Installing a loft floor

[GMM]

Sorry chaps, a bit of a long post but I have tried to get all the essential information in (!) Undoubtedly
there will be something I've missed out, even so....

I have to build a floor in a loft. At first this will provide a platform for some roof timber repairs to be
carried out but ultimately it will be used for storage, not living space, so there are no formal BR
requirements but, of course, I’d like it to be robust enough that it doesn’t all wind up in the bedroom below.

At present, there is a lath and plaster ceiling, with rather wimpy-looking ceiling joists, so not the most
stable platform to work on. It’s in pretty good condition (as a ceiling, not a floor) so I’d like to keep it
that way, which means installing joists clear of the ceiling timbers. This will, of course, create a void
for a good depth of insulation.

So question 1: To work above the ceiling, I’m thinking of supporting it with boards on acro jacks,
possibly moving these according to where I’m working as I go and putting some boards on the ceiling
joists to spread the load (mostly me!). Does this sound like a sensible thing to do – ie any better
suggestions?

One side of the space is an internal wall and the other side is a chimney breast with single-brick party
walls in the recesses each side (these look pretty weak). Each end is the outside wall of the house
running up about 60cm from the ceiling to the eaves of the roof. Since the shorter span (about 4m) is
from the internal wall to the chimney breast/party wall, the joists for the new floor will run this way (ie
parallel to the outside walls).

The plan is to use ledger boards (wall plates if you prefer) and joist hangers to mount 50 x 200mm
(-ish) joists at 400mm centres. I know that socketing into the brick is favoured by some but that isn’t
going to happen for many good reasons. To avoid going near the single-brick party wall, I’ll have to
use trimmers across the (approx.) 1.8m recesses each side of the chimney breast, mounting joist
hangers on these, so this is where the greatest loads will be: With 4 or 5 joists on each trimmer, their
mountings will be carrying approaching 1/3 of the floor loading. The short ledger boards that these trimmers will mount on will, of course, run at 90 degrees to the others (ie along the returns of the
chimney breast and along the outside walls).

So question 2: What’s the panel’s view on the best way to mount the ledger boards, particularly those
that carry the greatest loads? I’m thinking of resin studs between each pair of joist positions (in the
past I would have used expanding bolts but this is Victorian brick), but would a single large stud (M16)
be better or worse than a couple of smaller ones (M12) and is there any benefit in supplementing
these, with anything further (eg a pattern of multi-montis)? For the short ledger boards, I’m thinking
four studs in a rectangle around each joist hanger.
Using more mountings spreads the load, but over-perforating the boards would weaken them (though
I suspect it would take a lot for this to be a big issue).

Lastly, to get decent access into the loft space (currently a 2x2’ hatch in a cupboard), I need to create a
new doorway. The only way I can approach this is from the inside of the loft (due to obstructions I
can’t practically re-position until the opening is formed). Normally, cutting a new opening would be
best done using strongboys to support the triangle of brisk above until the lintel is installed but I’d
hesitate to jack against the top of the ceiling and getting them through the hatch wouldn’t be easy.

So question 3: Instead of using strongboys, would a board (4x1 or so) fixed to each brick (multimonti
into the brick centre) above the lintel do an adequate job of stopping everything moving while I get a
lintel in? (A bit non-standard I know, but the best option I can think of in the circumstances!)

Cheers

[John Rumm]

On 17/10/2012 11:49, GMM wrote:
> Sorry chaps, a bit of a long post but I have tried to get all the essential information in (!) Undoubtedly
> there will be something I've missed out, even so....
>
> I have to build a floor in a loft. At first this will provide a platform for some roof timber repairs to be
> carried out but ultimately it will be used for storage, not living space, so there are no formal BR
> requirements but, of course, I’d like it to be robust enough that it doesn’t all wind up in the bedroom below.
>
> At present, there is a lath and plaster ceiling, with rather wimpy-looking ceiling joists, so not the most
> stable platform to work on. It’s in pretty good condition (as a ceiling, not a floor) so I’d like to keep it
> that way, which means installing joists clear of the ceiling timbers. This will, of course, create a void
> for a good depth of insulation.
>
> So question 1: To work above the ceiling, I’m thinking of supporting it with boards on acro jacks,
> possibly moving these according to where I’m working as I go and putting some boards on the ceiling
> joists to spread the load (mostly me!). Does this sound like a sensible thing to do – ie any better
> suggestions?

Nothing wrong with it, but you may be able to get around the problem in
other ways, if you can get a few joists in without needing to load the
existing ceiling too much. Once that is done you can get a temporary
floor to work off and do the rest.

> One side of the space is an internal wall and the other side is a chimney breast with single-brick party
> walls in the recesses each side (these look pretty weak). Each end is the outside wall of the house
> running up about 60cm from the ceiling to the eaves of the roof.

So the existing ceiling is not at the top of the wall as such then, but
is suspended a bit below it?

> Since the shorter span (about 4m) is
> from the internal wall to the chimney breast/party wall, the joists for the new floor will run this way (ie
> parallel to the outside walls).
>
> The plan is to use ledger boards (wall plates if you prefer) and joist hangers to mount 50 x 200mm
> (-ish) joists at 400mm centres. I know that socketing into the brick is favoured by some but that isn’t
> going to happen for many good reasons.

Socketing is rarely done these days it seems... hangers are the norm.

> To avoid going near the single-brick party wall, I’ll have to
> use trimmers across the (approx.) 1.8m recesses each side of the chimney breast, mounting joist
> hangers on these, so this is where the greatest loads will be: With 4 or 5 joists on each trimmer, their
> mountings will be carrying approaching 1/3 of the floor loading. The short ledger boards that these trimmers will mount on will, of course, run at 90 degrees to the others (ie along the returns of the
> chimney breast and along the outside walls).

You can get strong shoes that rawl bolt to masonry, and are go for 10kN
and more... so one of those at each end of the joist will carry a
significant floor load. I needed to do one like this on my loft at my
previous place.

The bottom left of drawing:

http://internode.co.uk/temp/beam-layout.gif

Shows beam F terminating on one of these shoes. It in effect takes one
end of the entire floor load.

> So question 2: What’s the panel’s view on the best way to mount the ledger boards, particularly those
> that carry the greatest loads? I’m thinking of resin studs between each pair of joist positions (in the
> past I would have used expanding bolts but this is Victorian brick), but would a single large stud (M16)
> be better or worse than a couple of smaller ones (M12) and is there any benefit in supplementing
> these, with anything further (eg a pattern of multi-montis)? For the short ledger boards, I’m thinking
> four studs in a rectangle around each joist hanger.
> Using more mountings spreads the load, but over-perforating the boards would weaken them (though
> I suspect it would take a lot for this to be a big issue).

I can't see much advantage going for M16 over M12 since they are both
going to be significantly stronger than the timber (especially as the
loading is across the grain).

The type of hanger used also makes a difference. With a suitably rigid
one that will not attempt to "unwind" and flip the wall plate over, most
of the load is simply pushing the wall plate hard against the top of the
masonry - there should be very little lateral load.

Another option would be the masonry hangers that are designed to be
built into a leaf of brickwork. These can be retrofitted by raking out
some mortar and then mortaring back in. They don't require any bolting
as such.

> Lastly, to get decent access into the loft space (currently a 2x2’ hatch in a cupboard), I need to create a
> new doorway. The only way I can approach this is from the inside of the loft (due to obstructions I
> can’t practically re-position until the opening is formed). Normally, cutting a new opening would be
> best done using strongboys to support the triangle of brisk above until the lintel is installed but I’d
> hesitate to jack against the top of the ceiling and getting them through the hatch wouldn’t be easy.
>
> So question 3: Instead of using strongboys, would a board (4x1 or so) fixed to each brick (multimonti
> into the brick centre) above the lintel do an adequate job of stopping everything moving while I get a
> lintel in? (A bit non-standard I know, but the best option I can think of in the circumstances!)

That ought to do it. Is this a single leaf or double?


--
Cheers,

John.

/=================================================================\
| Internode Ltd - http://www.internode.co.uk |
|-----------------------------------------------------------------|
| John Rumm - john(at)internode(dot)co(dot)uk |
\=================================================================/

[Tim Lamb]

In message <DbidnZ6gBIzoXePN...@brightview.co.uk>, John
Rumm <see.my.s...@nowhere.null> writes

>On 17/10/2012 11:49, GMM wrote:
>> Sorry chaps, a bit of a long post but I have tried to get all the
>>essential information in (!) Undoubtedly
>> there will be something I've missed out, even so....

Couple of thoughts... someone in here reinforced their existing joists
rather than install new.

The other.. don't allow the new timbers anywhere near the lath and
plaster. We once had a chalet bungalow where the previous owners had
stiffened up the old loft floor to take the new bedrooms load. These
joists spanned the outside walls but were not attached to the old
ceiling ones. Inevitably the new timber moved and pinged nail head
plaster off the downstairs ceiling.

--
Tim Lamb

[GMM]

On Wednesday, October 17, 2012 3:28:07 PM UTC+1, John Rumm wrote:
> On 17/10/2012 11:49, GMM wrote:
>
> > Sorry chaps, a bit of a long post but I have tried to get all the essential information in (!) Undoubtedly
>
> > there will be something I've missed out, even so....
>
> >
>
> > I have to build a floor in a loft. At first this will provide a platform for some roof timber repairs to be
>
> > carried out but ultimately it will be used for storage, not living space, so there are no formal BR
>

> > requirements but, of course, I�d like it to be robust enough that it doesn�t all wind up in the bedroom below.

>
> >
>
> > At present, there is a lath and plaster ceiling, with rather wimpy-looking ceiling joists, so not the most
>

> > stable platform to work on. It�s in pretty good condition (as a ceiling, not a floor) so I�d like to keep it

>
> > that way, which means installing joists clear of the ceiling timbers. This will, of course, create a void
>
> > for a good depth of insulation.
>
> >
>

> > So question 1: To work above the ceiling, I�m thinking of supporting it with boards on acro jacks,
>
> > possibly moving these according to where I�m working as I go and putting some boards on the ceiling
>
> > joists to spread the load (mostly me!). Does this sound like a sensible thing to do � ie any better

>
> > suggestions?
>
>
>
> Nothing wrong with it, but you may be able to get around the problem in
>
> other ways, if you can get a few joists in without needing to load the
>
> existing ceiling too much. Once that is done you can get a temporary
>
> floor to work off and do the rest.
>

Indeed: Once I get the wall plates on, I'll bring phase 1 of the joists in through the (at that point open) roof, fix the roof timbers, then stash the rest of the floor materials before the roof is sealed again. Then I'll work from that 'platform'. What I'd like to do is avoid wrecking the ceiling at an early stage (!)

>
>
> > One side of the space is an internal wall and the other side is a chimney breast with single-brick party
>
> > walls in the recesses each side (these look pretty weak). Each end is the outside wall of the house
>
> > running up about 60cm from the ceiling to the eaves of the roof.
>
>
>
> So the existing ceiling is not at the top of the wall as such then, but
>
> is suspended a bit below it?
>
>

That's right: A bit unusual (and I haven't measured it) but about 2' of vertical wall above the ceiling. Just for a change, something that might make a job easier (famous last words!)

>
> > Since the shorter span (about 4m) is
>
> > from the internal wall to the chimney breast/party wall, the joists for the new floor will run this way (ie
>
> > parallel to the outside walls).
>
> >
>
> > The plan is to use ledger boards (wall plates if you prefer) and joist hangers to mount 50 x 200mm
>

> > (-ish) joists at 400mm centres. I know that socketing into the brick is favoured by some but that isn�t

>
> > going to happen for many good reasons.
>
>
>
> Socketing is rarely done these days it seems... hangers are the norm.
>
>
>

> > To avoid going near the single-brick party wall, I�ll have to

>
> > use trimmers across the (approx.) 1.8m recesses each side of the chimney breast, mounting joist
>
> > hangers on these, so this is where the greatest loads will be: With 4 or 5 joists on each trimmer, their
>
> > mountings will be carrying approaching 1/3 of the floor loading. The short ledger boards that these trimmers will mount on will, of course, run at 90 degrees to the others (ie along the returns of the
>
> > chimney breast and along the outside walls).
>
>
>
> You can get strong shoes that rawl bolt to masonry, and are go for 10kN
>
> and more... so one of those at each end of the joist will carry a
>
> significant floor load. I needed to do one like this on my loft at my
>
> previous place.
>
>
>
> The bottom left of drawing:
>
>
>
> http://internode.co.uk/temp/beam-layout.gif
>
>
>
> Shows beam F terminating on one of these shoes. It in effect takes one
>
> end of the entire floor load.
>

Do you have any details of what you used for that please? They could come in handy here...

>
>
> > So question 2: What�s the panel�s view on the best way to mount the ledger boards, particularly those
>
> > that carry the greatest loads? I�m thinking of resin studs between each pair of joist positions (in the

>
> > past I would have used expanding bolts but this is Victorian brick), but would a single large stud (M16)
>
> > be better or worse than a couple of smaller ones (M12) and is there any benefit in supplementing
>

> > these, with anything further (eg a pattern of multi-montis)? For the short ledger boards, I�m thinking

>
> > four studs in a rectangle around each joist hanger.
>
> > Using more mountings spreads the load, but over-perforating the boards would weaken them (though
>
> > I suspect it would take a lot for this to be a big issue).
>
>
>
> I can't see much advantage going for M16 over M12 since they are both
>
> going to be significantly stronger than the timber (especially as the
>
> loading is across the grain).
>
>
>
> The type of hanger used also makes a difference. With a suitably rigid
>
> one that will not attempt to "unwind" and flip the wall plate over, most
>
> of the load is simply pushing the wall plate hard against the top of the
>
> masonry - there should be very little lateral load.

It seems there's definitely a case for a good snug fit on the joists, which would ensure the forces all resolve in the right directions.

>
>
>
> Another option would be the masonry hangers that are designed to be
>
> built into a leaf of brickwork. These can be retrofitted by raking out
>
> some mortar and then mortaring back in. They don't require any bolting
>
> as such.

I had thought of these but sort of dismissed them as it's pretty unlikely that the mortar courses are level enough across the space to end up with a level floor. That may not be a very good excuse nor that I didn't fancy spending too much time up there chiselling out mortar (!) but it just looks like they might not be such a robust fix as the timber mounted ones. I could easily be wrong there.

>
>
>
> > Lastly, to get decent access into the loft space (currently a 2x2� hatch in a cupboard), I need to create a

>
> > new doorway. The only way I can approach this is from the inside of the loft (due to obstructions I
>

> > can�t practically re-position until the opening is formed). Normally, cutting a new opening would be
>
> > best done using strongboys to support the triangle of brisk above until the lintel is installed but I�d
>
> > hesitate to jack against the top of the ceiling and getting them through the hatch wouldn�t be easy.

>
> >
>
> > So question 3: Instead of using strongboys, would a board (4x1 or so) fixed to each brick (multimonti
>
> > into the brick centre) above the lintel do an adequate job of stopping everything moving while I get a
>
> > lintel in? (A bit non-standard I know, but the best option I can think of in the circumstances!)
>
>
>
> That ought to do it. Is this a single leaf or double?
>

Just a single, so not too much to fall down !


Many thanks for your input John.

Cheers

[GMM]

Absolutely - If there was too much danger of damaging the ceiling, I thin it would make more sense to just take the lot down to start with, then put a floor and new ceiling into the space, which was what the builder did next door in the same place. They were dropping the ceiling height however to accommodate a loft conversion.

[meow...@care2.com]

On Wednesday, October 17, 2012 11:49:06 AM UTC+1, GMM wrote:

> I have to build a floor in a loft. At first this will provide a platform for some roof timber repairs to be
> carried out but ultimately it will be used for storage, not living space, so there are no formal BR
> requirements but, of course, I’d like it to be robust enough that it doesn’t all wind up in the bedroom below.

> At present, there is a lath and plaster ceiling, with rather wimpy-looking ceiling joists, so not the most
> stable platform to work on. It’s in pretty good condition (as a ceiling, not a floor) so I’d like to keep it

2x8 joists is unusually large for loft storage. Almost all old joists are quite able to handle storage. So I wonder if you've fully and correctly evaluated the situation, ie what the current joist dimensions are (all 3).


NT

[GMM]

Just going by the standard joist span tables for moderately loaded floors. Of courser a lighter joist would be cheaper/easier/more desirable. On the other hand, I'd rather not have the whole thing collapse!

[John Rumm]

On 17/10/2012 20:42, GMM wrote:
> On Wednesday, October 17, 2012 3:28:07 PM UTC+1, John Rumm wrote:
>> On 17/10/2012 11:49, GMM wrote:

>> The bottom left of drawing:
>> http://internode.co.uk/temp/beam-layout.gif Shows beam F
>> terminating on one of these shoes. It in effect takes one end of
>> the entire floor load.

> Do you have any details of what you used for that please? They could
> come in handy here...

Its was a masonry fix shoe. In fact there is a picture there of it:

http://internode.co.uk/temp/shoe.jpg

I just had a quick look at the design load, and it was 8.47 kN. From
memory the shoe was rated at 12 or 15 kN.

>> The type of hanger used also makes a difference. With a suitably
>> rigid one that will not attempt to "unwind" and flip the wall plate
>> over, most of the load is simply pushing the wall plate hard
>> against the top of the masonry - there should be very little
>> lateral load.

> It seems there's definitely a case for a good snug fit on the joists,
> which would ensure the forces all resolve in the right directions.

Yup. You could by the sounds of it simply bolt a timber to the wall, and
then nail hangers to that.

>> Another option would be the masonry hangers that are designed to be
>> built into a leaf of brickwork. These can be retrofitted by raking
>> out some mortar and then mortaring back in. They don't require any
>> bolting as such.
>
> I had thought of these but sort of dismissed them as it's pretty
> unlikely that the mortar courses are level enough across the space to
> end up with a level floor. That may not be a very good excuse nor
> that I didn't fancy spending too much time up there chiselling out
> mortar (!) but it just looks like they might not be such a robust fix
> as the timber mounted ones. I could easily be wrong there.

Its easy enough to add a timber packer under the beam where it goes into
a shoe to tweak they height it sits in the shoe. For that matter you can
even trim the bottom of the a small amount if needs be.

[Tim Watts]

Did the OP ever say exactly what his current joists were?

--
Tim Watts Personal Blog: http://www.dionic.net/tim/

"History will be kind to me for I intend to write it."

[Tim Lamb]

In message <qmi3l9-...@squidward.local.dionic.net>, Tim Watts
<tw+u...@dionic.net> writes

>meow...@care2.com wrote:
>
>> On Wednesday, October 17, 2012 11:49:06 AM UTC+1, GMM wrote:
>>
>>> I have to build a floor in a loft. At first this will provide a platform
>>> for some roof timber repairs to be carried out but ultimately it will be
>>> used for storage, not living space, so there are no formal BR
>>> requirements but, of course, I’d like it to be robust enough that it
>>> doesn’t all wind up in the bedroom below.
>>
>>> At present, there is a lath and plaster ceiling, with rather
>>> wimpy-looking ceiling joists, so not the most
>>> stable platform to work on. It’s in pretty good condition (as a ceiling,
>>> not a floor) so I’d like to keep it
>>
>> 2x8 joists is unusually large for loft storage. Almost all old joists are
>> quite able to handle storage. So I wonder if you've fully and correctly
>> evaluated the situation, ie what the current joist dimensions are (all 3).
>>
>>
>> NT
>
>Did the OP ever say exactly what his current joists were?

Not that I noticed:-)

The ones holding up our lath and plaster ceiling are around 3"x2".
Raised tie/included attic so about 12' span.
>

--
Tim Lamb

[GMM]

On Wednesday, October 17, 2012 9:57:32 PM UTC+1, Tim Watts wrote:
> meow...@care2.com wrote:
>
>
>
> > On Wednesday, October 17, 2012 11:49:06 AM UTC+1, GMM wrote:
>
> >
>
> >> I have to build a floor in a loft. At first this will provide a platform
>
> >> for some roof timber repairs to be carried out but ultimately it will be
>
> >> used for storage, not living space, so there are no formal BR
>
> >> requirements but, of course, I’d like it to be robust enough that it
>
> >> doesn’t all wind up in the bedroom below.
>
> >
>
> >> At present, there is a lath and plaster ceiling, with rather
>
> >> wimpy-looking ceiling joists, so not the most
>
> >> stable platform to work on. It’s in pretty good condition (as a ceiling,
>
> >> not a floor) so I’d like to keep it
>
> >
>
> > 2x8 joists is unusually large for loft storage. Almost all old joists are
>
> > quite able to handle storage. So I wonder if you've fully and correctly
>
> > evaluated the situation, ie what the current joist dimensions are (all 3).
>
> >
>
> >
>
> > NT
>
>
>
> Did the OP ever say exactly what his current joists were?
>
>
>

The simple answer is there aren't any - apart from the skinny (3x2 or so) ceiling joists. Since these are 150 years old, I wouldn't walk on them without care, let alone base any kind of structure on them.

[GB]

On 17/10/2012 22:50, GMM wrote:

>> Did the OP ever say exactly what his current joists were?
>>
>>
>>
>
> The simple answer is there aren't any - apart from the skinny (3x2 or so) ceiling joists. Since these are 150 years old, I wouldn't walk on them without care, let alone base any kind of structure on them.
>

You'll be walking on those when you do the work. Will you support them
temporarily?

[meow...@care2.com]

On Wednesday, October 17, 2012 9:07:08 PM UTC+1, GMM wrote:
> On Wednesday, October 17, 2012 8:56:57 PM UTC+1, (unknown) wrote:

> Just going by the standard joist span tables for moderately loaded floors. Of courser a lighter joist would be cheaper/easier/more desirable. On the other hand, I'd rather not have the whole thing collapse!

This is a very common misunderstanding. BR requirements are based on sound transmission requirements, producing sizes far in excess of those required for safety. If the table says 2x8 you could use 2x4, fill the loft to the roof, and not be at the remotest risk. A 10' 2x4 can deflect at least 6" safely, a full loft on 10' 2x4s provides a fraction of its safe load limit.

Re noise transmission, the gap between the 2 joist layers has more effect than 2x8s. And I don't expect anyone lives in your loft anyway. :)


NT

[John Rumm]

On 18/10/2012 13:11, meow...@care2.com wrote:
> On Wednesday, October 17, 2012 9:07:08 PM UTC+1, GMM wrote:
>> On Wednesday, October 17, 2012 8:56:57 PM UTC+1, (unknown) wrote:
>
>> Just going by the standard joist span tables for moderately loaded
>> floors. Of courser a lighter joist would be cheaper/easier/more
>> desirable. On the other hand, I'd rather not have the whole thing
>> collapse!
>
> This is a very common misunderstanding. BR requirements are based on
> sound transmission requirements, producing sizes far in excess of
> those required for safety. If the table says 2x8 you could use 2x4,

They are based on acceptable* levels of deflection rather than sound
transmission (while there are building regs on sound transmission, they
don't figure at all in structural calculations of joist sizes at all as
far as I am aware)

(* Where acceptable is usually specified as a function of the joist
length, or an absolute number of mm if over a certain length. The
calculations will also check that the beam is not likely to fail in
shear at the ends, or bending in the middle)

> fill the loft to the roof, and not be at the remotest risk. A 10' 2x4
> can deflect at least 6" safely, a full loft on 10' 2x4s provides a
> fraction of its safe load limit.

As you highlight, the purpose of the deflection limits is not because
that is where a joist will fail catastrophically, but that is where any
more movement would become unacceptable. L&P ceilings can move a fair
amount, but don't expect one that is supported by a floor structure that
deflects 5 inches every time someone walks over it to last long!

> Re noise transmission, the gap between the 2 joist layers has more
> effect than 2x8s. And I don't expect anyone lives in your loft
> anyway. :)

Loft floors (once strengthened) are actually very good for noise
insulation since they typically have two separate sets of joists
carrying the floor and ceiling loads.

[GMM]

Well, you're right: Nobody does live in my loft ...at least not that I'm dare of...though you never know
who might sneak in there (according to the Daily Mail at least....(!) I'm anticipating that this floor will
ultimately be loaded with all the junk you normally find in lofts - broken toys, old masters and all that.

On the other hand, I'm not sure it would feel too good to walk on an unsupported 14' span of 2 x 4
and if it deflects 6', then I would have to mount them higher than 2 x 8s so they didn't bang on the
top of the ceiling.

I have been working from the BR tables as a) they would seem to give the best outcome and b) I
couldn't find anything else that gave useful information.

I shall take another look and see if I can find any useful info on this. Perhaps 2 x 6s would do the job
reasonably, although there is an attraction in 2 x 8s in that a future (currently completely off the radar)
loft conversion would be possible without having to take it all down and start again.

Of course, there is an argument that smaller timbers in themselves reduce the total dead weight of the
floor a bit but I should think this is a pretty minor effect as is the impact on overall cost.
Probably the most compelling argument (for me) could be the fact that I have to lug them up 3 floors
to get them in, but that only needs to be done once.

[Doctor Drivel]

GMM wrote:
> Sorry chaps, a bit of a long post but I have tried to get all the
> essential information in (!) Undoubtedly
> there will be something I've missed out, even so....
>
> I have to build a floor in a loft. At first this will provide a
> platform for some roof timber repairs to be
> carried out but ultimately it will be used for storage, not living
> space, so there are no formal BR
> requirements but, of course, I'd like it to be robust enough that it
> doesn't all wind up in the bedroom below.
>
> At present, there is a lath and plaster ceiling, with rather
> wimpy-looking ceiling joists, so not the most
> stable platform to work on. It's in pretty good condition (as a
> ceiling, not a floor) so I'd like to keep it
> that way, which means installing joists clear of the ceiling timbers.
> This will, of course, create a void
> for a good depth of insulation.
>
> So question 1: To work above the ceiling, I'm thinking of supporting
> it with boards on acro jacks,
> possibly moving these according to where I'm working as I go and
> putting some boards on the ceiling
> joists to spread the load (mostly me!). Does this sound like a

> sensible thing to do - ie any better

Put in the void as much insulation as possible. Also you might want to renew
the lighing cables while you are it, and install more in cae of new light
positions in the romms below. And install some 2.5mm cable ready to fit
sockets up there in case. Even some pipe ready to replace a tank, etc. You
don't want to rip it all up a few years down the line.

[GMM]

>
> As you highlight, the purpose of the deflection limits is not because
>
> that is where a joist will fail catastrophically, but that is where any
>
> more movement would become unacceptable. L&P ceilings can move a fair
>
> amount, but don't expect one that is supported by a floor structure that
>
> deflects 5 inches every time someone walks over it to last long!
>
>

Not to mention the possibility of seasickness every time anyone bounces around up there !

>
>
>
> > Re noise transmission, the gap between the 2 joist layers has more
>
> > effect than 2x8s. And I don't expect anyone lives in your loft
>
> > anyway. :)
>
>
>
> Loft floors (once strengthened) are actually very good for noise
>
> insulation since they typically have two separate sets of joists
>
> carrying the floor and ceiling loads.
>

I'm hoping this one will be good for heat insulation too, with a full load of insulation between the ceiling and the new floor.

[GMM]

>
>
> Put in the void as much insulation as possible. Also you might want to renew
>
> the lighing cables while you are it, and install more in cae of new light
>
> positions in the romms below. And install some 2.5mm cable ready to fit
>
> sockets up there in case. Even some pipe ready to replace a tank, etc. You
>
> don't want to rip it all up a few years down the line.

I couldn't agree more: Flooring over old cable would be a significant error and my first job will be to
put in some decent (and redundant in case one blows) lighting to replace the one batten holder there
at present, along with a socket or two, spurred from a handy point close by.
The (for want of a better term) internal gable end wall is adjacent to the stairs and top floor rooms so
the plan will be to put a layer of celotex over this once I have a floor I can stand on to do it.

[Alan Braggins]

In article <507fdf42$0$1139$5b6a...@news.zen.co.uk>, GB wrote:
>>
>> The simple answer is there aren't any - apart from the skinny (3x2 or so) ceiling joists. Since these are 150 years old, I wouldn't walk on them without care, let alone base any kind of structure on them.
>
>You'll be walking on those when you do the work. Will you support them
>temporarily?

He said he was thinking of boards on Acro props. Knowing exactly which bits
are supported from below when working above might be a bit tricky, but at the
least they'll effectively have a much shorter span.

[GMM]

Quite straightforward really: The loft is exactly the same as the room beneath, so props can be put
directly under the points where work is taking place and moved as it progresses. Using boards above
and below will spread the load.
As you say, even off (but close to) a board this will effectively shorten the span of timber so reduce stresses.
Certainly, in my view, preferable to buggering up the ceiling.

[meow...@care2.com]

On Thursday, October 18, 2012 1:49:59 PM UTC+1, John Rumm wrote:

> On 18/10/2012 13:11, meow2222 wrote:
> > On Wednesday, October 17, 2012 9:07:08 PM UTC+1, GMM wrote:
> >> On Wednesday, October 17, 2012 8:56:57 PM UTC+1, (unknown) wrote:

> >> Just going by the standard joist span tables for moderately loaded
> >> floors. Of courser a lighter joist would be cheaper/easier/more
> >> desirable. On the other hand, I'd rather not have the whole thing
> >> collapse!

> > This is a very common misunderstanding. BR requirements are based on
> > sound transmission requirements, producing sizes far in excess of
> > those required for safety. If the table says 2x8 you could use 2x4,

> They are based on acceptable* levels of deflection rather than sound

Whether that's true depends on how you define acceptable. Limiting deflection to 3mm over 10' has absolutely nothing to do with safety or what I would call acceptable deflection limits.

> > fill the loft to the roof, and not be at the remotest risk. A 10' 2x4
> > can deflect at least 6" safely, a full loft on 10' 2x4s provides a
> > fraction of its safe load limit.

> As you highlight, the purpose of the deflection limits is not because
> that is where a joist will fail catastrophically, but that is where any
> more movement would become unacceptable.

Well, not in any sense of acceptable that I would buy into. Lots of old houses have ceiling deflection levels numerous times as big as BR allows, and I don't see any problems or complaints resulting.

> L&P ceilings can move a fair
> amount, but don't expect one that is supported by a floor structure that
> deflects 5 inches every time someone walks over it to last long!

Timber that light isn't being discussed, and the OP's new joists won't be supporting a ceiling.


If 3.1mm deflection would upset the op, go with the tables. I personally wouldnt have any problem with half an inch on a heavily loaded section


NT

[meow...@care2.com]

On Thursday, October 18, 2012 2:17:37 PM UTC+1, GMM wrote:
> On Thursday, October 18, 2012 1:11:45 PM UTC+1, (unknown) wrote:
> > On Wednesday, October 17, 2012 9:07:08 PM UTC+1, GMM wrote:
> > > On Wednesday, October 17, 2012 8:56:57 PM UTC+1, (unknown) wrote:

> ultimately be loaded with all the junk you normally find in lofts - broken toys, old masters and all that.

> On the other hand, I'm not sure it would feel too good to walk on an unsupported 14' span of 2 x 4
> and if it deflects 6', then I would have to mount them higher than 2 x 8s so they didn't bang on the
> top of the ceiling.

I regularly put my weight on a single 12' unconnected 2x4, and there's no visible deflection. it helps if you bring reality into this.

In a loft floor structure the joists are connected by the boarding, and sometimes noggings, both of which spread loads over multiple joists

> I have been working from the BR tables as a) they would seem to give the best outcome and b) I
> couldn't find anything else that gave useful information.

> I shall take another look and see if I can find any useful info on this. Perhaps 2 x 6s would do the job
> reasonably, although there is an attraction in 2 x 8s in that a future (currently completely off the radar)
> loft conversion would be possible without having to take it all down and start again.

Timber requirements get ever deeper as the years roll by.

> Probably the most compelling argument (for me) could be the fact that I have to lug them up 3 floors
> to get them in, but that only needs to be done once.

Lots of victorian houses have 14' 2x4, 2x3 and even 1.5x3 loft joists, which all support a fully loaded loft without problem.


NT

[John Rumm]

On 18/10/2012 23:56, meow...@care2.com wrote:
> On Thursday, October 18, 2012 1:49:59 PM UTC+1, John Rumm wrote:
>> On 18/10/2012 13:11, meow2222 wrote:
>>> On Wednesday, October 17, 2012 9:07:08 PM UTC+1, GMM wrote:
>>>> On Wednesday, October 17, 2012 8:56:57 PM UTC+1, (unknown)
>>>> wrote:
>
>>>> Just going by the standard joist span tables for moderately
>>>> loaded floors. Of courser a lighter joist would be
>>>> cheaper/easier/more desirable. On the other hand, I'd rather
>>>> not have the whole thing collapse!
>
>>> This is a very common misunderstanding. BR requirements are based
>>> on sound transmission requirements, producing sizes far in excess
>>> of those required for safety. If the table says 2x8 you could use
>>> 2x4,
>
>> They are based on acceptable* levels of deflection rather than
>> sound
>
> Whether that's true depends on how you define acceptable. Limiting
> deflection to 3mm over 10' has absolutely nothing to do with safety
> or what I would call acceptable deflection limits.

(the limit is somewhat higher than 3mm in 10')

However, its not just about safety, but also producing an adequate
quality of building, where the floors don't bounce up and down, and all
the joints crack in the plaster because there is too much movement.

There are other requirements that are also taken into account with the
sizes commonly used, such as minimizing twisting of joists (although
strapping will still be required over 2.5m), and providing lateral
restraint of walls where required.

>>> fill the loft to the roof, and not be at the remotest risk. A 10'
>>> 2x4 can deflect at least 6" safely, a full loft on 10' 2x4s
>>> provides a fraction of its safe load limit.
>
>> As you highlight, the purpose of the deflection limits is not
>> because that is where a joist will fail catastrophically, but that
>> is where any more movement would become unacceptable.
>
> Well, not in any sense of acceptable that I would buy into. Lots of
> old houses have ceiling deflection levels numerous times as big as BR
> allows, and I don't see any problems or complaints resulting.

I have yet to be in a house where the floor deflects 6" while walking
across a room. Personally I would take that as an indication one should
get out fast!

However it is true that what was acceptable in the past would no longer
meet current standards. That is partly a reflection on more stringent
requirements for air tightness of properties, and also changes in
materials. Lime mortar, and soft bricks will accept more movement than
modern materials for example.

>> L&P ceilings can move a fair amount, but don't expect one that is
>> supported by a floor structure that deflects 5 inches every time
>> someone walks over it to last long!
>
> Timber that light isn't being discussed, and the OP's new joists
> won't be supporting a ceiling.

I was responding to your comments about the 2x4 which can deflect 6"
safely rather than the OPs storage floor. A 6" deflection may not cause
it to fail, but it would be excessive by any measure, and would not be
suitable even for loft storage.

The OP can use lighter timbers than those that would be required to meet
the standard of a floor if he wants, although if there was a suggestion
that at some point in the future it was upgraded to a habitable room it
would be sensible to build it to the required standards now obviously.

> If 3.1mm deflection would upset the op, go with the tables. I
> personally wouldnt have any problem with half an inch on a heavily
> loaded section

3.1mm would be fine even for a floor (the limit on that length would be
around 8.4mm IIRC).

[John Rumm]

4x2 joisted lofts usually feel reasonably solid underfoot. However it is
surprising just how much they creep over time, even without too much
weight on them.

If you look at:

http://internode.co.uk/loft/images/sag.jpg

That is looking at a new floor joist over a ceiling supported by 4.2m
4x2" joists.

one of the "D" joists as per:

http://internode.co.uk/temp/beam-layout.gif

The spacing under the ends of the new beam was approx 3/4 to 1" at the
wall plates. Mid span the gap under the joist is getting on for 4" - so
the ceiling had sagged around 3" mid span since when it was originally
built in 1956 (photo taken some time 2004). That was with just light
storage use on chipboard loft panels.

[meow...@care2.com]

On Friday, October 19, 2012 1:26:36 AM UTC+1, John Rumm wrote:

> On 18/10/2012 23:56, meow2222 wrote:
> > On Thursday, October 18, 2012 1:49:59 PM UTC+1, John Rumm wrote:
> >> On 18/10/2012 13:11, meow2222 wrote:
> >>> On Wednesday, October 17, 2012 9:07:08 PM UTC+1, GMM wrote:
> >>>> On Wednesday, October 17, 2012 8:56:57 PM UTC+1, (unknown)
> >>>> wrote:

> >>>> Just going by the standard joist span tables for moderately
> >>>> loaded floors. Of courser a lighter joist would be
> >>>> cheaper/easier/more desirable. On the other hand, I'd rather
> >>>> not have the whole thing collapse!

> >>> This is a very common misunderstanding. BR requirements are based
> >>> on sound transmission requirements, producing sizes far in excess
> >>> of those required for safety. If the table says 2x8 you could use
> >>> 2x4,

> >> They are based on acceptable* levels of deflection rather than
> >> sound

> > Whether that's true depends on how you define acceptable. Limiting
> > deflection to 3mm over 10' has absolutely nothing to do with safety
> > or what I would call acceptable deflection limits.

> (the limit is somewhat higher than 3mm in 10')

> However, its not just about safety, but also producing an adequate
> quality of building, where the floors don't bounce up and down, and all
> the joints crack in the plaster because there is too much movement.

No, its not. Millions of old houses have timber a fraction the size of the current requirements and don't suffer any of those problems.

> There are other requirements that are also taken into account with the
> sizes commonly used, such as minimizing twisting of joists (although
> strapping will still be required over 2.5m), and providing lateral
> restraint of walls where required.

> >>> fill the loft to the roof, and not be at the remotest risk. A 10'
> >>> 2x4 can deflect at least 6" safely, a full loft on 10' 2x4s
> >>> provides a fraction of its safe load limit.

> >> As you highlight, the purpose of the deflection limits is not
> >> because that is where a joist will fail catastrophically, but that
> >> is where any more movement would become unacceptable.

> > Well, not in any sense of acceptable that I would buy into. Lots of
> > old houses have ceiling deflection levels numerous times as big as BR
> > allows, and I don't see any problems or complaints resulting.

> I have yet to be in a house where the floor deflects 6" while walking
> across a room. Personally I would take that as an indication one should
> get out fast!

obviously that's not relevant

> However it is true that what was acceptable in the past would no longer
> meet current standards. That is partly a reflection on more stringent
> requirements for air tightness of properties,

I've never known a Victorian ceiling/loft floor be draughty

> and also changes in
> materials. Lime mortar, and soft bricks will accept more movement than
> modern materials for example.

Slight movement of loft floor does not move the brickwork. Modern PB is much more tolerant of movement than lath & plaster.

> >> L&P ceilings can move a fair amount, but don't expect one that is
> >> supported by a floor structure that deflects 5 inches every time
> >> someone walks over it to last long!

> > Timber that light isn't being discussed, and the OP's new joists
> > won't be supporting a ceiling.

> I was responding to your comments about the 2x4 which can deflect 6"
> safely rather than the OPs storage floor. A 6" deflection may not cause
> it to fail, but it would be excessive by any measure, and would not be
> suitable even for loft storage.

Obviously a floor structure that deflects 6" in use has never been proposed. My point was that deflection levels encountered in real life floors are a fraction of failure limits

> The OP can use lighter timbers than those that would be required to meet
> the standard of a floor if he wants, although if there was a suggestion
> that at some point in the future it was upgraded to a habitable room it
> would be sensible to build it to the required standards now obviously.

Standards change, so its not obvious, its just an option

> > If 3.1mm deflection would upset the op, go with the tables. I
> > personally wouldnt have any problem with half an inch on a heavily
> > loaded section

> 3.1mm would be fine even for a floor (the limit on that length would be
> around 8.4mm IIRC).

NT

[meow...@care2.com]

That's heavily at variance with my own experiences. I don't know why.


NT

[John Rumm]

True, so long as the fraction you have in mind is say 7/8ths...

Many places in the past would have used say 7x2 where these days 8x2
would be deemed adequate.

There are places which use significantly shallower timbers, but then
tend to have more frequent cross members and hence shorter effective spans.

Many of our downstairs floors here are only on 4x2" joists - but these
span dwarf walls often at 4' spacings - so they are quite rigid.

>> However it is true that what was acceptable in the past would no
>> longer meet current standards. That is partly a reflection on more
>> stringent requirements for air tightness of properties,
>
> I've never known a Victorian ceiling/loft floor be draughty

Modern buildings are in general very much less draughty than Victorian
places. There have been many changes in building practices to achieve
this. One example being joist support. It is no longer common to set
joist ends into openings in walls since (among other things) it also
creates a potential air gap. The switch to joist hangers also frees you
from the limitation of needing to use joist sizes that match a brick
height multiple in height.

>> and also changes in materials. Lime mortar, and soft bricks will
>> accept more movement than modern materials for example.
>
> Slight movement of loft floor does not move the brickwork.

I did not say it did - it was an example of building material that is
more tolerant of movement. Perhaps horsehair reinforced plaster would
have been a better example.

> Modern PB
> is much more tolerant of movement than lath & plaster.

Its different - not necessarily always better.

>>>> L&P ceilings can move a fair amount, but don't expect one that
>>>> is supported by a floor structure that deflects 5 inches every
>>>> time someone walks over it to last long!
>
>>> Timber that light isn't being discussed, and the OP's new joists
>>> won't be supporting a ceiling.
>
>> I was responding to your comments about the 2x4 which can deflect
>> 6" safely rather than the OPs storage floor. A 6" deflection may
>> not cause it to fail, but it would be excessive by any measure, and
>> would not be suitable even for loft storage.
>
> Obviously a floor structure that deflects 6" in use has never been
> proposed. My point was that deflection levels encountered in real
> life floors are a fraction of failure limits

No one (including the building regs) is attempting to suggest otherwise.

>> The OP can use lighter timbers than those that would be required to
>> meet the standard of a floor if he wants, although if there was a
>> suggestion that at some point in the future it was upgraded to a
>> habitable room it would be sensible to build it to the required
>> standards now obviously.
>
> Standards change, so its not obvious, its just an option

Indeed, but they rarely reduce, so its fairly obvious (to me at least).

Also building regs are not retrospective. So if a floor was designed as
a floor, and was compliant with the standards of the time, you would be
able to use as the basis of your room in the roof, it even if the
standards applying had changed since it was built.

>>> If 3.1mm deflection would upset the op, go with the tables. I
>>> personally wouldnt have any problem with half an inch on a
>>> heavily loaded section
>
>> 3.1mm would be fine even for a floor (the limit on that length
>> would be around 8.4mm IIRC).
>
>
> NT
>

[John Rumm]

Its possibly a reflection of the fact its actually quite difficult to
observe a sag of a few inches over a 4m+ span, unless you stick a
straight edge against it like I had in effect done there, or have some
way of sighting along the timber.

The sag was not visible in the room below, although I expect that had
you have gone round with a measuring stick you would have seen it.

[meow...@care2.com]

On Friday, October 19, 2012 3:53:36 PM UTC+1, John Rumm wrote:
> On 19/10/2012 13:38, meow...@care2.com wrote:
> > On Friday, October 19, 2012 1:26:36 AM UTC+1, John Rumm wrote:
> >> On 18/10/2012 23:56, meow2222 wrote:
> >>> On Thursday, October 18, 2012 1:49:59 PM UTC+1, John Rumm wrote:
> >>>> On 18/10/2012 13:11, meow2222 wrote:
> >>>>> On Wednesday, October 17, 2012 9:07:08 PM UTC+1, GMM wrote:
> >>>>>> On Wednesday, October 17, 2012 8:56:57 PM UTC+1, (unknown)
> >>>>>> wrote:


> >> However, its not just about safety, but also producing an adequate
> >> quality of building, where the floors don't bounce up and down, and
> >> all the joints crack in the plaster because there is too much
> >> movement.

> > No, its not. Millions of old houses have timber a fraction the size
> > of the current requirements and don't suffer any of those problems.

> True, so long as the fraction you have in mind is say 7/8ths...
> Many places in the past would have used say 7x2 where these days 8x2
> would be deemed adequate.
> There are places which use significantly shallower timbers, but then
> tend to have more frequent cross members and hence shorter effective spans.

I'm talking about 3, 4 & 5" timber where new builds would use double to triple the size

> >> However it is true that what was acceptable in the past would no
> >> longer meet current standards. That is partly a reflection on more
> >> stringent requirements for air tightness of properties,

> > I've never known a Victorian ceiling/loft floor be draughty

> Modern buildings are in general very much less draughty than Victorian
> places. There have been many changes in building practices to achieve
> this. One example being joist support. It is no longer common to set
> joist ends into openings in walls since (among other things) it also
> creates a potential air gap. The switch to joist hangers also frees you
> from the limitation of needing to use joist sizes that match a brick
> height multiple in height.

But one place draughts don't come in is through Victorian ceilings

> >> The OP can use lighter timbers than those that would be required to
> >> meet the standard of a floor if he wants, although if there was a
> >> suggestion that at some point in the future it was upgraded to a
> >> habitable room it would be sensible to build it to the required
> >> standards now obviously.

> > Standards change, so its not obvious, its just an option

> Indeed, but they rarely reduce, so its fairly obvious (to me at least).
> Also building regs are not retrospective. So if a floor was designed as
> a floor, and was compliant with the standards of the time, you would be
> able to use as the basis of your room in the roof, it even if the
> standards applying had changed since it was built.

There's no way a BCO will accept a loft conversion in a 1924 house on its original 3" loft floor joists.


NT

[John Rumm]

On 20/10/2012 01:49, meow...@care2.com wrote:
> On Friday, October 19, 2012 3:53:36 PM UTC+1, John Rumm wrote:
>> On 19/10/2012 13:38, meow...@care2.com wrote:

>>>> However it is true that what was acceptable in the past would no
>>>> longer meet current standards. That is partly a reflection on more
>>>> stringent requirements for air tightness of properties,
>
>>> I've never known a Victorian ceiling/loft floor be draughty

I just noticed that you superficially said *loft* floors there, and my
comments that followed were more generally about "floors" in general -
my fault for not reading carefully.

>> Modern buildings are in general very much less draughty than Victorian
>> places. There have been many changes in building practices to achieve
>> this. One example being joist support. It is no longer common to set
>> joist ends into openings in walls since (among other things) it also
>> creates a potential air gap. The switch to joist hangers also frees you
>> from the limitation of needing to use joist sizes that match a brick
>> height multiple in height.
>
> But one place draughts don't come in is through Victorian ceilings

There is a general drive to cut down air currents through void spaces in
direct contact with heated surfaces. If you have a penetration through a
wall, then air will flow through it. Even if that never discharges
directly into the living space, it lowers the thermal performance of the
building.

>>>> The OP can use lighter timbers than those that would be required to
>>>> meet the standard of a floor if he wants, although if there was a
>>>> suggestion that at some point in the future it was upgraded to a
>>>> habitable room it would be sensible to build it to the required
>>>> standards now obviously.
>
>>> Standards change, so its not obvious, its just an option
>
>> Indeed, but they rarely reduce, so its fairly obvious (to me at least).
>> Also building regs are not retrospective. So if a floor was designed as
>> a floor, and was compliant with the standards of the time, you would be
>> able to use as the basis of your room in the roof, it even if the
>> standards applying had changed since it was built.
>
> There's no way a BCO will accept a loft conversion in a 1924 house on its original 3" loft floor joists.

I doubt a loft with 3" joists would not have been deemed acceptable as a
proper floor for a habitable space - even in 1924.

However, my point was, that if you upgrade something now to the current
standards of a floor in a habitable room, then there would be no need to
upgrade it further if one later made the space habitable - even if the
standards for a floor have changed by then.

[meow...@care2.com]

On Saturday, October 20, 2012 4:42:00 PM UTC+1, John Rumm wrote:
> On 20/10/2012 01:49, meow...@care2.com wrote:
> > On Friday, October 19, 2012 3:53:36 PM UTC+1, John Rumm wrote:
> >> On 19/10/2012 13:38, meow...@care2.com wrote:


> >> Also building regs are not retrospective. So if a floor was designed as
> >> a floor, and was compliant with the standards of the time, you would be
> >> able to use as the basis of your room in the roof, it even if the
> >> standards applying had changed since it was built.

> > There's no way a BCO will accept a loft conversion in a 1924 house on its original 3" loft floor joists.

> I doubt a loft with 3" joists would not have been deemed acceptable as a
> proper floor for a habitable space - even in 1924.
> However, my point was, that if you upgrade something now to the current
> standards of a floor in a habitable room, then there would be no need to
> upgrade it further if one later made the space habitable - even if the
> standards for a floor have changed by then.

3x3 was the smallest standard habitable flooring joist size in Victorian houses. It was much used for short spans, such as across corridors & landings.

IIRC the 1924 BR didn't specify joist sizes, so 3x3 would still be compliant for habitation then. It could be used in loft floors above corridors, where the span was short.

I challenge you to find any BCO that would accept that in a loft conversion today.


NT

[John Rumm]

On 20/10/2012 20:33, meow...@care2.com wrote:
> On Saturday, October 20, 2012 4:42:00 PM UTC+1, John Rumm wrote:
>> On 20/10/2012 01:49, meow...@care2.com wrote:
>>> On Friday, October 19, 2012 3:53:36 PM UTC+1, John Rumm wrote:
>>>> On 19/10/2012 13:38, meow...@care2.com wrote:
>
>
>>>> Also building regs are not retrospective. So if a floor was
>>>> designed as a floor, and was compliant with the standards of
>>>> the time, you would be able to use as the basis of your room in
>>>> the roof, it even if the standards applying had changed since
>>>> it was built.
>
>>> There's no way a BCO will accept a loft conversion in a 1924
>>> house on its original 3" loft floor joists.
>
>> I doubt a loft with 3" joists would not have been deemed acceptable
>> as a proper floor for a habitable space - even in 1924. However,
>> my point was, that if you upgrade something now to the current
>> standards of a floor in a habitable room, then there would be no
>> need to upgrade it further if one later made the space habitable -
>> even if the standards for a floor have changed by then.
>
> 3x3 was the smallest standard habitable flooring joist size in
> Victorian houses. It was much used for short spans, such as across
> corridors & landings.

And it still might be acceptable now (for short lengths)

> IIRC the 1924 BR didn't specify joist sizes, so 3x3 would still be
> compliant for habitation then. It could be used in loft floors above
> corridors, where the span was short.

A loft floor is not a floor in the accepted sense though - its not
expected to carry significant load.

> I challenge you to find any BCO that would accept that in a loft
> conversion today.

A BCO would be happy with a loft using 3x2 - its a good deal better than
many a lofts built with modern trusses. However that is a very different
thing from a loft floor which going to be used for a habitable room. If
you are converting the loft, then the same spec as would apply to any
other floor in the building will kick in.

Out of interest I had an experiment with superbeam to see what you can
get away with on a 3x2 (well 72x47mm) and a typical floor load
(uniformly distributed 0.8kN/m on each joist). 1.3m seems to be about
the limit - so you could probably still do a landing with it and comply
with modern building regs. (having said that, its generally simpler to
use one depth all over to save having to buy lots of timber sizes)

[meow...@care2.com]

For clarity, lets take it a step further. Say the loft got 2x2s in 4' spans in 1924, hopelessly unsuitable for habitable rooms, but still compliant for them in 1924. So it was built in compliance with BR standards for habitation at the time, and you can indeed walk on them, just about. But no BCO in their right mind would accept a conversion to habitable now on 2x2s.


NT

[Tony Bryer]

On Sat, 20 Oct 2012 22:33:14 +0100 John Rumm wrote :
> Out of interest I had an experiment with superbeam to see what you can
> get away with on a 3x2 (well 72x47mm) and a typical floor load
> (uniformly distributed 0.8kN/m on each joist). 1.3m seems to be about
> the limit - so you could probably still do a landing with it and comply
> with modern building regs. (having said that, its generally simpler to
> use one depth all over to save having to buy lots of timber sizes)

Our old rule of thumb when I was a BCO which matched the tables pretty
well was that for floor joists double the depth in inches and subtract two
to get the permissible span in feet; flat roof joists, subtract one (2"
joists).

As you say, in most cases practicality requires all joists to be the same
depth (you need tops of joists to be level and want them all to bear on
wall at one level) so except for the largest span they are generally
oversized. There's also more in reserve in that for virtually all joists,
deflection governs the size, not bending stress and few floors are loaded
to BR design loads (1.5kN/m2 30lb/ft2).

--
Tony Bryer, Greentram: 'Software to build on',
Melbourne, Australia www.greentram.com

[Tim Lamb]

In message <VA.00004c5...@delme.greentram.com>, Tony Bryer
<to...@delme.greentram.com> writes

Slightly sideways... is it possible that the quality of timber available
to the Victorians was better than that of today?

I use a lot of recycled 4"x2" and generally find that the grain is more
dense and the knots smaller than that supplied new.
>

--
Tim Lamb

[stuart noble]

A dense grain means a timber gown in more northerly climes, and I don't
think the climate has changed significantly since Victorian times. There
has always been, and still is, a great variation in quality and price
for what is loosely termed "4x2"

[ne...@sylva.icuklive.co.uk]

On Sun, 21 Oct 2012 08:13:13 +0100, Tim Lamb
<t...@marfordfarm.demon.co.uk> wrote:

>Slightly sideways... is it possible that the quality of timber available
>to the Victorians was better than that of today?
>
>I use a lot of recycled 4"x2" and generally find that the grain is more
>dense and the knots smaller than that supplied new.

Most timber Victorians used in buildings would have been imported and
Baltic or Scandinavian, slow grown 14 rings to the inch and the snow
load knocks off dead suppressed branches.

We were major importers of wood on the global market as our economy
grew earlier than others.

Since the 1950s more home grown timber has come on line and it has
benefited from machine stress grading where in the past it failed
visual grading. Also I suspect much Canadian lumber is second growth
nowadays.


AJH

[stuart noble]

On 21/10/2012 11:13, ne...@sylva.icuklive.co.uk wrote:
> On Sun, 21 Oct 2012 08:13:13 +0100, Tim Lamb
> <t...@marfordfarm.demon.co.uk> wrote:
>
>> Slightly sideways... is it possible that the quality of timber available
>> to the Victorians was better than that of today?
>>
>> I use a lot of recycled 4"x2" and generally find that the grain is more
>> dense and the knots smaller than that supplied new.
>
> Most timber Victorians used in buildings would have been imported and
> Baltic or Scandinavian, slow grown 14 rings to the inch and the snow
> load knocks off dead suppressed branches.

Interesting about the snow.

>
> We were major importers of wood on the global market as our economy
> grew earlier than others.
>
> Since the 1950s more home grown timber has come on line and it has
> benefited from machine stress grading where in the past it failed
> visual grading. Also I suspect much Canadian lumber is second growth
> nowadays.
>
>
> AJH
>

Finnish softwood is as good as it's ever been but too pricey for the
general market

[GMM]

So for a 14ft span (as here), the 8" joists I originally proposed would be about right?
I wonder if the tables are constructed from that rule of thumb or from a complex calculation that gives the same result?
I guess most of the discussion (now) is about what you might be able to get away with, rather than what should be done, but I'd prefer to over-engineer than under, for the sake of a couple inches.
Of course, I'm equally concerned that they are mounted securely, as Mr R outlined......

[John Rumm]

I don't think 2x2 would have been used for the floor of a habitable room
in 1924 or at any other time.

Its seems to are engaging in a little reductio ad absurdum.

> So it was built in compliance with BR
> standards for habitation at the time, and you can indeed walk on
> them, just about. But no BCO in their right mind would accept a
> conversion to habitable now on 2x2s.

[John Rumm]

On 21/10/2012 08:13, Tim Lamb wrote:

Almost certainly... it may not have been "stress graded" but then it was
not the fast grown, low density, twisted to buggery stuff that often
masquerades as timber these days.

> I use a lot of recycled 4"x2" and generally find that the grain is more
> dense and the knots smaller than that supplied new.


--

[John Rumm]

On 21/10/2012 13:09, GMM wrote:
> On Sunday, October 21, 2012 2:16:39 AM UTC+1, Tony Bryer wrote:
>> On Sat, 20 Oct 2012 22:33:14 +0100 John Rumm wrote :
>>
>>> Out of interest I had an experiment with superbeam to see what
>>> you can get away with on a 3x2 (well 72x47mm) and a typical floor
>>> load (uniformly distributed 0.8kN/m on each joist). 1.3m seems to
>>> be about the limit - so you could probably still do a landing
>>> with it and comply with modern building regs. (having said that,
>>> its generally simpler to use one depth all over to save having to
>>> buy lots of timber sizes)
>>
>> Our old rule of thumb when I was a BCO which matched the tables
>> pretty well was that for floor joists double the depth in inches
>> and subtract two to get the permissible span in feet; flat roof
>> joists, subtract one (2" joists).
>>
>> As you say, in most cases practicality requires all joists to be
>> the same depth (you need tops of joists to be level and want them
>> all to bear on wall at one level) so except for the largest span
>> they are generally oversized. There's also more in reserve in that
>> for virtually all joists, deflection governs the size, not bending
>> stress and few floors are loaded to BR design loads (1.5kN/m2
>> 30lb/ft2).

> So for a 14ft span (as here), the 8" joists I originally proposed
> would be about right? I wonder if the tables are constructed from
> that rule of thumb or from a complex calculation that gives the same
> result? I guess most of the discussion (now) is about what you might
> be able to get away with, rather than what should be done, but I'd
> prefer to over-engineer than under, for the sake of a couple inches.
> Of course, I'm equally concerned that they are mounted securely, as
> Mr R outlined......

If you can rule out that the space will ever be converted to habitable,
then you can undersize a tad from the tabulated values. As Tony
mentioned above, its normally the deflection limits that dictate the
size rather than the shear or bending limits. (i.e the floor would be
likely to damage decorative finishes, feel to bouncy, and upset
inhabitants of rooms below, long before the timber is in danger of
actually failing)

For your application (i.e. with the new beams some distance above the
existing ceiling, and not ceiling to be mounted on the underside of the
new joists), deflection beyond normal limits is a non issue. So it
reduces to a problem of what is adequate in terms of bending and shear
loading on the timber (assuming you don't mind it feeling a little more
bouncy than "normal" given that you know it is still structurally sound).

Perhaps a play Tony's excellent bit of software might be in order
(assuming there is still a demo version available for download?)

[GMM]

Yes John - it may potentially be that 2 x 6 joists (which are significantly cheaper per metre) could do an
adequate job in this application, if they are available (my local timber yard couldn't supply that length
for the living room when I was costing it: There, the lower spec was due to the wall running along the
middle, effectively halving the span).

TBH, I didn't spot Tony's software - I thought he just mentioned the rule of thumb(!). If I did go below
spec (say 4m of 2 x 6), it would be great to have some idea of how wobbly such a floor would be. The
joist tables just give maximum length for size, as far as I can see.

I'd still feel a little uncomfortable that it would deny the option of making the space habitable in the
future though, even though the rest of the house is big enough that it shouldn't be an issue.
Although I get the point that BR specs change over time, they surely can't ask for joist that are much
deeper than they require now, so I would have thought the current specs won't change much.

(Apologies if my posts are hard to read. Someone told me a while ago they weren't wrapping, whilst in
a recent thread someone else told me they had a lot of empty lines!)

[John Rumm]

On 21/10/2012 17:47, GMM wrote:
> On Sunday, October 21, 2012 4:25:44 PM UTC+1, John Rumm wrote:
>> On 21/10/2012 13:09, GMM wrote:

>> If you can rule out that the space will ever be converted to habitable,
>> then you can undersize a tad from the tabulated values. As Tony
>>
>> mentioned above, its normally the deflection limits that dictate the
>> size rather than the shear or bending limits. (i.e the floor would be
>> likely to damage decorative finishes, feel to bouncy, and upset
>> inhabitants of rooms below, long before the timber is in danger of
>> actually failing)
>>
>> For your application (i.e. with the new beams some distance above the
>> existing ceiling, and not ceiling to be mounted on the underside of the
>> new joists), deflection beyond normal limits is a non issue. So it
>> reduces to a problem of what is adequate in terms of bending and shear
>> loading on the timber (assuming you don't mind it feeling a little more
>> bouncy than "normal" given that you know it is still structurally sound).
>>
>>
>>
>> Perhaps a play Tony's excellent bit of software might be in order
>> (assuming there is still a demo version available for download?)

> Yes John - it may potentially be that 2 x 6 joists (which are significantly cheaper per metre) could do an
> adequate job in this application, if they are available (my local timber yard couldn't supply that length
> for the living room when I was costing it: There, the lower spec was due to the wall running along the
> middle, effectively halving the span).

A timber merchant ought to be able to get 6x2 in 5.3m lengths at least.

> TBH, I didn't spot Tony's software - I thought he just mentioned the rule of thumb(!). If I did go below
> spec (say 4m of 2 x 6), it would be great to have some idea of how wobbly such a floor would be. The
> joist tables just give maximum length for size, as far as I can see.

I just checked, there is still a demo available. It has printing
knobbled - but that won't be a handicap for your needs.

http://www.superbeam.co.uk/sbwdemo.htm

(the usual caveats about it letting you design unsafe structures faster
apply, if you don't stick in sensible values!)

It will show you the calculated deflection for whatever load you apply,
and also tell you when you are exceeding the safe working limits on the
timber.

If you model your longest timber that should let you get a feel for the
changes. The loadings to apply for a normal floor appear further up the
thread (if you know what you are storing etc you may be able to use
lower figures)

> I'd still feel a little uncomfortable that it would deny the option of making the space habitable in the
> future though, even though the rest of the house is big enough that it shouldn't be an issue.

Might be worth working out what you could "get away with" and then
comparing the cost difference to doing it to full spec.

> Although I get the point that BR specs change over time, they surely can't ask for joist that are much
> deeper than they require now, so I would have thought the current specs won't change much.

I would not expect them to change in substance at all really. They may
grow to include more on composite joists (i.e. man made beams with
struts top and bottom and some sheet material webbing)

> (Apologies if my posts are hard to read. Someone told me a while ago they weren't wrapping, whilst in
> a recent thread someone else told me they had a lot of empty lines!)

Its a combination of not wrapping and all the lines being double spaced.
The former is easy to fix in a reply with a quick CTRL + R in
thunderbird. That latter takes slightly more editing!

[meow...@care2.com]

That's exactly the point. All Victorian loft floor joists were compliant for habitable use at time of building, but no BCO is going to accept them in a conversion today.


NT

[GMM]

Thanks for the link - Now I have to find a Windows PC I can spark up and run the software - should be interesting.

I suspect a narrower joist could be significantly cheaper for these timbers, although it might not make a substantial impact on the job cost overall.
I've rather taken a 'do it properly and do it once' approach to this house, rather than being cost-driven.

I did toy at one stage with the prospects for making up composite joists in situ, given the access issues, but decided that would just add another variable (and potential disaster) to the equation (!)

Must have another go at Thunderbird for this group. I had it working, then it wouldn't post for some reason.

[John Rumm]

I don't quite follow the line of thought. A loft floor in 1924 was *not*
designed for habitable use then, so its no surprise it would not be
considered adequate now.

Your point seems to be that the definition of "habitable" did not exist
in 1924. However to cut through the confusion, look at what would have
been installed in 1924 for a normal 1st floor set of joists, since that
will have been designed for what we would today call habitable. If the
loft had similar spans but thinner joists, (which which is a safe bet)
then its not habitable - then or now.

[John Rumm]

On 22/10/2012 10:19, GMM wrote:

> Thanks for the link - Now I have to find a Windows PC I can spark up
> and run the software - should be interesting.

Probably not as "industry standard" or trustworthy, but there are a
number of smart phone apps about that will do calcs at least on a single
beam. Might be worth trying one of those if you have platform for that.

> I suspect a narrower joist could be significantly cheaper for these
> timbers, although it might not make a substantial impact on the job
> cost overall. I've rather taken a 'do it properly and do it once'
> approach to this house, rather than being cost-driven.

It depends a bit on the layout and how many joist hangers you need etc.
Those will be the same price regardless.

> I did toy at one stage with the prospects for making up composite
> joists in situ, given the access issues, but decided that would just
> add another variable (and potential disaster) to the equation (!)
>
> Must have another go at Thunderbird for this group. I had it
> working, then it wouldn't post for some reason.

It ought to be fairly painless for newsgroups if you are using your ISPs
newsserver. If using a third party one, you may need to turn on the
"request authentication" option in the account settings.

[meow...@care2.com]

I don't think you've followed what I'm saying at all.


NT

[GMM]

On 22/10/2012 14:43, John Rumm wrote:
> On 22/10/2012 10:19, GMM wrote:
>
>> Thanks for the link - Now I have to find a Windows PC I can spark up
>> and run the software - should be interesting.
>
> Probably not as "industry standard" or trustworthy, but there are a
> number of smart phone apps about that will do calcs at least on a single
> beam. Might be worth trying one of those if you have platform for that.
>
>> I suspect a narrower joist could be significantly cheaper for these
>> timbers, although it might not make a substantial impact on the job
>> cost overall. I've rather taken a 'do it properly and do it once'
>> approach to this house, rather than being cost-driven.
>
> It depends a bit on the layout and how many joist hangers you need etc.
> Those will be the same price regardless.
>
>> I did toy at one stage with the prospects for making up composite
>> joists in situ, given the access issues, but decided that would just
>> add another variable (and potential disaster) to the equation (!)
>>
>> Must have another go at Thunderbird for this group. I had it
>> working, then it wouldn't post for some reason.
>
> It ought to be fairly painless for newsgroups if you are using your ISPs
> newsserver. If using a third party one, you may need to turn on the
> "request authentication" option in the account settings.
>
>

I should take a browse through the phone apps (without wishing to turn
the thread into another debate about which phone is best!)
Your comments about ISP newservers made me realise that I set
Thunderbird up for home, then might have failed to post over a different
WiFi connection, so if this posts, problem solved (ish)

[John Rumm]

So what is your point then? I thought we had covered:

1) Standards have edged up?
Agreed. But not hugely.

2) Skimpy joists in a loft were acceptable as a habitable room floor in
the past?
Nope, don't buy it for a moment. They were acceptable as a loft floor
and that is all - but there was no anticipation that was going to be
used as a bedroom of office etc. They were also to a much lower standard
than was deemed necessary for all the other "proper" floors in the same
building.

3) Joists will survive loadings far beyond those specified in the
building regs?
Yup agreed. The specs are primarily intended to limit deflection, and
excessive deflection occurs long before structural failure.

4) You can still use skimpy timbers if they are short enough?
Yup agreed, nothing much has changed there.

5) Joist sizes are specced on sound transmission?
I don't believe that really comes into it. Specs on sound transmission
have got much stronger in recent times, but joist sizes themselves have
not increased as a result. Joist support techniques have changed -
reducing wall penetrations, ensuring adequate insulating material is
included in floor construction to reduce noise. Eliminating gaps and air
paths are all partly related to reducing noise transmission as well as
improving thermal performance. Fire protection rules have also
tightened, and that has had a knock on on ceiling coverings (i.e. 1/2"
PB and skim, not 9mm etc, intumescent covers over ceiling penetrations).

Have I missed any?

[John Rumm]

On 22/10/2012 21:48, GMM wrote:
> On 22/10/2012 14:43, John Rumm wrote:
>> On 22/10/2012 10:19, GMM wrote:
>>
>>> Thanks for the link - Now I have to find a Windows PC I can spark up
>>> and run the software - should be interesting.
>>
>> Probably not as "industry standard" or trustworthy, but there are a
>> number of smart phone apps about that will do calcs at least on a single
>> beam. Might be worth trying one of those if you have platform for that.
>>
>>> I suspect a narrower joist could be significantly cheaper for these
>>> timbers, although it might not make a substantial impact on the job
>>> cost overall. I've rather taken a 'do it properly and do it once'
>>> approach to this house, rather than being cost-driven.
>>
>> It depends a bit on the layout and how many joist hangers you need etc.
>> Those will be the same price regardless.
>>
>>> I did toy at one stage with the prospects for making up composite
>>> joists in situ, given the access issues, but decided that would just
>>> add another variable (and potential disaster) to the equation (!)
>>>
>>> Must have another go at Thunderbird for this group. I had it
>>> working, then it wouldn't post for some reason.
>>
>> It ought to be fairly painless for newsgroups if you are using your ISPs
>> newsserver. If using a third party one, you may need to turn on the
>> "request authentication" option in the account settings.
>>
>>
> I should take a browse through the phone apps (without wishing to turn
> the thread into another debate about which phone is best!)

Hence why I did not mention a platform. I have seen them for mine, so I
presume they exist for the others... I wonder if Tony is planning on an
Android or iOS port of superbeam proper?

> Your comments about ISP newservers made me realise that I set
> Thunderbird up for home, then might have failed to post over a different
> WiFi connection, so if this posts, problem solved (ish)

Normally if the newsserver accepts authentication then it will allow
posts from any network. If it does not, then it will usually only accept
them from its "own" network address block.

BTW the quoting looks much better - no double spaced lines anymore ;-)

[Tony Bryer]

On Tue, 23 Oct 2012 01:43:39 +0100 John Rumm wrote :
> Hence why I did not mention a platform. I have seen them for mine, so I
> presume they exist for the others... I wonder if Tony is planning on an
> Android or iOS port of superbeam proper?

No current plans. Would it be an interesting project? Probably. Would it
make any money, given what people are used to paying for mobile apps?
Probably not. If I was looking for an alternative platform, then the most
likely route would be a web-based app, accessible via any browser.

[GMM]

>>>
>> I should take a browse through the phone apps (without wishing to turn
>> the thread into another debate about which phone is best!)
>
> Hence why I did not mention a platform. I have seen them for mine, so I
> presume they exist for the others... I wonder if Tony is planning on an
> Android or iOS port of superbeam proper?
>
>> Your comments about ISP newservers made me realise that I set
>> Thunderbird up for home, then might have failed to post over a different
>> WiFi connection, so if this posts, problem solved (ish)
>
> Normally if the newsserver accepts authentication then it will allow
> posts from any network. If it does not, then it will usually only accept
> them from its "own" network address block.
>
> BTW the quoting looks much better - no double spaced lines anymore ;-)
>
>

Excellent! Problem solved then (now to work on the typos....)

[John Rumm]

On 23/10/2012 06:10, Tony Bryer wrote:
> On Tue, 23 Oct 2012 01:43:39 +0100 John Rumm wrote :
>> Hence why I did not mention a platform. I have seen them for mine, so I
>> presume they exist for the others... I wonder if Tony is planning on an
>> Android or iOS port of superbeam proper?
>
> No current plans. Would it be an interesting project? Probably. Would it
> make any money, given what people are used to paying for mobile apps?
> Probably not. If I was looking for an alternative platform, then the most
> likely route would be a web-based app, accessible via any browser.

One option would be to only sell/give away the mobile app as an extra
for the full PC version - gives you the sales hook for some added value
on the main product, stops folks who already use it going to alternative
platforms since they can still use the mobile version of the "industry
standard", but without cannibalising your desktop sales.

[GMM]

On 23/10/2012 01:43, John Rumm wrote:

>>>
>> Probably not as "industry standard" or trustworthy, but there
are a number of smart phone apps about that will do calcs at least on a
single beam.
Might be worth trying one of those if you have platform for that.

It's interesting that there seem to be more apps available than
web-based calculators. Unfortunately most of these seem to be in
American and I'm not sure how C16 compares with their specs but, putting
a trial dimension of 4m in, using Redwood or Southern Pine (which seem
to come out the same and probably equate broadly) gives a deflection of
69mm for a 6x2 for a light floor loading, which seems an awful lot. For
some reason, the one I was using seems not to accept 2" beams after the
first use so I can't go back and check but the deflection for 8x2 was
about 20-something mm so about a 2" difference.
So to keep the floor clear of the ceiling, either size would have to be
mounted at the same height (roughly), which is interesting. So it might
well be best to go for the bigger size and do it 'properly' after all.
I suspect this is a worst case as adding boards and struts would stiffen
the whole thing. On the other hand, there is bound to be a bit of
gradual sag over time, as you saw, even in an unloaded joist, which
might offset this.
I suppose an alternative (to use narrower timbers) would be to change
the joist pattern but the only way I can see to do this would be to put
something very substantial across the middle of the 6m dimension (or
split it into 3) and run the joists at right angles to this. This would
make most joists around 3m, where 6x2 would probably be fine. I'm not
sure this would be wise though as it would increase the loading on the
walls to a great extent where this beam was mounted and would generate
problems with actually getting everything into place.
Thinking back to my last place, it's certainly true that we had a loft
which I floored using chipboard panels onto 4x2 ceiling joists with no
problems. I think the differences there were a) the spans were much
smaller and b) I didn't give it any thought at all!

[John Rumm]

On 24/10/2012 09:55, GMM wrote:
> On 23/10/2012 01:43, John Rumm wrote:
>
>>>>
>>> Probably not as "industry standard" or trustworthy, but there
> are a number of smart phone apps about that will do calcs at least on a
> single beam.
> Might be worth trying one of those if you have platform for that.
>
>
> It's interesting that there seem to be more apps available than
> web-based calculators. Unfortunately most of these seem to be in
> American and I'm not sure how C16 compares with their specs but, putting
> a trial dimension of 4m in, using Redwood or Southern Pine (which seem
> to come out the same and probably equate broadly) gives a deflection of
> 69mm for a 6x2 for a light floor loading, which seems an awful lot. For
> some reason, the one I was using seems not to accept 2" beams after the
> first use so I can't go back and check but the deflection for 8x2 was
> about 20-something mm so about a 2" difference.

Give me the specs of a beam if you like and I will see what numbers I get...

> So to keep the floor clear of the ceiling, either size would have to be
> mounted at the same height (roughly), which is interesting. So it might
> well be best to go for the bigger size and do it 'properly' after all.
> I suspect this is a worst case as adding boards and struts would stiffen
> the whole thing. On the other hand, there is bound to be a bit of
> gradual sag over time, as you saw, even in an unloaded joist, which
> might offset this.

When you add floor boards, you can use load sharing in the calcs - since
you tie a number of them together in effect, the statistical variation
is likely to reduce, so you can assume stiffness's no so close to worst
case.

> I suppose an alternative (to use narrower timbers) would be to change
> the joist pattern but the only way I can see to do this would be to put
> something very substantial across the middle of the 6m dimension (or
> split it into 3) and run the joists at right angles to this. This would
> make most joists around 3m, where 6x2 would probably be fine. I'm not
> sure this would be wise though as it would increase the loading on the
> walls to a great extent where this beam was mounted and would generate
> problems with actually getting everything into place.
> Thinking back to my last place, it's certainly true that we had a loft
> which I floored using chipboard panels onto 4x2 ceiling joists with no
> problems. I think the differences there were a) the spans were much
> smaller and b) I didn't give it any thought at all!

Well indeed. Also a loft just used for storage will not be loaded to
anything close to what one needs to allow for a room.

[Martin Bonner]

On Wednesday, October 17, 2012 11:49:06 AM UTC+1, GMM wrote:
>
> I have to build a floor in a loft.
> I’d like it to be robust enough that it doesn’t all wind up
> in the bedroom below.
>
>
>
> At present, there is a lath and plaster ceiling, with rather
> wimpy-looking ceiling joists,

> So question 1: To work above the ceiling, I’m thinking of supporting
> it with boards on acro jacks, possibly moving these according to
> where I’m working as I go and putting some boards on the ceiling
> joists to spread the load (mostly me!).
> Does this sound like a sensible thing to do – ie any better
> suggestions?

I'd put at a beam at right angles to the joists in the middle of the ceiling, supported by acro-props. This will halve the effective length of the joists, and make them plenty man enough for the job.

You'll then need boards on top of the joists to stand on.

> Lastly, to get decent access into the loft space I need to create a
> new doorway. The only way I can approach this is from the inside
> of the loft. Normally, cutting a new opening would be
> best done using strongboys to support the triangle of brisk above
> until the lintel is installed but I’d
> hesitate to jack against the top of the ceiling and getting them
> through the hatch wouldn’t be easy.
>
> So question 3: Instead of using strongboys, would a board (4x1 or so)
> fixed to each brick (multimonti into the brick centre) above the lintel
> do an adequate job of stopping everything moving while I get a
> lintel in?

Sounds like a reasonable plan. I'd probably just use a rawl-plug
and No 10/ No 12 screw. It doesn't have to take much weight, and it
doesn't have to take it for long.

[GMM]

On 24/10/2012 16:55, John Rumm wrote:
> On 24/10/2012 09:55, GMM wrote:
>> On 23/10/2012 01:43, John Rumm wrote:
>>
>>>>>
>>>> Probably not as "industry standard" or trustworthy, but there
>> are a number of smart phone apps about that will do calcs at least on a
>> single beam.
>> Might be worth trying one of those if you have platform for that.
>>
>>
>> It's interesting that there seem to be more apps available than
>> web-based calculators. Unfortunately most of these seem to be in
>> American and I'm not sure how C16 compares with their specs but, putting
>> a trial dimension of 4m in, using Redwood or Southern Pine (which seem
>> to come out the same and probably equate broadly) gives a deflection of
>> 69mm for a 6x2 for a light floor loading, which seems an awful lot. For
>> some reason, the one I was using seems not to accept 2" beams after the
>> first use so I can't go back and check but the deflection for 8x2 was
>> about 20-something mm so about a 2" difference.
>
> Give me the specs of a beam if you like and I will see what numbers I
> get...
>

Thanks John, that would be very useful. All lengths are 4m (a little
variation but no more than 100mm) and the TRADA tables say a 47 x 195
joist is good for that at 400 centres but of course give no further
information. (A number of apps etc simply give the same data in their
calculators - you put in the size and it tells you the max span.)

Clearly very small ones would be inadequate at this length, but it would
be interesting to know what the performance of 7" (ie 47 x 170) and 6"
(47 x 145) would be at 4m for C16 timber. From what the tables
indicate, the advantage of C24 is pretty minimal, so hardly worth the
trouble of sourcing etc except in very marginal situations).

I'll most likely cover with chipboard (8x2 t/g sheets), which comes in
18 or 22mm. I'm assuming that there's little difference here for 400mm
joist centres (and would go for 18 as it's lighter) and the more
important factor is to screw it down to every joist but I could easily
be corrected on this (!)

It's also pretty clear that a span of 4m will benefit from strutting at
mid span, so that's on the agenda too.

Friday night - must be time for a cold one (!)

Cheers!

[John Rumm]

On 26/10/2012 19:11, GMM wrote:
> On 24/10/2012 16:55, John Rumm wrote:
>> On 24/10/2012 09:55, GMM wrote:
>>> On 23/10/2012 01:43, John Rumm wrote:
>>>
>>>>>>
>>>>> Probably not as "industry standard" or trustworthy, but there
>>> are a number of smart phone apps about that will do calcs at least on a
>>> single beam.
>>> Might be worth trying one of those if you have platform for that.
>>>
>>>
>>> It's interesting that there seem to be more apps available than
>>> web-based calculators. Unfortunately most of these seem to be in
>>> American and I'm not sure how C16 compares with their specs but, putting
>>> a trial dimension of 4m in, using Redwood or Southern Pine (which seem
>>> to come out the same and probably equate broadly) gives a deflection of
>>> 69mm for a 6x2 for a light floor loading, which seems an awful lot. For
>>> some reason, the one I was using seems not to accept 2" beams after the
>>> first use so I can't go back and check but the deflection for 8x2 was
>>> about 20-something mm so about a 2" difference.
>>
>> Give me the specs of a beam if you like and I will see what numbers I
>> get...
>>
>
> Thanks John, that would be very useful. All lengths are 4m (a little
> variation but no more than 100mm) and the TRADA tables say a 47 x 195
> joist is good for that at 400 centres but of course give no further
> information. (A number of apps etc simply give the same data in their
> calculators - you put in the size and it tells you the max span.)

Well if I model that with a uniformly distributed load of 0.8kN/m (long
term load), and treat them as a load sharing system, you get a
deflection of just under 11mm. Go down to 175mm and the deflection goes
up to just under 15mm. However its worth noting that it is also flagged
as failing in maximum permitted bending stress at that point. 150mm and
deflection is nearly 23mm. 100mm and deflection is approaching 80mm!

That is a full floor load though - unless storing a large magazine or
record collection its unlikely you would reach that.

> Clearly very small ones would be inadequate at this length, but it would
> be interesting to know what the performance of 7" (ie 47 x 170) and 6"
> (47 x 145) would be at 4m for C16 timber. From what the tables
> indicate, the advantage of C24 is pretty minimal, so hardly worth the
> trouble of sourcing etc except in very marginal situations).

C24 at 175 for example would be a "pass" on 175mm (deflection 7.61mm)
and on 150mm (11.98). (125mm and you are back to failing in bending stress)

Note also you can push it firther with short term loads - so you don't
necessarily need to make adjustments to factor in someone clod hoppering
around up there shifting the boxes.

> I'll most likely cover with chipboard (8x2 t/g sheets), which comes in
> 18 or 22mm. I'm assuming that there's little difference here for 400mm
> joist centres (and would go for 18 as it's lighter) and the more
> important factor is to screw it down to every joist but I could easily
> be corrected on this (!)

It interlocks, so only really needs enough screwing in this application
to stop it sliding about.

> It's also pretty clear that a span of 4m will benefit from strutting at
> mid span, so that's on the agenda too.

It will make it bounce less. How much that matters for storage space is
debatable.

> Friday night - must be time for a cold one (!)

Sounds like a good plan ;-)

[GMM]

Thanks for doing that John, it's very interesting: It looks like 150 at
C24 is pretty much comparable to 195 at C16 in terms of deflection: I
didn't think the difference between the two would be anywhere near as
much, from the other information I've seen.
Of course, the C24 may be more expensive even in the smaller size (I'll
have to get a quote) but if it's not too much more, the extra 2" might
be a useful space saving: It would certainly reduce the step up onto
this floor from the new access door, which might make it more user friendly.

(It took me a while to get back - I've been away for a few days and it
seems Firefox can't connect to the server from other networks)

All the best

[John Rumm]

I remember when I did my loft conversion that there was some C24
specified (most was C16). It took the timber merchant about an extra
week to order that in for me. Hence it may not be as readily stocked.

> (It took me a while to get back - I've been away for a few days and it
> seems Firefox can't connect to the server from other networks)

Thunderbird perhaps?

If your mail server supports authentication then you may be able to use
it by turning on the "Always request authentication when connecting to
this server" checkbox in the news server server settings dialogue.