Making an extra strong beam with wood and metal

[DrKrettin]

I want to make a kind of pergola with a tiled roof on one side of the house, 6 m along the side of the house and 3 m wide. Usually, these pergolas have supporting posts every 3 metres, but I particularly want a span of 6m between the corner posts, which means I need a strong supporting beam to hold up the roof. A builder friend suggested that I get two 6m laminated wooden beams of cross section 100x200 mm and sandwich a metal sheet (thickness 3 - 6 mm) between them, bolting the three together along the length of the beam. This metal sheet would be vertical, thus effectively working as an I-shaped RSJ, but the whole thing would be much more attractive than just an RSJ. 

I have no idea whether I am over-engineering this. The roof weight will be 30 sqm of tiles at 50 kg per sqm = 1500 kg including roof beams, about two thirds of which weight will rest on this 6m beam. Any comments on these quantities would be appreciated.

What I do wonder about is something the builder claims is simply not a problem - what happens to this wood-steel-wood sandwich when the steel expands and contracts under temperature changes? Does the wood expand with the steel, or what happens?

Edited August 8, 2017 by DrKrettin

[Manticore]

Any expansion problem would be more likely to be caused by the wood expanding and contracting with changing moisture content. I don't think I would worry too much unless you live somewhere with long wet and dry seasons.

[studiot]

This type of construction is known as a Flitch Beam. Yes they are a traditional way to strengthen wooden beams.

https://www.google.co.uk/search?site=&source=hp&q=flitch+beam&oq=flitch&gs_l=psy-ab.1.0.0l4.1264.2745.0.5141.6.6.0.0.0.0.183.1030.0j6.6.0....0...1.1.64.psy-ab..0.6.1020...0i131k1.gFzds4YGAuw

Success in its working relies on substantial close (shear) connection between the wood and the metal, over the length of the beam.
This means many through bolts, tightened up over large area bearing plate washers.

Post some facts and figures for better design info, courtesy SF design services.

 

[StringJunky]

Quote

Disadvantages[edit]

Due to the high cost of labor, use of this type of beam has greatly declined. The advent of high-strength engineered lumber which uses modern adhesives and lower cost wood fibers has rendered this system largely obsolete. For example, compare the capacity of 2 beams spanning 18 feet: the allowable strength of a 3½ʺ × 11¼ʺ laminated veneer lumber (LVL) member has a tabulated allowable applied load of 398 plf (pounds per linear foot); compare this to a 2 × 12 flitch beam constructed of ½ʺ steel plate with two laminations of #2 SPF with an allowable applied load of 386 plf.

Engineered lumber is cut to length and installed similar to sawn lumber; the flitch requires shop fabrication and/or field bolting. This coupled with a much increased self weight of the beam (11.4 pounds (5.2 kg) for engineered wood vs. 25.2 pounds (11.4 kg) for a flitch beam) decreases the viability of the system. https://en.wikipedia.org/wiki/Flitch_beam

 

[DrKrettin]

All the above is really useful information, thanks everyone. Now I've got the name of the beam, I can do a little research.  Where I live (Tenerife) the biggest problem is supply of materials. Wood is not only expensive, there is very little choice of shapes and sizes, and so I have to make do with what is available.

Today, I had worked out exactly what I needed to make most of the construction from metal, using 100x100 box metal, 3mm thick. Off to the only company selling metal on the island, hanging around to be served, only to be told that they were out of stock, can't order it, no idea when it might arrive, bugger off.  They also denied that 6m lengths of steel for a flitch beam actually existed. I failed to explain in Spanish what I wanted to do, but here there is no customer support or any help whatsoever.

So off to the biggest wood supplier, where I found I could get 100x200 or 70x200 laminated beams, and very little else. But oddly what they did have was a piece of metal ideal to use for a filch beam, exactly what the "experts" denied existed.

So my idea is to use two 70x200 laminated beams and the filch beam to span the 6 metres. This would have to take the weight of about 1000 kg (tiles) evenly spread. I just can't imagine that this is not strong enough, but if anyone disagrees, please let me know.

[studiot]

Two things should be noted.

The economics of commercial construction will be very different from the economics of DIY, where labour cost is not usually counted.

The proposed flitch beam can never act like an I beam or RSJ.
The webs in such beams carry almost all the (vertical) shear stress, whilst the flanges carry almost all the bending stress.
The proposed design has no webs, which is why I asked for some facts and figures.

[DrKrettin]

Noted, thanks. My alternative is to construct the thing in metal using an actual I beam (which they actually have in stock, at least today), but the whole thing would look awful, and cladding it in wood would be time-consuming. I find it hard to believe that I need to do that. 

[studiot]

I am currently in the north of Scotland, but expect to be back home by the weekend when I will have the data tables and time to check calculations if that is any good.

I could easily envy your choice of domicile, except I found the strawberries oversize and tasteless.

Edited August 8, 2017 by studiot

[Bender]

Putting the strongest and stiffest material in the middle is, from an engineering perspective, a very odd thing to do. Putting a wooden beam between two steel plates is a lot stronger (about the same as an I beam with similar dimensions). You also don't need that much bolting all over the length of the beam, since the only purpose of the wood is to keep the steel plates apart.

Edited August 12, 2017 by Bender

[studiot]

1 hour ago, Bender said:

Putting the strongest and stiffest material in the middle is, from an engineering perspective, a very odd thing to do. Putting a wooden beam between two steel plates is a lot stronger (about the same as an I beam with similar dimensions). You also don't need that much bolting all over the length of the beam, since the only purpose of the wood is to keep the steel plates apart.

I assume you mean putting steel plates on the underside and topside of the timber and bolting through vertically as opposed to the original proposal

Quote

This metal sheet would be vertical,

How would you disguise the bottom steel?

Edited August 12, 2017 by studiot

[John Cuthber]

On 8/8/2017 at 5:56 PM, studiot said:

I am currently in the north of Scotland, but expect to be back home by the weekend when I will have the data tables and time to check calculations if that is any good.

I could easily envy your choice of domicile, except I found the strawberries oversize and tasteless.

The requirements for snow loading may also differ.

[studiot]

6 minutes ago, John Cuthber said:

The requirements for snow loading may also differ.

You are quite right the snow loading requirements will be different in Scotland, Somerset and Tenerife.

http://www.snow-forecast.com/resorts/Teide/6day/mid

That is why correct building load assessment includes a location.

Location also affects wind loading.

[John Cuthber]

Indeed- the point I was making was that it'snot just the weight of the roof you need to account for.
Someone might want to stand on the roof to work on it.

[studiot]

OK so I'm back in Somerset.

Can you provide some architectural setails to go with the material you put in the first post?

Is the roof single or double pitched, hipped or what?

How is the side thrust from the rafters resisted ?

Six metres is a very long length of structural timber to obtain, I note your builder suggests laminated.

What  fixing down arrangements are available?

What will be seen of the beams and rafters. Are 200 (high?) x 100 (wide) beams the max or could they be bigger?
They seem small to me from connection considerations.

With the loads discussed, I don't think you need steel. I think suitable beams could be made up on site from laminated facings (that don't have to be 6m long) enclosing spacer blocks of timber.

This arrangement would have many advantages.
Timber needs spread out connections.
It tends to relax under compression and loose some load transference capacity of connecting bolts bearing directly on timber.
There might be a difficulty fixing rafters to a timber steel sandwich only 100 wide.