wood roof bracing?

[mats12]

Hello, I dont have much experience in this field, so Id like to hear some opinions.

Id like to know if there is any need to brace the columns of the roof (marked as red circle in the picture).
Are wood columns OK, and if so, do I need some additional bracing? Columns are anchored in to concrete slab.
What about concrete columns?

 

[EngineeringEric]

Hate to be this guy, but... Do the timber columns have sufficient capacity? Do you know how to determine that based on timber properties? How is the load entering the column (bearing, hanger, other?)? How do you transfer the load out of the column? there typically is no reason to make these concrete, steel if necessary, concrete if you want the look.

I am concerned by the placement of your columns being off center of those live valley beams, the beams(green) you are supporting may not actually have load pending your rafter/collar/thrust design. Do you have a real ridge board is that just a ridge plate?

So your question of bracing. If the column is properly attached to the rafter system, or beam, then it is braced at the top by the beam. Lower it looks like you are braced in one direction by the wall (what braces the wall? So do not know your Lb1 or Lb2 but these will govern the size of your column.... i digress, so the top of column bracing, if this is braced by the beam then you need to verify that the bottom of beam won't roll or be a hinge to the column (think collapse or lift off). the top of your roof beams/rafters/ridge/valley is braced by your roof sheathing assuming that is adequately connected to your lateral restraint system.

Honestly, none of this is complicated and you can verify most of these with judgement but you do need a clear well defined load path. If this isn't something you are comfortable with designing then as always, get professional help, pay them, and learn from their work.

 

[molibden]

Columns are ok. Just make a correct connection top and bottom.

This roof needs structural sheating on top of rafters or steel strap bracing, because your main roof beams are not braced by walls. You need to transfer lateral forces from roof to external (low) walls.

Is this masonry or timber frame strucure?

Looks like you have internal wall under ridge drawn directly to the ridge. This will not get constructed and the contractor will terminate it below horizontal ties.

 

[oldestguy]

This basic layout of rafters and "beams" looks goophy, crazy. There appears to be no useful purpose for the "beams". With rafters sufficiently strong enough to carry the roof load and secured at the lower ends for thrust, it would be pretty standard construction and no "beams" needed.

 

[HouseBoy]

HOLD ON THERE oldestguy!
"With rafters sufficiently strong enough to carry the roof load and secured at the lower ends for thrust" ??!!
THAT is a tall order for the graphic depiction that I see!

I don't see anything that would suggest to me that the walls are anything that could resist the substantial thrust that I would expect there, let alone the rafter "strength" and the connection that would be needed. Maybe they are but that would require some special walls (to resist thrust) I think.

It looks to me that wood columns would likely be adequate.

EE - The valley beams will have some load but that will be much less than the two perpendicular beams that are "crossing" each other (at the column locations.

The way I understand the illustration, the parallel roof beams will be doing the vast majority of the work. I'd be curious about the size of those beams. Just guessing, but I'd say deflection of the roof beams will be a "thing".

 

[molibden]

This configuration of roof beams and rafters is standard EU practice for residential homes, so it could look a bit odd to you guys from US.

 

[EngineeringEric]

House guy. If the rafters are built the way they are modeled, they will have a tension thrust collar tie. if the rafters have sufficient bending capacity and tension capacity in the tie then there is no thrust (minimal) if that is the case then those green beams are not taking any load however the live valleys are, which are also cantilevered past the post towards the ridge (not supported near the ridge).

I don't think it is unconventional for us in the US, but i would say there are minor issues or details which look off... that may be a states/EU thing.

 

[HouseBoy]

Eric, based on relative stiffnesses, I do not expect the collar ties to develop any tension in them before the rafter want to deflect out and cause thrust on the sidewalls. Presumably before that happens, the green beams will start supporting the vertical load (thus negating the thrust).
The collar ties are too high up on the rafters to be effective as tension members. Of course it is theoretically possible to proportions the components so that the forces don't follow my scenario but, based on the appearance of things in the sketch and my expectation of the ordinary proportions and member sizes, I think the great beams will be doing the bulk of the work.

 

[EngineeringEric]

Unless we are working in the center span of the green beams that have already deflected 2"(5cm) under self weight alone on a 10m span...

I was only trying to point out that if the elements are stiff enough then the green beams are not needed. if the CT is 1/3 down from ridge it isn't too unbelievable to utilize what appears to be a bulky rafter in strong axis bending. I've had to design rafters like this multiple times in vaulted rooms without the 'green beam'. And why else would we put a collar tie on each side of the rafter other than to handle additional tensile connection transfer.

I believe the above problem requires a full on stiffness matrix to solve but i am sure we all would design either or to handle 100% of the load based on residential budgets!

(still nervous about the live valley)

But we can agree that it is doable!

 

[molibden]

Well if you put "green beams" away then the "collar tie" is in compression and you get big horizontal forces at the supports pointing from the inside out.

 

[Mellimac]

Re: OP original question: As long as you have a sufficiently sized post, not seeing the issue. And the beams are resting on post, and properly attached with suitable hardware. The distance from those load points to concrete(?) or shear walls is what? I do hope you plan to block between top of rafters to limit racking? General rule of thumb for collar ties is 1/3 down from apex-wall bearing.

As for the style, those beams remind me of purloins (just wrong side). I would be inclined to have a stout ridge beam, as you have multiple load points (perpendicular walls) to stick in posts. Then you can get rid of those 'beams' or not. I'd still keep the collar ties, but they'd see less tension load with a well supported ridge beam. Plus, your valley beam (I see that as an issue as it currently stands) will have something substantial to attach to (ridge beam).
You are sort of mixing building styles with what I see above...

 

[HouseBoy]

Eric, I'm not seeing enough info to recognize 2" deflection of the "green" beams under self weight.
Where do you see that?
I agree with molibden.
1/3 from the ridge is too high for a collar tie to be effective as a tension member.
The rafters will deflect too much.
IF the walls can resist the thrust (which I would expect to be uncommon) then the collar tie will be in compression as molibden writes.

 

[Mellimac]

Houseboy - Are you confusing a collar tie with rafter tie?

https://www.nachi.org/collar-rafter-ties.htm?loadbetadesign=0

BTW - I also believe the rafters need some thrust protection at heel, but I'm assuming the walls are concrete, and the OP has enough perpendicular wall ties to overcome rafter thrust. That being said, I noted my dislike of the setup. A ridge beam would alleviate my issues.

 

[HouseBoy]

Perhaps I am but in name only.
I stand by my assertion that the collar tie will be in compression under gravity loads.
Thanks of the link. It describes things well. I note the collar ties are in tension under wind type loading.

The conversation I usually have is when folks want the color tie to do the work of the rafter tie. THAT is what I intended by the comments that in order to provide suitable thrust resistance, the tying members need to be very close to the bottom of the rafters.

In a ridge beam solution, the "live valleys" all be spanning even further and therefore would be a greater concern.

 

[EngineeringEric]

So I had to check HouseBoy... Given the spans noted, we would need a serious heavy timber rafter, like 7.5"x9.5"deep @ 48" o.c. with 24ft clear span to use a collar tie as thrust restraint with it located at 1/3 depth... So i will concede that my statement of "if rigid enough" still stands as valid; but i also agree with your statement of "probably not the case" as those are overly huge members. if we go to half depth it is a much more reasonable sized member (as expected by all).

I still like to think of it is a system... but for another day!

 

[spermar]

No need to brace the columns if designed correctly. The size of the columns may be so ridiculous that you need to do something, but I'd start by sizing the columns, unbraced, and going from there.

What's the height of the columns and what's the load to the columns?

 

[mats12]

OP here.
tnx for opinions!

Columns are cca 300 cm high. They are anchored to the concrete slab above floor storey.

I calculated green beams (Eurocode) and dimensions are: b/h = 24/26 cm

Collar tie are placed just above green beams. They are elements in tension (taking horizontal component of gravity load + self weight and snow load) - so the green beams dont have to - thats why ties are placed just above them . When there is a window load this elements (collar ties) are in compression.
Greem beams are designed as biaxial bending (self weight + gravity load as vertical load and wind as horizontal load).

 

[molibden]

Designing the green beams in biaxial bending is overly conservative, as the rafters with collar tie form an "A" structure, that is anchored at the wall supports, so no horizontal bending will form in the green beam.
You can model the "A" structure in one plane (rafters + collar tie + hinge wall supports) and model green beams as sliding supports in horizontal direction. You will see it works.

 

[HouseBoy]

I'm curious about your description of the columns. I'm not familiar with (or I don't recognize) some of the terminology.

Do the columns bear on top of a wall? Reason I ask is because the distance from the floor to the green beams looks to be more than 300cm.
At 300cm actual column height, I would not expect too much of a slenderness problem with "normal" sized components but if there is a column element bearing on the top of a wall, you most likely do need some sort of bracing there.

Also, I think designing the green beams for the lateral load might NOT be conservative overall.
I would expect that the green beams will not see the load (so, conservative in that regard) but the component that will see the load, needs to be proportioned for that.

Load-paths will be the key (for both gravity loads and lateral loads).