Notching joists.

[Gilesda]

Hi all.
I want to drop the floor in a 2.4m x 3m room about 48mm. The current floor is suspended and sits on 102 x 47 sawn timber joists.
These joists sit in joist brackets that are built into the walls. The joists are 2.4m long but are supported in the middle on a wall that runs down the centre. So they span about 1.2m.
The easiest way for me to drop them 48mm is to notch out the two ends by 48mm depth. Leaving 54mm of timber to sit in the brackets.
Is this a really bad idea? If they spanned the full 2.4m width I wouldn't even have considered it. But I'm wondering if it would be ok over that 4ft distance.
Thanks.

 

[johnny2007]

Bad, very bad idea.
They will split like an easy madam's legs...

 

[tony1851]

That will work, as long as there are no obvious splits or loose knots at the ends.

 

[johnny2007]

No way.
Try it on.
Notch one end, rest it on a brick and rest the 1.2 length on another brick.
Now jump on it.

 

[tony1851]

No way.
Try it on.
Notch one end, rest it on a brick and rest the 1.2 length on another brick.
Now jump on it.

It will work, and can be proved using the data from the BS Code for structural timber design.

 

[JohnD]

I want to drop the floor in a 2.4m x 3m room about 48mm.

That's interesting.

Why?

 

[johnny2007]

It will work, and can be proved using the data from the BS Code for structural timber design.

Does it say you can notch timber at one end and rest in on a hanger?
Come on man, there must be a limit to bodging things.

 

[tony1851]

It's not a bodge, and it works with the end in a hanger.

The only issue at the end of a timber joist is the shear stress. As the span is short, the load at each end will be very low, and the remaining section of timber (47mm wide x 54mm deep) will be adequate to take the shear stress, with some to spare.

Remember also that the boarding helps to distribute the load to adjoining joists, which further reduces the shear stress. Even without that 'load-sharing', the remaining timber will be fine, subject to there being no obvious splits etc.

 

[Gilesda]

That's interesting.

Why?

If you look at the pictures I posted you can see a ledge of wood which is the house floor level. The floor I want to drop sits 48mm above this level. The floor is in a conservatory that I'm doing a few changes to.
The reason it's too high is when it was built it wasn't realised that the floor in the house sits below the dpc. I don't know if that is common but that's how it was built late 60s. So the joist hangers sit on the dpc and the floor is too high.
The patio doors that you can just about see the bottom corner of in the pictures was originally sat two bricks higher than that. So you had to step over it to get into the conservatory.
I dropped the patio doors down to floor level and now the floor being higher is a problem. Well by problem I mean it looks a bit crappy. If I leave it the edge of flooring in the conservatory will show through the glass.

 

[Gilesda]

That will work, as long as there are no obvious splits or loose knots at the ends.

The joists are very good I can't see any loose knots anywhere.
As an aside can you still buy sawn timber? I only seem to see cls these days.

 

[Bouy]

If you actually did the calc for a c16 47 x 54 joist it shows that it fails strength wise by 5% and deflection wise by 35%, shear wise its OK by 50%. Obviously the fact that its still 102mm deep in the majority of the joist suggests that strength and deflection would not be an issue. The only bug factor is that where the notch is all sorts of secondary stresses start to occur and this can lead to the propagation of splitting planes and ultimately to failure. Possibly by not notching at 90 degree but splaying off at 30 degrees (say) may help to prevent excessive secondary stresses occurring

 

[jacko555]

I'm not a builder or structural engineer, however, if there are valid concerns about secondary stresses on the notch points, surely this could be solved by just adding additional joists and hangers?

 

[tony1851]

@Bouy, as you say, a 47 wide x 54 deep joist would fail in bending stress and deflection, but be OK in shear (permissible shear stress 0.67 N/mm2, actual shear stress 0.28 N/mm2 based on 2kN/m2) so - yes - OK by around 50%.

But when notching the joist, the Code allows for the possibility of defects by reducing the allowable stress by the K5 factor of he/h = 0.52 in this case. So the allowable stress would be 0.52 x 0.67 = 0.34 N/mm2, a bit nearer the limit. but then we could also apply the load-share factor of 1.1, and depth factor (around 1.1) so there is a reasonable margin in the OP's case.

 

[tony1851]

I'm not a builder or structural engineer, however, if there are valid concerns about secondary stresses on the notch points, surely this could be solved by just adding additional joists and hangers?

Yes, that would be a perfectly valid way; double the number of joists - say @ 200 centres rather than 400 centres, and you halve the load on each joist and reduce the shear stress by half.

 

[Gilesda]

I though about fixing new brackets alongside the originals set at the height I want. So the joists move along a few inches from where they are now. Fixing new brackets though I'm not sure how id do that given the originals were put in as the walls were built.. I suppose I'd have to cut mortar out of the joints and slide them in.