Google Groups no longer supports new Usenet posts or subscriptions. Historical content remains viewable.
Dismiss

Remove 13 ft. bearing wall - Beam choices?

1,030 views
Skip to first unread message

Bill

unread,
Apr 24, 2008, 2:06:07 PM4/24/08
to
Before I go and hire an architect, get building permit, etc. I would like to
get a ballpark on what this might look like...

I have a bearing wall (2x4 studs) which is 13 ft. long in my living room
which I am thinking of tearing out and replacing with a beam. (2 story
house - living room first floor.)

But looking at "span tables", this is looking like it would need a solid
wood beam like 4 x 10? Well a 10" beam would come down a bit far and not
look so good...

So how about a steel I-beam? I looked at span tables for I-beams and it
looks like a "W6x9" would do the trick? This would come down less and look
better.

Anyway I don't know a thing about steel I-beams. Does "W6x9" mean 6 inches
wide and 9 inches high?

Is there a smaller I-beam which would work for this span?

And I have 2x4 walls that this I-beam would connect to on each side. Would
the I-beam just rest on say 4 x 6 wood posts? Or have holes drilled in the
bottom of the I-beam and lag screw it to the wood posts?

And how would I fasten the joists resting on the top of the I-beam to the
I -beam?


hal...@aol.com

unread,
Apr 24, 2008, 2:13:37 PM4/24/08
to

you must support wall on both sides before removing.. temporary walls.

what you need is probably a engineered wood beam pair with a stell
plate insert.

get this evaluated by a structural engineer before proceeding. they
can give you specifics

jloomis

unread,
Apr 24, 2008, 2:43:23 PM4/24/08
to
You can upgrade the beam, and install it in the cieling......That is joist
hang the ceiling joists to the beam and install it in the cieling.....Yes
you need to support the cieling with temporary.
Also where the beam is supported on either side, you need to make sure you
have solid post and underfloor support i.e. blocking and pier support
directly under posts.....
What size are the cieling joists......
With that span of 13 ft. and your live load upper part, you should be able
to calc out a beam and or oversize it. What about a 6x10? Laminate beams
are stronger but not decorative.
with I beams usually wood is attached that the joists are attached to.
I have done many types of wall / ceiling supports with and without
engineering.
4x4 4 feet
4x6 6 feet
4x8 8 ft....
4x10 10 ft
a 4x12 will span 12 ft.......
When using 6x......they will increase the span....
jloomis
"Bill" <billnoma...@yahoo.com> wrote in message
news:67biegF...@mid.individual.net...

Howard

unread,
Apr 24, 2008, 2:46:23 PM4/24/08
to

As you are planning please remember that the "posts" supporting the
new beam must transfer their load onto a proper foundation or
location. That load may include the attic and roof loads. The
foundation beneath the first floor wall will need to be reviewed
whether it is a slab, basement or crawlspace. You can also place the
beam in the structure between the first and second floor by supporting
the ceiling/floor as described above, cutting all the floor joists,
inserting the beam and using joist hangers to tie the joists to the
side of the new cross beam.

I did this once and used an engineered wood beam because it is easier
to attach the joists and the supporting posts. I'm a Professional
Engineer and knew that I had to check the foundation, side supports,
the connections, etc. Oh and start looking for any pipes, vents and
electrical lines in the wall you are removing, all will have to be
relocated.

Robert Allison

unread,
Apr 24, 2008, 3:30:36 PM4/24/08
to
Bill wrote:
> Before I go and hire an architect, get building permit, etc. I would like to
> get a ballpark on what this might look like...
>
> I have a bearing wall (2x4 studs) which is 13 ft. long in my living room
> which I am thinking of tearing out and replacing with a beam. (2 story
> house - living room first floor.)
>
> But looking at "span tables", this is looking like it would need a solid
> wood beam like 4 x 10? Well a 10" beam would come down a bit far and not
> look so good...
>
> So how about a steel I-beam? I looked at span tables for I-beams and it
> looks like a "W6x9" would do the trick? This would come down less and look
> better.
>
> Anyway I don't know a thing about steel I-beams. Does "W6x9" mean 6 inches
> wide and 9 inches high?

No, W6x9 means that it is 6 lbs per foot and nine inches tall.

> Is there a smaller I-beam which would work for this span?

An engineered wood beam multiplied may do it, but you would have
to check with the span tables for them.

> And I have 2x4 walls that this I-beam would connect to on each side. Would
> the I-beam just rest on say 4 x 6 wood posts? Or have holes drilled in the
> bottom of the I-beam and lag screw it to the wood posts?

Typically, a metal beam on a wood post would have clips welded
onto it to fasten to the support posts. There are various ways
of attachment.

> And how would I fasten the joists resting on the top of the I-beam to the
> I -beam?

This would again entail clips either welded or bolted to the beam
and fastened to the joists.

As other posters have pointed out, you must make sure to transfer
the load to your foundation appropriately. It does not sound
like you have the experience to judge all the possible
configurations for doing this properly. An engineer can well
make this project doable and keep within a budget for doing so.

When I do things like this on my own, I always overengineer,
because I cannot properly do the calculations, so I overbuild to
compensate. The engineer can save you money by specifying
enough, without overdoing it. The cost of the engineer can often
be saved by the savings from less material and less work. YMMV.

As far as what it will cost, if you do the legwork (make accurate
drawings, take a lot of photos of both the foundation under the
support posts and the floor above and the situation itself) then
the engineer can design a system for you without having to do all
of that himself. Design costs; probably 300 to 500 bucks if it
is simple. Build costs; depends, but if it is simple and the
foundation can support everything as it is, then 1,000 bucks or so.

Well worth the piece of mind to hire the engineer. He sees
things that you can't.

--
Robert Allison
Rimshot, Inc.
Georgetown, TX

Dave in Houston

unread,
Apr 24, 2008, 4:35:43 PM4/24/08
to

you must support wall on both sides before removing.. temporary walls.

You don't. Here's the way we do it:
http://www.pbase.com/speedracer/image/2622653

Set your beam in the attic space directly over the existing wall. Bolt all
your ceiling joist to the new beam as in the photo.
Tear down existing wall - nothing moves.

Dave in Houston


Joseph Meehan

unread,
Apr 24, 2008, 4:51:29 PM4/24/08
to
<hal...@aol.com> wrote in message
news:28fc8532-7f0c-4356...@m73g2000hsh.googlegroups.com...

> On Apr 24, 10:06�am, "Bill" <billnomailnosp...@yahoo.com> wrote:
...

>
> get this evaluated by a structural engineer before proceeding. they
> can give you specifics

That is the most important part.


--
Joseph Meehan

Dia 's Muire duit

hal...@aol.com

unread,
Apr 24, 2008, 5:06:47 PM4/24/08
to

get pro help the roof rafters MIGHT not be strong enough to take tha
added load..........

Wayne Whitney

unread,
Apr 24, 2008, 5:13:00 PM4/24/08
to
On 2008-04-24, hal...@aol.com <hal...@aol.com> wrote:

> On Apr 24, 12:35�pm, "Dave in Houston" <DeeJayEmCee...@att.net> wrote:
>
> > Set your beam in the attic space directly over the existing
> > wall. Bolt all your ceiling joist to the new beam as in the photo.
> > Tear down existing wall - nothing moves.
>

> get pro help the roof rafters MIGHT not be strong enough to take

> that added load..........

The roof rafters don't come into it--although in the photo they appear
to be right above the new beam, there is no connection between the
rafters and the new beam.

Cheers, Wayne

dpb

unread,
Apr 24, 2008, 5:52:49 PM4/24/08
to
Robert Allison wrote:
...

> When I do things like this on my own, I always overengineer, because I
> cannot properly do the calculations, so I overbuild to compensate. ...

If you don't do the calculations, how do you know you're "over"-engineered?

:)

--

Robert Allison

unread,
Apr 24, 2008, 6:09:12 PM4/24/08
to

Not to mention that there is an entire floor between the wall he
wants to take out and the attic.

Robert Allison

unread,
Apr 24, 2008, 6:14:51 PM4/24/08
to
dpb wrote:

I can do the calculations on alot of things, but, from my
experience, anything I design gets downsized if an engineer looks
at it. I have worked with a lot of engineers, sometimes I use
them, some times I don't. I know generally what is required and
how to use span tables, but....

For instance, if I need to put a footer under a post load, my
footer will be much larger than what an engineer would design. I
do that "just to make sure".

Every time I take something that I designed to my favorite
engineer he always asks me "how many Abrams tanks I was planning
to support with this?".

hogheavenfarm

unread,
Apr 24, 2008, 7:10:46 PM4/24/08
to
I have just completed the same thing. Took out a 14' section of wall
(exterior) between the living room and a porch that was added on. I
used engineered (LVL) or as we say Microlam beam built of three
2x12's. Remember this is nominal size, steel isnt. When you get a 9" I
beam , it is 9". A 12" LVL is only 11 1/4" avtual size.
I looked at the "beam-over" pics, good idea, but I was concerned
about the bottom chord of the rafter overlap, looked like only 4-6",
should be more. Others have correctly pointed out, that the posts must
rest on solid bases, all the way down.I'm using 4 2x4's on each side
carrying the load to a double bottom plate which rests directly on the
steel I beam in the basement. If you are doing an outside wall, there
will naturally be something underneath, the problem comes when you
open up an interior wall (load bearing) and post it on nothing. LVL
beams are easier to work with since you can put them up one at a time
and nail them together, no heavy lifting like a steel beam, and no
boxing in problems due to the steel.
For your application a 9" nominal LVL would work, (doubled or
tripled), which would give you the headroom you need.

Dave in Houston

unread,
Apr 24, 2008, 7:52:15 PM4/24/08
to

"hogheavenfarm" <wmhqu...@verizon.net> wrote in message
news:d9cda74c-f354-49aa...@26g2000hsk.googlegroups.com...


> I looked at the "beam-over" pics, good idea, but I was concerned
> about the bottom chord of the rafter overlap, looked like only 4-6",
> should be more.

What you cannot see in the picture is that the angle iron brackets are
thru-bolted to both sets of ceiling joists, on both sides of the beam.

Others have correctly pointed out, that the posts must
> rest on solid bases, all the way down.

Here's another shot of the new space minus the old wall. The end
support for the beam in this case consisted of four 2X4s, three ganged side
by side in the outside wall cavity and one sistered on the interior side
perpendicular which you can see in the pic.

http://www.pbase.com/speedracer/image/2622654

Dave in Houston


dpb

unread,
Apr 24, 2008, 8:22:28 PM4/24/08
to
Robert Allison wrote:
> dpb wrote:
>
>> Robert Allison wrote:
>> ...
>>
>>> When I do things like this on my own, I always overengineer, because
>>> I cannot properly do the calculations, so I overbuild to compensate.
>>> ...
>>
>>
>> If you don't do the calculations, how do you know you're
>> "over"-engineered?
>>
>> :)
>>
>> --
>
> I can do the calculations on alot of things, but, from my experience,
> anything I design gets downsized if an engineer looks at it. ...

You miss the smiley? Hope not... :) It was only intended as a joke,
not an aspersion...

--

Robert Allison

unread,
Apr 24, 2008, 9:32:48 PM4/24/08
to

No, I saw it. I just wanted to be clear to everyone else. I
figured that you knew what I was talking about!

BobK207

unread,
Apr 25, 2008, 3:02:55 AM4/25/08
to

Dave-

That is a pretty cool design concept to replace a wall with a beam.

But the joists will move downward until the beam deflects enough to
take the load that the wall was supporting.....depending on the beam
sizing & the ceiling load (actually in this case, the 2nd story floor
& ?). The deflection at mid span could be in the 3/8 to 1/2" range.

Plus in the OP's I'm pretty sure that your concept would place the
beam on the floor of the 2nd story.

But still a neat concept.

cheers
Bob

BobK207

unread,
Apr 25, 2008, 3:06:30 AM4/25/08
to

Unless someone else has fielded this

W6x9 means a wide flange beam nominally 6" deep & weighing 9 lbs/ft

cheers
Bob

Wayne Whitney

unread,
Apr 25, 2008, 3:12:01 AM4/25/08
to
On 2008-04-25, BobK207 <rkaz...@gmail.com> wrote:

> But the joists will move downward until the beam deflects enough to
> take the load that the wall was supporting.....depending on the beam
> sizing & the ceiling load (actually in this case, the 2nd story
> floor & ?). The deflection at mid span could be in the 3/8 to 1/2"
> range.

This could be addressed by using a pre-cambered beam, like a glulam.

Cheers, Wayne

Dave in Houston

unread,
Apr 25, 2008, 3:37:00 AM4/25/08
to

"BobK207" <rkaz...@gmail.com> wrote in message
news:690fc0eb-2cf6-4ea5...@k10g2000prm.googlegroups.com...

Dave-

That is a pretty cool design concept to replace a wall with a beam.

But the joists will move downward until the beam deflects enough to
take the load that the wall was supporting.....depending on the beam
sizing & the ceiling load (actually in this case, the 2nd story floor
& ?). The deflection at mid span could be in the 3/8 to 1/2" range.

Plus in the OP's I'm pretty sure that your concept would place the
beam on the floor of the 2nd story.

But still a neat concept.


Truth to tell, I missed the fact that the OP was dealing with another
story vs. an open attic or I probably never would have posted this as a
solution.
As far as your concerns about deflection, we measured absolutely zero in
this particular application. I do recognize the necessity of sizing the
beam to carry the load. But, in this case the load is 2X6 ceiling joists
and half-inch painted drywall and the beam consists of 3/4 inch CDX plywood
sandwiched between SYP 2X12s though I recall there being either two 10 foot
2X12s on either side of three pieces of plywood or an eight and a ten on one
side and a ten and an eight on the other. I don't recall the length of the
span except that it was around 18 or 20 feet. We must have put a couple of
hundred 3 1/4 inch nails in it in addition to the thru-bolts in each angle
iron brackets. We probably even considered a 1/4 inch steel flitch plate in
lieu of the 3/4 plywood. Assembling the beam in the attic space was hard
enough w/o having to contend with a couple of pieces of 1/4 inch by 11 X
eight or ten foot steel plate.

Dave in Houston


Smitty Two

unread,
Apr 25, 2008, 5:46:34 AM4/25/08
to
In article <Px5Qj.230$1b7...@newssvr13.news.prodigy.net>,

One thing I don't see discussed very often on ahr is sheer strength.
AIUI, houses with a lot of doors and windows often use an interior wall
not only for load bearing, but also for sheer. It's worth considering.
When I added a window and a pocket closet door in the bedroom, I had to
make sheer strength modifications to what was left of the walls.

David Nebenzahl

unread,
Apr 25, 2008, 7:23:41 AM4/25/08
to
On 4/24/2008 10:46 PM Smitty Two spake thus:

> One thing I don't see discussed very often on ahr is sheer strength.
> AIUI, houses with a lot of doors and windows often use an interior wall
> not only for load bearing, but also for sheer. It's worth considering.
> When I added a window and a pocket closet door in the bedroom, I had to
> make sheer strength modifications to what was left of the walls.

I think you meant to type "shear". But your points stand.


--
The best argument against democracy is a five-minute
conversation with the average voter.

- Attributed to Winston Churchill

Message has been deleted

RicodJour

unread,
Apr 25, 2008, 11:50:58 AM4/25/08
to
On Apr 25, 4:51 am, sonofa...@nonetoday.com wrote:
>
> People who rip out load bearing walls are normally idiots. The wall
> was put there for support and is intended to stay there. If you want
> a visual of the next room, put in a few smaller windows so only every
> other stud is removed and beef up those that stay.
>
> People who think that a house is going to remain solid and survive in
> severe storms leave their houses with their original structures. Only
> those Saturday morning home re-make shows knock out load bearing walls
> to create lots of open space. Of course the tv viewer never sees the
> house a few years later when the roof sags, or sees what occurs during
> a tornado.

An excellent example of frontier gibberish spoken with the conviction
of someone who has no knowledge of construction. Aspiring trolls
please note.

R

Bobk207

unread,
Apr 25, 2008, 2:33:21 PM4/25/08
to
On Apr 24, 8:12 pm, Wayne Whitney <whit...@post.harvard.edu> wrote:

Wayne-

Yes, the beam, if cambered properly, would deflect to a straight
condition.

But the ceiling / 2nd story floor would have to be jacked along with
the beam for the entire system to have no deflection....you would need
to enforce displacement compatibility (& a raised starting point) to
prevent a sag when the walls were removed.

It's kinda like getting a sistered joist to really share the load.

cheers
Bob

Smitty Two

unread,
Apr 25, 2008, 2:39:40 PM4/25/08
to
In article <481186d2$0$19801$8226...@news.adtechcomputers.com>,
David Nebenzahl <nob...@but.us.chickens> wrote:

> On 4/24/2008 10:46 PM Smitty Two spake thus:
>
> > One thing I don't see discussed very often on ahr is sheer strength.
> > AIUI, houses with a lot of doors and windows often use an interior wall
> > not only for load bearing, but also for sheer. It's worth considering.
> > When I added a window and a pocket closet door in the bedroom, I had to
> > make sheer strength modifications to what was left of the walls.
>
> I think you meant to type "shear". But your points stand.

Yep, thanks. Not too many homonyms fool me, and that's a bad one to get
wrong since it changes the meaning of the word. Shear it is, and I'll
remember it.

Bobk207

unread,
Apr 25, 2008, 2:44:50 PM4/25/08
to
On Apr 25, 1:51 am, sonofa...@nonetoday.com wrote:
> On Thu, 24 Apr 2008 07:06:07 -0700, "Bill"

>
>
>
> <billnomailnosp...@yahoo.com> wrote:
> >Before I go and hire an architect, get building permit, etc. I would like to
> >get a ballpark on what this might look like...
>
> >I have a bearing wall (2x4 studs) which is 13 ft. long in my living room
> >which I am thinking of tearing out and replacing with a beam. (2 story
> >house - living room first floor.)
>
> >But looking at "span tables", this is looking like it would need a solid
> >wood beam like 4 x 10? Well a 10" beam would come down a bit far and not
> >look so good...
>
> >So how about a steel I-beam? I looked at span tables for I-beams and it
> >looks like a "W6x9" would do the trick? This would come down less and look
> >better.
>
> >Anyway I don't know a thing about steel I-beams. Does "W6x9" mean 6 inches
> >wide and 9 inches high?
>
> >Is there a smaller I-beam which would work for this span?
>
> >And I have 2x4 walls that this I-beam would connect to on each side. Would
> >the I-beam just rest on say 4 x 6 wood posts? Or have holes drilled in the
> >bottom of the I-beam and lag screw it to the wood posts?
>
> >And how would I fasten the joists resting on the top of the I-beam to the
> >I -beam?
>
> People who rip out load bearing walls are normally idiots.  The wall
> was put there for support and is intended to stay there.  If you want
> a visual of the next room, put in a few smaller windows so only every
> other stud is removed and beef up those that stay.  
>
> People who think that a house is going to remain solid and survive in
> severe storms leave their houses with their original structures.  Only
> those Saturday morning home re-make shows knock out load bearing walls
> to create lots of open space.  Of course the tv viewer never sees the
> house a few years later when the roof sags, or sees what occurs during
> a tornado.


>>>>>> People who rip out load bearing walls are normally idiots. <<<<<<

They could very well be idiots but normally? No.

Modifying a structure after it's built is not that dissimilar to
modifying while it's still in the design phase....though one needs to
use constrcution tools rather than a keyboard or an eraser.

>>>>The wall was put there for support and is intended to stay
there.<<<

Yes that is (was) true when the house was built.

But as long as one determines the required structural capacity &
replaces (or in some case because of code changes, increases it) the
structure will be fine.....that's why we have design folks. :)


>>>People who think that a house is going to remain solid and survive in
severe storms leave their houses with their original structures. <<<<

Improperly done, wall removals can weaken a structure but done
properly you'll wind up with a structure that is as strong usually
stronger due to increases in capacity demanded by the code.

cheers
Bob

Wayne Whitney

unread,
Apr 25, 2008, 3:08:28 PM4/25/08
to
On 2008-04-25, Bobk207 <rkaz...@gmail.com> wrote:

> On Apr 24, 8:12 pm, Wayne Whitney <whit...@post.harvard.edu> wrote:
>
> > This could be addressed by using a pre-cambered beam, like a glulam.
>

> But the ceiling / 2nd story floor would have to be jacked along with
> the beam for the entire system to have no deflection....you would
> need to enforce displacement compatibility (& a raised starting
> point) to prevent a sag when the walls were removed.

Good point. I guess if the joists are unloaded at the time of
installing the precambered beam, it wouldn't be too hard. You could
just clamp/jack each joist to the pre-cambered beam one at a time
before fastening them.

But if the joists are attached to each other transversely at the time
of beam installation, then you have to jack the whole system, and in
the shape of the pre-camber. That could be trouble, but doable.

Cheers, Wayne

Bill

unread,
Apr 25, 2008, 3:25:41 PM4/25/08
to
"Robert Allison" wrote in message
>
> ...As far as what it will cost, if you do the legwork (make accurate
> drawings, take a lot of photos of both the foundation under the support
> posts and the floor above and the situation itself) then the engineer can
> design a system for you without having to do all of that himself....
>

That sounds like a good idea. I can tear out a foot or so of the drywall on
the ceiling on either side of the bearing wall and remove a foot of drywall
on the side walls both directions. Then take pictures of everything
including the crawl space under all this. Then provide this along with
measurements to the engineer.

Note: Under all this is are TREE TRUNKS! This is an old house built in the
1930's. For support beams under the house... Under each outside wall and
under each bearing wall, there is an entire 11 inch round tree trunk with
the top side sawed flat and pier supports under it. Actually there are
already pier supports at both ends of where the bearing wall ends (Where
supports for a beam would go).

The 2nd floor (1st floor ceiling) has closely spaced 2 x 4 joists resting on
this bearing wall (the 2 x 4's span 10 ft.). And not evenly spaced either. I
don't think this is enough support (I would think 2 x 6's would be better),
but the house is still standing - just "creaks" a bit when walking upstairs!

Note that the upstairs sub-floor is nailed into these 2 x 4's with zillions
of nails. So ripping out the 2 x 4's and replacing with 2 x 6's would be
fun!

I understand that I would need to support everything on either side before
removing the bearing wall and that I would need to transfer the load down
through the subfloor.


0 new messages