How you design your Chevron Braces

[Forgot2Yield]

Hi there,
I've read a few forums on this topic and see that some people create models where chevron braces are not taking any axial load from gravity forces. My question is if there is considerable gravity load acting at the point where your chevron braces are connecting to your beam, (say we are talking about a machine frame where the heavily loaded feet stand directly above your chevron brace) and in reality a lot of this load will be transferred to the brace then what would be your preferred method of design:
1)design the braces for those axial loads resulting in large braces compared to the beam sizes
2)specify in the procedure that the braces be installed only after the beams have been fully loaded so that they do not take any of the axial

 

[AskTooMuch]

Design the chevron brace taking both lateral (wind/seismic) and gravity load from tbe equipment like your typical global analysis. Do a separate local beam/connection design without the brace for gravity/equipment load only.
This is written in standard design guide for vertical brace of EPC firms.

 

[ads0221]

Design the braces to take both gravity and lateral forces because most likely you won’t have control over the order they erect and set everything. We like to keep our unity ratios to a 0.5 max on WT’s rotated because of eccentricity, double angles and W beams is fine to go up to 1.

 

[human909]

Unless the bracing is extremely light then I definitely design the braces to take the axial load. I certainly don't design the beam for gravitational in the absence of the braces. That is akin to designing the cord on a warren truss to take the full bending load. It makes no sense.

If it is a tiny frame with minimal lateral loads then the lateral bracing just goes along for the ride. But I'm talking a 3x4x2.5m equipment frame.

 

[AskTooMuch]

Human909, have you had any experience in petrochemical industry to say it doesn't make any sense? It makes a lot of sense and the people who made these design guides have several years of experience in the industry. The reasoning is there are times vertical braces maybe asked to me removed during maintenance or turanaround of equipment. And it doesn't really cost much extra to design these few beams for gravity without the bracesa and good redundant design in this case.

 

[dold]

I certainly don't design the beam for gravitational in the absence of the braces. That is akin to designing the cord on a warren truss to take the full bending load

It is a requirement for seismic design. The intent being that during a design level earthquake we assume that vertical braces yield during the event and are no longer able to support the gravity load (or at least lose a large portion of their strength). So you must design the beam to support the full gravity load assuming the braces are gone.

 

[sandman21]

Its not yielding small amounts of drift will shed load to the beam and column. However, if you do have a large point load at the brace connections and I do mean large you should consider including that load in the braces. This is when considering seismic loading

 

[Forgot2Yield]

Thanks for all the info everyone. I will take both loading situations into account and design the braces for the axial loads and the beam for gravity/equipment loads without the bracing. AskTooMuch, do you know if there are any of those design guides of EPC firms you mentioned for free download online?

 

[human909]

Human909, have you had any experience in petrochemical industry to say it doesn't make any sense?

No, but where I work doesn't matter. Do you have much experience in the bulk handling of bulk materials? Does it matter?

It makes a lot of sense and the people who made these design guides have several years of experience in the industry.

Justifying something as because the design guide says so isn't the best justification.

The reasoning is there are times vertical braces maybe asked to me removed during maintenance or turanaround of equipment.

Well that does make sense, if those are the requirements. Thank you. In cases where that is required I too have allowed the redundancy.

It is a requirement for seismic design. The intent being that during a design level earthquake we assume that vertical braces yield during the event and are no longer able to support the gravity load (or at least lose a large portion of their strength). So you must design the beam to support the full gravity load assuming the braces are gone.

I don't know what code you are using but surely that depends on your specific code and your seismic design approach and assumed ductility?

EG, one person's chrevron brace might be another person's inclined column.

Certainly it is common to support 300+ tonne silos using chevron braces, where the beam above has no chance to support the load. And while you might look at that and this that is a soft bottom story, well yeah it is. That is a necessary requirement so it is designed as such.

Thanks for all the info everyone. I will take both loading situations into account and design the braces for the axial loads and the beam for gravity/equipment loads without the bracing.

No problems. Can't really go too far wrong doing that.

 

[TLHS]

Given that I regularly put chevron braces in specifically as a vertical load path, especially for retrofit work, it would be weird if I ignored the vertical load on them in other times when it's not convenient.

The braces are going to be a stiffer load path than the beam, in most cases, so you'd need some really tricky detailing to shed vertical load out of them.

 

[canwesteng]

I can see that a large ECP (starts with an f) would put that in design criteria, but that doesn't mean it's common practice or even logical. If you need the chevron not to take dead load, you can ask them to install the equipment first, but otherwise it is always designed for this. Given that any time bracing is removed without engineering input it has the possibility to create serious instability, the beam yielding when the bracing is removed is the least of your worries.

 

[human909]

Given that I regularly put chevron braces in specifically as a vertical load path, especially for retrofit work, it would be weird if I ignored the vertical load on them in other times when it's not convenient.

The braces are going to be a stiffer load path than the beam, in most cases, so you'd need some really tricky detailing to shed vertical load out of them.

Thanks for chiming supporting vertical load in chevron braces. Sometimes I start to feel insane when I get shot down by multiple other people.

But like many disagreements on eng-tips. It is a matter of different experiences and different needs. One approach barely fits all.

 

[dold]

I don't think anyone is saying to not design the braces for the imposed load. But for US seismic design codes (ASCE 7, AISC 341, etc) you can't rely on the vertical part of the seismic force resisting system to carry gravity loads for certain types of vertical bracing. Steel specifically. IIRC, braces are expected to have around 30% or 40% of their compressive strength remaining after a design level earthquake. Buckling of the brace and yielding of the gusset plates are part of the mechanism. Tension yielding of the brace the other component.

I'm not saying there's anything wrong with using a vertical brace or leaning column as a part of the gravity system like your hoppers, just saying that in some circumstances you can't rely on that brace's contribution support gravity loads. Emphasis on "in some circumstances".

 

[human909]

Thanks for elaborating on that dold. I might have a look specifically at those provisions to further inform me, they aren't applicable to me but more knowledge is certainly better.

Most of my chevron braces are primarily lateral. But I have done several were they are predominantly gravitational. When the latter is the case I've always kept a very healthy margin above code requirements. That margin is my 'sleep comfortably at night' margin. My clients don't complain, I save them money elsewhere.

 

[JoshPlumSE]

Chevron braces for wind resistance are very different than chevron braces for seismic resistance. The reason being that for seismic design you expect your members to experience demands beyond the normal loading you design for. And, you rely on the SYSTEM (i.e. Chevron bracing) to have ductility post member failure.

And, there is A LOT of good literature showing what happens to Chevron braces after one side buckles (during an earthquake). The buckled brace loses almost all of it's compressive resistance, but the structure still deforms in that direction. Which then means that the tension force in the other brace keeps increasing until it pulls down the "gravity only" beam.... unless the beam is designed to resist these additional loading.

 

[Forgot2Yield]

[/But like many disagreements on eng-tips. It is a matter of different experiences and different needs. One approach barely fits all.]
The more I learn the more this seems to ring true.
Just out of curiosity has anyone used or seen others use slotted holes in their chevron top connections specifically for the purpose of not having to design the bracing for the equipment/gravity loads?

 

[TLHS]

Bringing the braces to a common plate and then using slots from that plate to the beam is theoretically a thing. I've seen studies on that for earthquakes. I've discussed it for a retrofit once but never done it.

https://www.researchgate.net/public...races_with_Vertically_Slotted_Beam_Connection

 

[canwesteng]

Josh has described more what my experience would be in high seismic regions/categories. The beam is designed for an additional load at mid span separately. This is substantially different than designing a beam to span for dead and live without the brace though.

 

[JoshPlumSE]

Matty54 -

I have not used slotted bolt holes with a chevron. Nor have I seen it done. That's not to say there couldn't be a reason to do it. I just don't see a ton of "value added" to the project by saving a few buck on the braces it that way.

CastwestEng -

I just modified my post to make it clear that I was elaborating about the seismic braces.

Wind chevron's don't have much in the way of special requirements that I know of. Though, in heavy industrial projects (where the braces may get moved when equipment gets swapped out or such), I have seen the beam designed for the full gravity load. Though I have always seen the braces for BOTH the gravity and wind loads combined.