Continue to Site

Eng-Tips is the largest engineering community on the Internet

Intelligent Work Forums for Engineering Professionals

  • Congratulations pierreick on being selected by the Eng-Tips community for having the most helpful posts in the forums last week. Way to Go!

Three-Sided (U-shaped) Rooftop Screen

Status
Not open for further replies.

dold

Structural
Aug 19, 2015
613
I have a 3-sided prefab roof screen to accompany new RTUs (see attached pdf sketch) which will replace existing RTUs

Haven't been able to find any threads about this particular situation - some similar questions about shielding, but I think this warrants a new thread..

How does the wind act on the shielded screenwall in the north/south wind scenario(see sketch), and the SE/NE oblique?

For the sake of time I'm inclined to just design for a standalone wall (rooftop equip/structure/appurt.) and double it for reactions at base of the RTU, even though 150% still sounds conservative. I'm looking for loads on my framing/curb holddown - not the screen design. This is obviously overkill but quite frankly "sake of time - not money" is a client-based prerogative. Don't worry...i'm off the clock.

I'm good with my wind loads on single surface/standalone RTU (ignore seismic for now), but how does the air flow behave?(see sketch for geometry of screens) I've read rules of thumb that say 'around 1.5 to 2.0*Height_of_Screen', the flow will re-form/re-join/un-separate and around that limit one should consider full 'exposure' again. When does that apply: 4-sides, 3-sides, 2-parts? And if that's the case, wouldn't the windspeed affect the change in y-axis(vertical) velocity in a 2-D cut of the wall? 'Angle of refraction' of the flow as the flow passes the edges of surface seems like an good analog.

Rationally it almost seems like there would be an eddy creating a net negative pressure inside the volume of the 3-sided structure (even more-so with 4-sided(?)) considering this particular geometry, but perhaps the screens would need to be a little taller for a net neg-pressure to develop? And that theory would only apply to a gust normal to the north/west/south faces, particularly at higher windspeeds. And when we're talking about windspeed - as opposed to wind pressure - we are considering how windspeed will affect the behavior of the flow around the surface, i.e. how much wind hits the shielded wall at a given distance/separation, which in turn will effect some +/- pressure inside the volume. Lets just leave that theory as a topic of discussion - back to the question.

Are there any studies/threads/links/codes/commentaries that I cant find talking about this? I'd like to whittle down my calcs a little bit from ham-hand status. Has anyone really dug into this before? Links to old threads?

Otherwise what sort of thoughts do yall have? Other than "contact supplier"?


Thanks!
 
 http://files.engineering.com/getfile.aspx?folder=71899aee-431b-49f8-8eca-f771188ad74e&file=eng-tips_--_3_sided_roof_screen_sketch_[flat].pdf
Replies continue below

Recommended for you

First, I am confused. You are putting an open structure (3-sided, no roof) around a RTU and expecting the wind to behave differently on the RTU than if the screen were not there? The wind coming from the E/W direction can hit the RTU at full blast? I might be missing something because in your isometric, the white open side has a shadow like it would be adjoining a wall or something on the roof.

To save everyone from having to download your attachment, I am uploading into my post:
Capture_qunsok.png




Juston Fluckey, SE, PE, AWS CWI
Engineering Consultant
 
If this is for everyday, productive engineering work I would say that you are way overthinking it. Calculate the wind load on the screen as if the RTU was not there. Calculate the wind load on the RTU as if the screen was not there. In other words, neglect shielding effects. Done.

If this is for your own science project, I would engage a firm that specializes in wind tunnel testing if you are looking for anything more than whats in the code or rules of thumb.
 
obviously, there will be turbulence from the screen but without wind tunnel tests there is no way to accurately predict it...so the question would be ..would the presence of turbulence be more severe than assuming no shielding?....I agree with MotorCity that assuming no shielding would be a reasonable way to proceed....
 
Status
Not open for further replies.

Part and Inventory Search

Sponsor