Thermal expansion

[Legion]

I'm thinking of building a roofed deck. Since I love working with steel, that's what I'd like to use. The area could be maybe 8m x 6m. I'm aware that beams that long will expand, by several mm (I'm in Perth, so a maximum of 40° temperature range). What I'm curious about is if I should design slip joints into it or if it's OK to just weld it all up as one monolithic structure. If I concreted the uprights in with a single structure, would the expansion be enough to crack the concrete and cause problems over time?

[Styx]

Whether temperature effects will be a problem depends upon the type of structure and how it is constructed and anchored.

If, for example, it is a steel roof structure on flexible steel columns, it is unlikely to be a problem.
However, if the steel structure is confined between rigid concrete abutments, or between buildings, it could cause problems.

Steel has an approximate thermal coefficient of expansion of 0.013 mm per metre length per degree Celsius.

For example, a structure 8 m long with a temperature change of 40 Celsius degrees, the change in length would be approx:
0.013 x 8 x 40 = 4.2 mm

However, steel exposed to the sun seems to get much hotter than the ambient air temperature, often too hot to handle.
So the actual temperature of the steel could be much higher, say even 150 C.
The temperature range should be considered from the temperature at construction to the extremes of high and low.
So, say construction is at 25 C and the lowest expected temp. is zero, and the highest expected temp. 150 C,
the range would be -25 and +125.
In the above example the range might be then 125 Celsius degrees instead of 40 Celsius degrees.
The expansion would be 0.013 x 8 x 125 = 13 mm, and the contraction 0.013 x 8 x 25 = 2.6 mm.

The above estimates are total length changes, so if the centre of the structure is considered stationary, the end changes would be half the above amounts.

So the top of a slender steel column (embedded in a concrete footing) might need to move sideways, say, 13/2 = 6.5 mm due to temperature effects. Whether this is a problem or not depends on the ability of the column to sway this amount.

Styx