious problems worked out before hand to reduce the number of major revisions.

Witterbound
Good eye. Your guess is exactly correct. the main floor has 8' to the bottom of the beam and a 5' knee wall on the second floor. I was trying to use the least amount of timbers for this design but you may be right that a higher wall and lower pitch with a tie beam might be much easier. 

[ponderosae]

I was just looking at this topic the other day, along with the Southern Pine tables for ridge beams, which were published since, so maybe that information can add to this discussion.

Their assumptions for table development do not seem to include the slope of a roof (and neither do their tables for allowable roof loads). Perhaps the framing works the same at any slope, it's just that the areas between framing can be problematic for flat roofs especially.

For the beams, 'a 3.0" bearing length is assumed, except for the sizes noted which require a 4.5" bearing length'.

'Beam size is based on the load transferred from 1/2 the span of the supported roof framing. Calculations assume the worst case of simple- or continuous-span glued laminated timber, but only simple-span Southern Pine lumber beams'.

'Note that reference design values have not been adjusted for buckling. To use these tables, therefore, the compression edge of the header or beam must be laterally supported at intervals of 24" or less. In addition, lateral support must be provided at bearing points'.

I guess that covers most of what determines how well a ridge beam works, then.

There's also an article which goes into more detail about the design, and mentions the roof slope, but basically it states:

"The only way that low ends of rafters can move outward is if high ends also move downward. Therefore, if high ends of rafters are supported to prevent downward movement, low ends will not move outward."

Further information about that is on the Questions & Answers page there:

(Q) How much load is on a ridge beam?

"For downward ('gravity') loads, calculating design loads for a ridge beam is relatively easy for simple conditions; uniform dead load, uniform live load and two supports.

Snow load is defined (by code) along the horizontal span of roof rafters or joists, not the sloped length. Dead load is most often also applied along the horizontal span, although, to be completely accurate, weight of dead load should first be calculated along the sloped length and then calculated for the horizontal span. For relatively low slope (up to about 30 degrees from horizontal), there is almost no practical difference"...