UKC

I'm looking to make an indoor wall that is 6 metres wide, 30 to 35 degrees overhanging and about 3m high. I can't support it from the ceiling so want to bolt it into concrete walls at each end (proper concrete, not breeze block) and run a steel beam along the top edge (so 6m long steel beam). There will be 2 by 6 pine/spruce struts running perpendicular to the beam to support the 18mm plywood facing.

With the above design, I'm looking for help with a few questions:

1) does the beam profile matter? I'd rather go with a box cross section as it's easier to bolt to other things than an i-beam.

2) how beefy a bar should I use? I.e. what width and what thickness of steel?

3) The bottom edge of the wall will be firmly supported but would it be a good idea to have a second 6m beam running across the wall at half height? Or is this overkill?

I feel like I want to overengineer it a fair bit because a collapse while in use would likely be fatal.

The concrete walls at the ends are fairly old but they are a foot thick so I'm pretty confident they will not be the weak point.

I can't speak about steel beams with any expertise but it should be possible to find data somewhere on the relative stiffness of I-beams and box-sections of varying thickness. I think it wouldn't need to be that big and I'm sure if you size it correctly then a box section would be fine.

Having built my own wall I can confidently say that if you are using 2*6s then a second steel beam half way up will not be necessary. However using a bunch of pieces of 2*4 to connect the 2*6s horizontally helps to provide a bit of stiffness, and if you do it at the point where your sheets ofmply join, then it gives you some extra wood to screw the ply to and keep the joint flush.

You might need a structural engineer for this - having got a 6.5m universal steel I beam holding up my roof I can testify that such a beast weighs a (metaphorical) ton (maybe >50Kg/m) and the load from the self weight alone is substantial. You might need to hack holes in the walls and use a pad blocks to support the beam and that might require you to have the beam made in two parts in order to install it. 

On the other hand you could get away with an all-wood construction and a centre post.  If you've got the height or a decent wall behind you could even place that post away from the lip of the wall and use it to support other training kit.

Spanning 3.6m (1.5 panels) with wood is definitely easy to do.

However, if you are dead set on free spanning it then a home-made timber I-beam construction is probably easier to manage then a steel and you could have one to resist the vertical forces and one to resist the horizontal made from the same piece hidden in the construction. You can also get these supplied to size, they are commonly used for spanning large gaps in house construction.

Thanks. Good to know the middle beam won't be needed. I've not put in cross pieces of 2 by 4 in the design, partly because it will be completely jammed between 2 walls, so it will not really have any way to move or deform but it's a good shout putting some in on the ply joins.

Hmm... I can deal with a 50kg beam. My plan is to support it at both ends using a similar box beam bolted to the wall at 30 degrees with bolts that will probably go right through the wall and out the other side. So supporting it at the ends should be ok... I'm thinking the joint between the beams will be held with 4xM12 bolts at each end or something like that.

There is the option of just cutting a square hole in both walls, making the beam 6.6m and threading it through, so no side beams, but I'm not sure I can easily cut a hole like that through concrete.

There is no wall behind the board, the roof of the building is curved and insubstantial (walls only at either end). I really don't want to put a supporting strut in the middle, for various reasons.

I feel intuitively more comfortable using steel because its a single strut that holds the whole wall up and I'd rather not run the risk of it rotting somewhere down the line, and wood also tends to just snap, where as steel is going to bend first and let me know I need to stop using it before it lies down on me.

Do you know what the dimensions of the beam holding your roof up?

50Kg per metre - ~300Kg in total...

For working out the size of the beam you can find all necessary information in the below databook.

http://www3.eng.cam.ac.uk/DesignOffice/databooks/2002/structural.pdf

The load case is a combination of the 4th and 5th examples in section 3.5 - you can consider the dead and live loads separately and then add the deflections and stresses.

Box section will be much heavier than I-beam, and bolting to I beams is definitely easier. In fact you could just lie your uprights onto one of the flanges, using coach screws to prevent movement.

The key number you want from the tables in section 8.3 is the second moment of area. This is the I that you put back into the beam deflection formulae. Probably easier is to decide on an acceptable deflection, then work out the minimum I to achieve this, then pick a beam with at least that value.

Thanks. I'll see if I can work out those equations. I was hoping not to have to do learning uni engineering but I guess I'll have to give it a shot!

a 6m steel beam would be massively heavy and expensive. Could you not use LVL instead which would be much lighter and cheaper. Span table might give you some idea:

https://wesbeam.com/Wesbeam/media/Documents/Design%20Info%20Docs/WESB0362_e...

If I had that much space i would consider a full kilter board setup

You absolutely sure your walls are solid concrete? It would be highly unusual, it implies they were shuttered in place and concrete poured (really big job) or they are prefab panels like lots of garages. Either way it wouldn't be unusual to encounter steel inside.

Your foot thick walls, are they external walls? in which case they will likely be cavity (or rubble filled).... and your ceiling, is that just a ceiling or is there a floor above? Solid foot thick concrete walls with just a ceiling above seems like a very strange building. I'd just get a structural engineer around and pay, I was a builder for 20 years and never once had the arrogance to think that I could do a structural engineers job with a few weeks of internet research. Even if you can work out the downward pull there's a lot more to it because it may affect other aspects of the building. 

If you do use a steel it will be heavy, will have to be posted 100mm into each wall and will probably have to sit on a padstone. Best advice is to get someone who knows what they're doing to come and have a look, it'll definitely be fatal if your steel collapses because it's not installed correctly. 

I can't be 100% sure how they were built but drilling through them as I have already done needs a serious amount of effort.... The "concrete" is composed of something sticking the stuff together and a lot of small stones. It's an agricultural building that is about 50 years old.

They are external walls and it's an agricultural shed. I don't think there is a rubble cavity because it's quite hard to drill all the way through. The ceiling is just curved corruagated iron with a relatively light duty steel frame to hold it on.

If the beam is too heavy to support easily on the walls, I can back up that support by putting struts from the beam down to the floor at each end so it isn't weighting the walls.

I don't for a minute think I could do a structural engineers job without years of training but this is a fairly simple structure and I'm happy to overengineer it a bit to make sure I have a margin of error.  So, for example, I'm pretty sure I could design a safe swing set with a little internet research, I'm pretty sure I wouldn't even want to try to design a safe rollercoaster without years of professional training and experience.

I know the sort. I would go for putting the beam 100mm plus into the walls on pads (concrete block) which act as bearers spreading the load in the wall, but are also easier to level than a rough hole. 

You might be surprised at the amount you need to remove at one end to swing the beam in. 

I had this prob with a home wall where I couldn’t use the roof trusses.
The pragmatic thing I did in this situation was to split the wall into 3.6m and 2.4m sections which matches panel sizes nicely and put in a supporting post. With those dimensions, you can do it all with timber. Run your 2x6s down to a foot plate bolted into the floor and it’s nicely triangulated. Ok you don’t have an uninterrupted 20 foot board, but that’s a minor problem.

Fair enough it is an interesting building. In that case could you not just bolt two timber rails on the back wall and tie it all back to those at 600mm intervals? There's plenty of commercial walls which build out further and higher without the need for a steel support beam. The only person who could tell you for sure would be a structural engineer though.

Can you bolt into the back wall? If so I would run a timber beam along the back at 1.5m and 3m heights with horizontal beams coming out to meet the wooden supports the ply fronting would be attached to.

A single human being isn't that heavy so the danger of over-engineering is you make a structure so heavy it cannot support itself.

So, the building is a rectangle with walls on all sides up to about 80cm. The end walls are full height. The side walls stop at 80cm and turn into a corrugated iron semicircle roof (if you Google "nissan but" you get the approximate shape, only my one has short concrete walls at the bottom of the sides. I just can't get a link to paste here that isn't 1000 characters long).

I want to put the wall along the length of the building, so there is no wall behind to attach it to (if I use the short side wall the anchor point would be so low that the leverage forces would be terrifying).

Is the top of the 3m climbing wall at the apex of the curved roof?

If not could you have a central longer 6 x 2 going higher than the wall and then support this with a timber  6 x 2 prop butted against the base of your 80cm high wall?

I think making it a single span will be a lot more costly than splitting it with a support. 

I would have thought that if you split it a long scaffold pole would suffice. Popped through each wall it won't snap. 

Having been in a nissian hut or two I see the issue!

A supporting framework spanning one side wall to the other may be the answer - but instead of a single beam a truss system could achieve the same stiffness for less weight. I appreciate as well the space behind the wall is minimal, due to the curved roof, but to make it stiff, ideally you would have some depth in your structure, so it isn't just a flat surface.

The whole thing will stiffen up when you add the ply too - monocoque is the term when building aircraft frames I think.

That's a pretty cool space to have but I can see your problems. If you go with a steel I'd definitely advise getting a structural engineer around because you'll need those concrete walls assessed before you add anything.