Thread: Advice sought on shed truss- camber or not?

As I have too much money and not enough to do, I decided to drop a tree on my machinery shed a couple of months ago. I'm fabricating a new truss and am wondering whether I should build some camber into it. The photo shows the damaged truss and the structure generally:
machinery shed damage.jpg

The span is just under 9 meters, height of the truss is 500mm and material is 50x50x3mm SHS with 25x25x2 SHS webbing.

I've welded up the long rectangle and cut the material for the webbing but, having zero experience with this kind of truss, I'm wondering if it is necessary / advisable to pull some camber (arch) into the main frame before I weld in the bracing. As you can see, it supports the centre truss in the middle of the span without any posts in the shed. I'm thinking maybe 40-50mm of camber but would appreciate input if anyone has any experience / knowledge with this.

my knee jerk reaction was yes, why not.
but the original manufactures wouldn,t have, imo
a camber will cause complications with the roof sheet replacement.
i would use the original as a exact design guide, meaning, use the same material (all mild steel i expect) , thickness, welds, spacings etc, etc. beef it up in areas you think appropriate.

Pete I think we used to call those a gillian truss?
There is very little weight on a roof like that in the scheme of things, the lift created by the wind flowing over the roof would be a much higher load in the upwards direction than the weight of the roof would be in the downwards direction.
So no I dont think you need to put a set in it, even if it did sag 10mm who would know.....

ps. check your other trusses by pulling a string line beneath them so you can measure/check those and make the new one the same, it should have very little sag being a steel truss. A 9 mtr timber truss has a bit of set, but wood aint steel....

instead of making the whole length trust, i,d think about fixing just the bent stuff. you can weld in plates to the original stuff so you just bolt in the new trust. m16 construction bolts will do.

With steel prices where they are at the moment, I’d probably cut and weld at the centre, plate the join for good measure.
I wouldn’t be putting any camber in it, as pointed out already you’re just making your life more difficult than it needs to be putting the corro on.

Thanks for the replies, it seems that the 'no camber' vote wins. I guess it won't deflect much over that span. Until I lift an engine from it. Kidding.
Having given this a lot of consideration and inspected how the shed is constructed I decided it would be far simpler to fabricate a whole new truss and lift it into place, weld it to the posts as has been done with the original. To cut-and-shut the existing truss would save some material but mean a lot more complicated task. I'm following the original design very closely except for the rust and the chook-poo welds.
I don't think a bit of arc in the front truss would make sheeting the roof difficult though, I spent a few years erecting sheds much larger than this and the sheets can be shrunk & stretched during laying to cover over a multitude of sins.

You could use span/250 as a rule of thumb to evaluate if your deflection is too high. If it's less than that you're probably thinking about solving a problem that isn't there.

If I'm interpreting the above post correctly, that would suggest 37mm camber.

I'm thinking there are a few possibilities;
If I give it no camber and it sags a little, no-one will ever know.
If I give it 40mm of camber and it sits there with a 40mm upward deflection, no-one will ever know.
if I give it 40mm of camber and it drops to dead straight, I will point this out to everyone I know.

Will ponder this over the next few days, my work week starts tomorrow so I'll get back to the truss on about Wednesday.

Why not drop a similar sized tree on the other bay, then turn both trusses upside down with built in camber. Easy and cheap.

Originally Posted by Pete O

If I'm interpreting the above post correctly, that would suggest 37mm camber.

Not really. It's just a simple indicator. In a broader sense it's just recognising that structures are flexible and some deflection is expected and totally acceptable.

More than 37mm would be an unexpected result, much less than 37mm probably normal and a grey area in the middle.

Originally Posted by Reidy41

Why not drop a similar sized tree on the other bay, then turn both trusses upside down with built in camber. Easy and cheap.

Why didn't someone suggest this before I finished dropping the row of trees? Can't believe I didn't think of it.

Anyone here have a CAD system that can do FEA (finite element analysis) - and is competent at using it? That's really what's needed here to answer the question.
A quick CAD model of a 9m truss, 500 high, made from the steel specified, the spacing of the bracing derived from the photo will give the total weight of the truss. Add to that the weight of the purlins to be used and the weight of the roofing to be installed as a uniformly distributed force vertically down. (So some info missing so far).
FEA will tell us the initial static deflection. That's probably the camber you should build in.
You can look up the wind zone your shed is in, and apply the wind lift force to the FEA model to see how far it will deform the truss up, If it is more than the static downward deflection, leave it straight. If not, use the static deflection as your camber.
Wind lift will simply reverse that and put the metal back to the shape you welded it in lying down.....
All this takes 5 minutes for someone who knows what they are doing. Sadly that counts me out. I can use CAD (Fusion 360 and Inventor) for modelling but have got totally lost every time I have tried FEA for anything I designed....

If someone here wants to give me a quick hand to learn how to, I'll do the model and FEA for this member's truss.

It's called a Warren truss. Apparently adding camber to them is quite common. Since it's the end one (that everyone sees) the perfectionist in me would want it to be straight, not sagging. There's quite a bit of info on the web about them. Perhaps call a steel truss-maker and innocently ask whether they would add camber if you were to engage them to make a replacement.

Some examples of a Warren Truss:

Originally Posted by shedhappens

check your other trusses by pulling a string line beneath them so you can measure/check those and make the new one the same, it should have very little sag being a steel truss.

The problem with that John, is it won't tell Pete if, or how much, the trusses were pre-cambered. If the string line shows the truss to be dead straight, it means it almost certainly was pre-cambered, but it won't tell him how much.

Originally Posted by Pete O

As I have too much money and not enough to do, I decided to drop a tree on my machinery shed a couple of months ago. I'm fabricating a new truss and am wondering whether I should build some camber into it. The photo shows the damaged truss and the structure generally:


The span is just under 9 meters, height of the truss is 500mm and material is 50x50x3mm SHS with 25x25x2 SHS webbing.

I've welded up the long rectangle and cut the material for the webbing but, having zero experience with this kind of truss, I'm wondering if it is necessary / advisable to pull some camber (arch) into the main frame before I weld in the bracing. As you can see, it supports the centre truss in the middle of the span without any posts in the shed. I'm thinking maybe 40-50mm of camber but would appreciate input if anyone has any experience / knowledge with this.

I think a number of the replies have, like I initially did, may have missed that line, and not seen it in the somewhat low res photo, and they're envisioning something more like this:

1653005498258_1400x1050.jpg

If you had this layout where trusses are one behind another all supported by the end walls, I'd say you're way overkill with the steel, as that is an 8m span made out of folded steel spot welded together.

But what you actually have is a pretty horrible bit of design as far as I'm aware - IF you think of it as a truss. Obviously it's stood up thus far fine, but trusses to my understanding are never intended to have a single point load applied to them, which is exactly what you have on that. I'd suggest this looks like a chicken and quacks like a duck - as in, it looks like a truss, but is built to be a beam. (Is it heavier gauge construction than the other three?). You could actually completely delete that 'truss' and simply put a single post in the middle to hold up the end of the centre truss that runs front to back, and the roof would function fine sitting on the three trusses that run front to back. I'm sort of wondering how much of the load was being carried by the damaged 'truss' in the first place, and how much was hanging off the roof timbers running left to right.

Anyway, the point is that whether or not camber is normally built into these things (I highly doubt it looking at mine) is completely irrelevant for your situation - what you're building should be considered a BEAM, not a truss, and work from that viewpoint. Or alternatively, if you can deal with it being there, save yourself a huge amount of effort and just whack a post there.