Heavy clay roof. Need some advice on the calculator

[Planeiron]

Not sure what's going on but I can't upload to my gallery. Can't see the option even. Just getting the error message "
Error
You don't have permission to perform this operation."

Have wiped cookies and tried different browsers but no joy. Anyone else getting this? Probably just me!   :

[Jim_Rogers]

Picture gallery is currently broken due to some behind the scenes upgrades.
No one can access it as of right now.

Jim Rogers

[Planeiron]

Thanks Jim. Will there be an update when it's back online?

[Jim_Rogers]

Thanks Jim. Will there be an update when it's back online?

the picture gallery is working now.
Jim Rogers

[Planeiron]

Thanks Jim.

The more I read, the more questions I end up with but I suspect that's the nature of studying most things. Anyway I've sketched a transverse section and (apart from those braces looking a bit weedy) I'm fairly happy with everything until I get to the tie and king post truss. I'm in metric land (and my sawyer likes metric) so I've labelled the left side in metric and the right in imperial.



 

I've left out common rafters in the diagram for clarity. This is a 12:12 roof with a 10' span so the distance from ridge apex to outer arris of the plate is around 85". With this smallish distance I hope to avoid a midway purlin as I'm planning to install a couple of reclaimed (from my old kitchen) velux windows and these are about 40" long. 

So to calculate whether the common rafters can support the weight of the roof is it a question of finding the weight of the roof about half way either side of each rafter as for example the one marked below between the 2 green lines?



 

Next question relates to the king post truss. I read in a TFEC publication that it might be best to toe in the principle rafters to the tie if there's enough meat beyond (arrow on first picture above). Then is it feasible to have the join between P rafter and tie not over the jowled post as shown. The tie is 6 x 9" white oak.



 


Can I use the calculators to decide whether I need to strut this relatively small gap between principles (6 x 6") and king post. I think Don, from your previous replies, the ridge beam carries 1/2 the weight of the entire roof and the plates a 1/4 each. What would the calculation be for each toe or heel of each of the 3 king posts?

Apologies for all these questions. It must be a pain!

[Jim_Rogers]

Your green lines showing the rafter support area looks right.
The normally way of figuring it is the area is usually the same as the spacing from center line of the rafter to center line of the next rafter.

Jim Rogers

[Don P]

Yup, Between those green lines is the area tributary to the rafter the load flows to the rafter in the middle. "Tributary" is exactly the right word  .

Uhh, yeah, whatever on the toe being the best bearing. Then worry about shearing stress and try to move to a heel connection  ;). On this, I'd bet you are fine either way. As the span and load increase, shear exceeds wood's strength and steel in the heel joint restraint enters in. Or another way to think about it, if you can design that connection you can probably build the truss.

What would the calculation be for each toe or heel of each of the 3 king posts?

These are both pretty funky, I think this will give what you are asking for as "Lower half of Top Chord, Heeljoint Compression"
Untitled (forestryforum.com)

Untitled (timbertoolbox.com)

[Planeiron]

 

 Thanks Jim and Don. I think I'm close to getting this but I would appreciate a skilful eye cast over this drawing just to say if the joinery is in the right ballpark. Dashed lines are internal joinery. Just looking for a general yay or nay and then might ask a few details about angles and depths of those top chord to king and top chord to tie heels. Just for context, green european white oak, span is 10' and pitch 12/12. King is 10 x 6" at the joggle tapering to 6 x 6", chords and tie 6 x 9", struts 6 x 6", ridge 4 x 6". Many thanks

[Jim_Rogers]

I'm sorry but your background is bad, and I can't hardly see the lines at all. 
You need to draw darker lines or get a better picture of the drawing.
Standard for peg holes is 2" off the bearing end of the tenon and 1 1/2" off the shoulder.

Jim Rogers

[Planeiron]

I've tried to modify this picture. Thanks about the peg hole placement. Is it any better? ;D Looks OK on my screen... but will maybe redraw and try to scan it in if this is no better.



 

[Don P]

There's more than one way so take this for what its worth.

From the top, check the ridge, it seems light. I'd lose the spurs on the top chord to kingpost and reduce the tenon to a stub 2" long, leaving more meat there. This is a compression joint, no peg.   

The struts have stub tenons, no pegs.

The bottom of the top chord, lose the tenon completely, This is where I was saying it is hard to get all of that to actually bear. Keep it simple. Swing the spur to a bisect of the angle and size it to resist the thrust in compression and shear. Compression is checked using the axial compression running in the top chord hitting the end grain of that notch at some angle between perpendicular and end grain compression. To find the allowable compression for the bisect angle you'll use the "Hankinson formula". Shear beyond the notch in the bottom chord is checked using the tension in the bottom chord against the allowable shear.

Bottom of the kingpost is in tension, put the pegs below the neutral axis of the bottom chord. Or go completely thru and key under. Mainly I'm saying the pegs as drawn are a "tension perp" problem, tension perpendicular to grain. The NDS says get on the other side of center to the pull. I do see the half dovetail which is sufficient without pegs. 

I think I just removed all the pegs and a good bit of the tenons chasing timber strength. As I tend to do. That is fine for gravity but does not take uplift or reversals into account. I'm guessing the dead load is greater than the wind load's uplift force but to be on the safe side I would hide allthread across the top chord/king and heel joint connections. Being green oak, hot dipped galvy or stainless for any metal. This will look quite a bit different in 5 years than the day you make things "dollar bill tight".

[Planeiron]

Many thanks, brilliant info and now convinced that I'm out of my depth :D. That Hankinson formula certainly seems to be what I'm after but after hitting the online books :P I'm bottom of the class and can't work out what numbers to plug in. Gonna try and find an engineer to size these timbers and detail the notching although I'll be surprised if I can commission anyone local as there is very little timber framing here. Modified my design (not yet to scale) based on suggestions above. Love the keyed tenon idea and could be a very nice feature. Ridge sitting entirely on the king and without spurs seems very sensible and much easier to cut. I've left in the tenon on the tie on this modified drawing but I think I'm coming round to the threaded rod idea so I can sleep easier.



 

[Jim_Rogers]

Recently, I had two frame go through engineering review and here is a picture of one of the king posts at the ridge on a 10/12 roof pitch:



 
You can see that these top cords go to the king post with a tenon on each side. The tenons are in a through mortise:



 

This shape of the king post passed.
Another frame with a king post and ridge beam over the king looked like this:


 

Again this frame the roof pitch was 4/12 but the king post had a centered through mortise for the two top cords to seat into. with tenons.



 

Again, you can see the through mortise and two tenon on the top cord to king post detail.

And here is another couple of photos of a completed king post top:


 



  

I hope this has helped you to do your frame.

Jim Rogers

[Planeiron]

Thanks Jim, that's really helpful and it's great to see the finished work.

[Don P]

That lowered ridge is nice, that helps uncrowd things at the peak. The callout for 2 timber screws in each rafter up top was tough to fit. If I were sawing my own I'd be tempted to go thicker than 2x just to be able to stagger screws.

I wandered down to the land of lost toys and took a few pics of how I've done a couple.
I'm not pegging the peak and don't like to do any more removal than needed so this is a shallow mortise on one side of a kingpost top;


 

The lower row there is the top chords with stub tenons. I believe that is a tie with a full length tenon on the row above them;




The above was in 6x stock, this is all 8x10's but shows the basic proportions of what I usually do. I passed the simple straight bottom tenon on the kingpost on these through the bottom chord several inches and planned on pegging low in the bottom chord. (I would have to dig for an hour to get a bottom pic, the maid quit :D). If you do a key below the bottom chord check shear/ provide enough relish in that tenon beyond the key. Your short bottom chord really has almost no stress in bending, that's all really more for information. I'm also not certain you can assemble that half dovetail, a lot is coming together for it to be kicked to the side a bit as its coming home. Make sure you can get it together before cutting them all.



 

Hankinson... One thing you can do is use the compression perp to grain values and see if you need to dig into it at all. I think it would be safe to use our white oak group design values with your white oak. I'm assuming #2 grade. Compression perpendicular to grain is limited to 475 psi. Parallel to grain it is 800psi and I'm just too lazy to write it out, Mr Hankinson sez at 22.5° it is 727 psi allowable. If anyone is interested in how holler  .

It looks like the bent spacing is 8' and with common rafters half the load is on the truss, half on the plates soo, 40 square feet x 70 lbs per square foot=2800 lbs on the truss  :D I think we're ok.

Kingpost truss calc here, you have a load to plug in;
Untitled (forestryforum.com)

Heeljoint compression (top chord axial compression) controls in that joint vs bottom chord tension so we need to resist about 1485 lbs of compression. We can handle 727 lbs per square inch safely. 1485/727=2.04 square inches req'd... squat.
You are 6" wide.. 2.04/6=.34" deep, obviously go deeper with the notch, at least 1.5" x 6" wide so it cant jump a tad and slide. That is also why just about every one of these has some form of clamp or clasp or mechanical connection. The notch depth limit is 1/4 timber depth out near the ends of the bottom chord.

The webs are decorative in this arrangement there is no bending load on the top chord, that is carried by the commons. The webs can be omitted.

[Tom King]

Pam and I were in a restaurant today in a building built in the 3rd quarter of the 19th Century.  I was reminded of this thread when I saw this "coffee table".

It had been taken out of part of the building that the restaurant is in when they opened up part of the second floor to make a high ceiling over the dining room.

Notice the low depth of the mortise.  We had an interesting discussion with the owner when I explained why the mortise was not so deep, and why there was no through tenon.



 

[Planeiron]

That lowered ridge is nice, that helps uncrowd things at the peak. The callout for 2 timber screws in each rafter up top was tough to fit. If I were sawing my own I'd be tempted to go thicker than 2x just to be able to stagger screws.

I wandered down to the land of lost toys and took a few pics of how I've done a couple.
I'm not pegging the peak and don't like to do any more removal than needed so this is a shallow mortise on one side of a kingpost top;


 

The lower row there is the top chords with stub tenons. I believe that is a tie with a full length tenon on the row above them;




The above was in 6x stock, this is all 8x10's but shows the basic proportions of what I usually do. I passed the simple straight bottom tenon on the kingpost on these through the bottom chord several inches and planned on pegging low in the bottom chord. (I would have to dig for an hour to get a bottom pic, the maid quit :D). If you do a key below the bottom chord check shear/ provide enough relish in that tenon beyond the key. Your short bottom chord really has almost no stress in bending, that's all really more for information. I'm also not certain you can assemble that half dovetail, a lot is coming together for it to be kicked to the side a bit as its coming home. Make sure you can get it together before cutting them all.



 

Hankinson... One thing you can do is use the compression perp to grain values and see if you need to dig into it at all. I think it would be safe to use our white oak group design values with your white oak. I'm assuming #2 grade. Compression perpendicular to grain is limited to 475 psi. Parallel to grain it is 800psi and I'm just too lazy to write it out, Mr Hankinson sez at 22.5° it is 727 psi allowable. If anyone is interested in how holler  .

It looks like the bent spacing is 8' and with common rafters half the load is on the truss, half on the plates soo, 40 square feet x 70 lbs per square foot=2800 lbs on the truss  :D I think we're ok.

Kingpost truss calc here, you have a load to plug in;
Untitled (forestryforum.com)

Heeljoint compression (top chord axial compression) controls in that joint vs bottom chord tension so we need to resist about 1485 lbs of compression. We can handle 727 lbs per square inch safely. 1485/727=2.04 square inches req'd... squat.
You are 6" wide.. 2.04/6=.34" deep, obviously go deeper with the notch, at least 1.5" x 6" wide so it cant jump a tad and slide. That is also why just about every one of these has some form of clamp or clasp or mechanical connection. The notch depth limit is 1/4 timber depth out near the ends of the bottom chord.

The webs are decorative in this arrangement there is no bending load on the top chord, that is carried by the commons. The webs can be omitted.

Thanks as ever Don. I think that I might start to sleep easier now! It seems then that the webs/struts can be entirely omitted so that would take care of the wedged half dovetail assembly issue - well spotted  :D

[Planeiron]

Pam and I were in a restaurant today in a building built in the 3rd quarter of the 19th Century.  I was reminded of this thread when I saw this "coffee table".

It had been taken out of part of the building that the restaurant is in when they opened up part of the second floor to make a high ceiling over the dining room.

Notice the low depth of the mortise.  We had an interesting discussion with the owner when I explained why the mortise was not so deep, and why there was no through tenon.



 

That's really nice and really quite shallow... Coffee table could be something that I could do if I make an expensive layout mistake...  :D
I guess this had metal work joining king and tie. Looks flat on those possibly 45 degree end cuts.

[Tom King]

The owner said the building was a hardware store for over a hundred years, and there was stuff hanging from that piece and the beams that were sitting on it.  She thought that's what the holes were for.

[Planeiron]

Kingpost truss calc here, you have a load to plug in;
Untitled (forestryforum.com)

Heeljoint compression (top chord axial compression) controls in that joint vs bottom chord tension so we need to resist about 1485 lbs of compression. We can handle 727 lbs per square inch safely. 1485/727=2.04 square inches req'd... squat.
You are 6" wide.. 2.04/6=.34" deep, obviously go deeper with the notch, at least 1.5" x 6" wide so it cant jump a tad and slide. That is also why just about every one of these has some form of clamp or clasp or mechanical connection. The notch depth limit is 1/4 timber depth out near the ends of the bottom chord.

I think I'm going to use the top chord heel connection as suggested. In terms of mechanical connection is something like that suggested in the image below sufficient only perhaps longer? Is the diameter

 OK and what about number per join? Top chord is 6x6 and bottom 9x6. All green oak so will be stainless steel. We don't have "Log hogs" in UK so will need to source something similar 

[Jim_Rogers]

Some engineers will require at least 2 or 3 inches into the lower timber. And some will specify at least 2 per rafter. 

I'm not sure in your case but that is what has happened to me before.

Jim Rogers

[Don P]

I'm going to use this to show something about design thinking as well.

"I've seen this done in that situation". That was pretty much how design was done through the 1700's. it's also known as "Empirical Design." It often works, knowing where gets tricky for me.

"Rational Design" is what came about as loads went from a wagon on a bridge to a train. Most of it gets down to, Quantify the loads and provide resistance.

What is it you want that connection to do?
Assume the notch has sheared.
You have figured out what the tension in the bottom chord is at the design load when designing that notch.
The screws come with tables of strength. This is in withdrawal, at some angle to grain.
There lies the path  . I have my doubts it'll work in much more than a pretty small truss.

Hmm, they have a deck screw for going through the rim, railing post, and a joist or blocking and then though a T nut type of end nut. This develops higher strength with a piece of ugly in view that needs a plug.

[BrentH]

You have probably taken action by now already, but I thought I would point out, on page 28 of Will Beemers "Learn to Timber Frame" the author specifies that the plans are specified to handle 50 lbs/sqft of snow load. It seems like your tiles still come up well short of that weight, though I may be missing something. 

I'm currently trying to build his plans as well, but my roof load is closer to 120lbs/sqft, so Im having to do some heavy alterations!

Brent

[Planeiron]

Hi Brent. I built my last barn using the Beemer book and it turned out great. I was worried about load and spread on the posts of this next one so the folks here provided some great information. Lots of ideas and ways to get to the end goal. I've started making English tying joints which will have 3 KP trusses. What are your plans for the roof?

[BrentH]

I have not yet decided on what I am going to do. I was hoping to avoid a king truss, as I will be putting a window in the gable to help ventilate the sleeping loft that will be up there. Most simple (least joinery and fewest loads to calculate) would be just to make a cruck arch that transfers the ridge beam loads down to the tie beam. I will be oversizing all of the timbers (black ash) in the structure, but my biggest hesitation is how much load the wedged half-dovetail joint can sustain.