# Wind load distribution on a sawtooth steel roof

I have a steel structure (see picture below) and its roof is a type of sawtooth roof. The plain/flat part is from glass and the inclined surfaces are some kind of pergolas, but I prefer them to be from steel sheet. (Pergolas are not good for areas with snow.)

My question is, what is the wind load distribution?

In my opinion there will be no wind in the direction of 0 degrees (because there are existing objects which prevent forming of the wind), only in the direction of 90 degrees. Also I'm not sure that the object will be fully closed (with facade walls) or partly open. I'm aware that this is important (the surrounding), so we can make solutions according to the two situations.

• How do you figure the wind stops at the elevation of the roof edge? It continues, so the exposed surfaces (horizontal and vertical) will have wind effects acting on them - windward and leeward pressures and uplift. Please get your hands on the governing building code to learn "how to".
– r13
Sep 9 at 16:25
• It's my mistake. The objects (left and right of the main object) in the section above, are higher (almost double) than the main object. That's how I know (at least that's my understanding) that there is no significant wind load from that direction.
– Xa0c
Sep 9 at 19:39
• Note, wind is dynamic in nature and is similar to water flow, it can overtop a barrier. Depends on the distance between the left and right structures and the central structure, the pressure can either be greater or lower than the windward pressure. When designing for the wind, be prudent, as we don't really have a good understanding of it except for a few standard shapes.
– r13
Sep 9 at 19:55
• The left object is about 3-5cm apart from the central object (so the central object can be autonomous object) and the right object is apart less than 1m. You are apsolutely right that it's a complex matter. However, I think that in this case, the wind load from the perpedincular direction (90 degrees) will be the dominant and most important for calculation.
– Xa0c
Sep 9 at 20:05
• It is totally up to you to interpret and use your design code prudently. My last remainder is the wind blows in all directions when it hitting a tall wall, it will flow over on the top edge (as mentioned before) and it will have a down thrust component that affecting the structure below. Again, it depends on the geometries.
– r13
Sep 9 at 20:15

## A foreword

There are two issues that you need to consider :

• standards and norms (legislation for your specific country)
• what really happens with wind.

When you are building you should consider both, but depending on the structure you might focus more on one of the two.

## standards and norms

Each country has its own standards and norms for buildings. For example in europe you can have a look at Eurocode EN1991-1-4 (Action no structrues - general actions Part 1-4) and the annex particular to the country you intend to raise the building.

In particular section 7.2 (pressure coefficients) is the most relevant to your problem (although you might need to do some work because the arrangement is not common).

What you need to be careful in this scenario, is that you need to consider all possible directions of the wind, especially the directions with the worst scenarios.

## What really happens with wind

1. The first thing that sprung to mind is that your drawing has the 0 degrees perpendicular to the building.

In real life, the wind usually has a strong preference with respect to direction (where I am from for example strong winds are usually from north-west and north and one month in the year from the south. If I were erecting a structure, I would take that directionality into account.

1. Its not only the downward pressure that you should worry about (not as common but still important if your area has that type of weather)

Usually the downward pressure is more important. However, especially with seesaw roofs (not only them actually), the upward pressure (or more specifically the pressure difference between the top and bottom of a roof) can be very important. Particularly if you are planning on wooden structures. Again, this becomes even more so important when you are building in areas with hurricanes.

According to Eurocode, the wind from the 0 degrees will definitely have some downward pressure. This is because the wind will create vortices over the pitch roof, and the turbulent flow with create a pressure difference, which will create a force.

Additionally, the 90 degrees wind direction as you guessed will also have a wind pressure.

The exact wind pressure values and/or the procedure is a matter of an experienced engineer that will see the requirements for the local building codes. IMHO it's not a possible to provide an estimate without inspecting the actual blueprints/location.

• First of all, thank you for your elaboreted answer. On the other matters. Actualy, the picture is wrong. The objects near the main object (left and right) are 2x higher than the main object. That's why I think that in that direction wind load is insignificant. On the other hand, the 90 degrees direction will be important. The surounding is urban zone, so the object is lower than most of the neighbouring objects. Also I'm well aware of the national regulations in my place, but unforunately, they are not very clear for this kind of roof. That's why I asked for help.
– Xa0c
Sep 9 at 19:43
• Also, before posting here, I saw the Eurocode (EN 1991-1-4: 2005), but as you noticed, there is no similar situation and frankly I don't feel very confident to make some combinatios of different sketches on my own. So I know that we can't discuss numbers (values), but we can discuss directions of wind load and discuss whether it will be suction or pressure, if you get my point. And I think that the easiest way will be - with pictures. For example (in words) if the wind comes from this direction - the wind load will look like this etc.
– Xa0c
Sep 9 at 19:52
• @Xa0c like you said the example does not exist so clean cut in Eurocodes (if they are relevant to your situation). However, Eurocode offers a methodology for that, which is not easy and requires technical education, experience and time to perform and get a number of the wind load. If on the other hand you are more interested in the wind load distribution, then you can setup a CFD analysis and obtain (at least) qualitative results (again caution in that you need significant expertise AND experience to do that). Sep 10 at 6:15
• I have technical education, I'm а civil engineer. But that's very wide spectrum, so in this specific area i can't draw my own conclusions, that's why i was hoping that some of you have faced very similar situation (if not the same - when it comes to roof appereance), so you can share your personal experience. However, thanks for your effort and advices.
– Xa0c
Sep 10 at 11:59
• Unfortunately, my experience is with solar trackers so it was a different can of worms. Sep 10 at 12:01

Below are contours of wind pressure on the mono-slope roof and sawtooth roof. Note the direction of the wind is from left to right, and the high edge of the roof is on the left.

In the US, $$C_p$$ is the external pressure coefficient. Please check your design code for the equivalent coefficient and how to apply it.

Pressure diagram of a mono-slope roof (per ASCE7):

Downward pressure on lower structure:

Wind can blow in any direction, the influence area and distance of a wind stream and the intensity at the contact surfaces, except experiment through the wind tunnel study, are of anybody's guesses. This sketch is provided for your further thinking. (Note the height of the wall and its distance from the building will influence the wind flow and intensity to a great extend. You could be right that the effects other than the 90 degrees case are negligible. But we have no way to prove it.)

• Thank you for you answer. But as you noticed, there is a big difference between one kind of sawtooth roof and another, it's not just matter of how many spans the roof is made of.
– Xa0c
Sep 9 at 19:56
• My answer was intended to draw your attention to the potential of wind distribution on this type of roof. For your special and rather complicate roof configuration, unless there is a wind tunnel study, nothing can be made certain, but I would refrain from claiming "there is no pressure".
– r13
Sep 9 at 20:03
• Don't get me wrong, i'm grateful for your effort. All that i'm saying is that in my opinion the 90 degrees direction will be more important in these case. 90 and 270 degrees. And i'm not sure on which part of the roof will be suction, on which pressure. Maybe together we can come to some acceptable solution.
– Xa0c
Sep 9 at 20:10