Design principles

[Alexander Lamaison]

Are there any basic rules governing the design of a wooden structure?

I'm trying to improve my wood working skills, and my next project is a
log store for the garden. My design so far is here [1], here [2] and
here [3].

Usually I decide on the structure by vague 'intuition' about weight and
load, but I'd much rather apply some tried and tested strategy. For
instance, how do you decide on the relative size of the members? And
which members transmit load to others?

Any advice is greatly appreciated.

[1] https://dl.dropboxusercontent.com/u/6028779/Wood%20store%201.png
[2] https://dl.dropboxusercontent.com/u/6028779/Wood%20store%201.png
[3] http://sketchup.google.com/3dwarehouse/details?mid=b771b29ae4bc6fd91153797f3361d543

Alex

--
Swish - Easy SFTP for Windows Explorer (http://www.swish-sftp.org)

[Lee Michaels]

"Alexander Lamaison" <aw...@doc.ic.ac.uk> wrote in message
news:86ppwy3...@doc.ic.ac.uk...

I see that you are from the UK. In the states we use a little different
terminology. I assume that this is for firewood. That is wood that will be
burned in a stove or fireplace? Is that correct?

Rather than get all concerned about design principles and strength, I would
first define EXACTLY what the structure is used for. Then go about building
something to suit that purpose. As I have built a number of "woodsheds" and
"firewood stands", I will tell you what I think.

1) Structures to house firewood simply do not have to meet a higher
standard used for houses, etc. They are often built out of whatever is
lying around or recycled materials. And if the only thing that is needed is
to keep the rain off, you don't need much.

2) I find it interesting that you have two layers. I have never done this
or found it necessary. Having it open front and back is nice to air out the
wood and let it dry. The increased circulation, especially in a wet
climate, would definitely help with that. But when the rainy season begins,
you may need to cover the front and back to protect the wood.

3) The last "firewood structure" I built was simply a platform to keep the
wood off the ground with a simple 2 X 4 frame at the ends and over the top.
I then put up some heavy duty plastic tarps over it and tied them to the
frame with the embedded grommets. I built the frame based on the size of
available tarps. This was nice because I did not have to wrestle with the
tarps when retrieving wood. I just pulled the front flap aside and got out
my wood.

4) Unless you have some severe size restrictions or need to create
something unusually pretty, you are really over thinking a simple, basic
structure. By not having the second level you substantially simplify the
structure. This makes it much simple and cheaper to build.

5) A lot of the time, I just put some wood down to keep the fire wood off
the ground. And then covered it with a tarp. That is as simple and basic
as it gets. You want to do more, fine. I just don't think a pretty little
cottage is all that necessary. But you do what you (or the missus) want.

Any way, just tell us a little more information. Exactly what you are
trying to do and what the structure will be used for. We can then fine tune
out advice a little more.

[Keith Nuttle]

The basic problem I see with your design is angular bracing. As I see
your design, it would seem that over time the unit would wobble and the
joints loosen up, with ultimate collapse.

There would be two ways to prevent this. The simplest would be to add
pieces in the corners 45 degrees to the up rights. This could be
plywood triangles or short piece cut at 45 degrees so they would fit
into the corners.

If you want to keep the openness of the structure I would half lap all
of the joints. Half laps are quick and easy to cut and are very strong
joints. They may not take any more time that cutting and installing
all of the 45 degree braces above.

It you really want to get fancy you could make mortise and tennon
joints, but I think that is over kill, unless you want the practice for
a future project.

[Bill]

Alexander Lamaison wrote:
> Are there any basic rules governing the design of a wooden structure?

This is a very, very, very old problem. Trees didn't come with
directions. Indeed, all trees are different and many
of them are probably not very happy about what you have in mind! : )

More specifically with regard to your question, I'm reminded of the
words "form" and "function".
Ignoring "basic rules" there are entire cultures of ideas of how to
transform the flesh of a tree into something more noble.

I like this web site--after numerous visits, I still haven't seen
eveything on it.
http://www.museumfurniture.com/

Bill

[Alexander Lamaison]

"Lee Michaels" <leemichaels*nadaspam* at comcast dot net> writes:

> "Alexander Lamaison" <aw...@doc.ic.ac.uk> wrote in message
> news:86ppwy3...@doc.ic.ac.uk...
>> Are there any basic rules governing the design of a wooden structure?
>>
>> I'm trying to improve my wood working skills, and my next project is a
>> log store for the garden. My design so far is here [1], here [2] and
>> here [3].
>>
>> Usually I decide on the structure by vague 'intuition' about weight and
>> load, but I'd much rather apply some tried and tested strategy. For
>> instance, how do you decide on the relative size of the members? And
>> which members transmit load to others?
>>
>> Any advice is greatly appreciated.
>>
>> [1] https://dl.dropboxusercontent.com/u/6028779/Wood%20store%201.png
>> [2] https://dl.dropboxusercontent.com/u/6028779/Wood%20store%201.png
>> [3]
>> http://sketchup.google.com/3dwarehouse/details?mid=b771b29ae4bc6fd91153797f3361d543
>>
> I see that you are from the UK. In the states we use a little
> different terminology. I assume that this is for firewood. That is
> wood that will be burned in a stove or fireplace? Is that correct?

Quite so.

> Rather than get all concerned about design principles and strength, I
> would first define EXACTLY what the structure is used for. Then go
> about building something to suit that purpose. As I have built a
> number of "woodsheds" and "firewood stands", I will tell you what I
> think.
>
> 1) Structures to house firewood simply do not have to meet a higher
> standard used for houses, etc. They are often built out of whatever
> is lying around or recycled materials. And if the only thing that is
> needed is to keep the rain off, you don't need much.
>
> 2) I find it interesting that you have two layers. I have never done
> this or found it necessary.

The reasoning is that the logs are not all coming from the same source
or at the same time. The layers allow new unseasoned logs to be added
without making it hard to get already seasoned logs out.

> Having it open front and back is nice to air out the wood and let it
> dry. The increased circulation, especially in a wet climate, would
> definitely help with that. But when the rainy season begins, you may
> need to cover the front and back to protect the wood.

Originally it had a back but, as I'm going to be putting it against the
brick wall of the garage, I got rid of it to cut the cost. It's pretty
much always rainy season here but if it gets especially nasty I could
attach tarpaulin to the front which rolls down - seen that done a few
times.

> 3) The last "firewood structure" I built was simply a platform to
> keep the wood off the ground with a simple 2 X 4 frame at the ends and
> over the top. I then put up some heavy duty plastic tarps over it and
> tied them to the frame with the embedded grommets. I built the frame
> based on the size of available tarps. This was nice because I did not
> have to wrestle with the tarps when retrieving wood. I just pulled
> the front flap aside and got out my wood.
>
> 4) Unless you have some severe size restrictions or need to create
> something unusually pretty, you are really over thinking a simple,
> basic structure. By not having the second level you substantially
> simplify the structure. This makes it much simple and cheaper to
> build.
>
> 5) A lot of the time, I just put some wood down to keep the fire wood
> off the ground. And then covered it with a tarp. That is as simple
> and basic as it gets. You want to do more, fine. I just don't think
> a pretty little cottage is all that necessary. But you do what you
> (or the missus) want.
>
> Any way, just tell us a little more information. Exactly what you are
> trying to do and what the structure will be used for. We can then
> fine tune out advice a little more.

I realise I'm thinking about this too much given what the project it,
but that's because I want to learn things I can apply more generally to
future projects, not just this little old log store. Even if I go with
a simple platform+tarps jobbie, I'm curious what I should be thinking
about if I were building something more substantial.

But let me narrow the requirements:
- Keep logs dry
- Allow air to circulate round logs
- Hold weight of logs when full
- Last, say, 10 years with preservative occasional sprayed on it

Thanks for the advice,

[Alexander Lamaison]

Thanks for the advice. So that's bracing _or_ half-lap joints?

[Alexander Lamaison]

Bill <BILL_...@whoknows.net> writes:

> Alexander Lamaison wrote:
>> Are there any basic rules governing the design of a wooden structure?
>
> This is a very, very, very old problem. Trees didn't come with
> directions. Indeed, all trees are different and many
> of them are probably not very happy about what you have in mind! : )

Being such an old problem, I don't know why there seems to be little
advice. All woodworking books I've got my hands on spend the their time
describing the million-and-one ways of making a joint. None give advice
on how to structure the pieces you are joining. Most don't even explain
when to choose one joint over another.

> More specifically with regard to your question, I'm reminded of the
> words "form" and "function".
> Ignoring "basic rules" there are entire cultures of ideas of how to
> transform the flesh of a tree into something more noble.

I guess I'm interested in function more than form. How you design
something with to use minimal effort/material but still be confident it
will do the job. Up till now I either hope or overengineer it. I'd
like a better method.

> I like this web site--after numerous visits, I still haven't seen
> eveything on it.
> http://www.museumfurniture.com/

I'll trawl that this evening.

Thanks,

[dpb]

On 5/10/2013 1:36 PM, Alexander Lamaison wrote:
...

> Being such an old problem, I don't know why there seems to be little
> advice. All woodworking books I've got my hands on spend the their time
> describing the million-and-one ways of making a joint. None give advice
> on how to structure the pieces you are joining. Most don't even explain
> when to choose one joint over another.
>

...

You're looking in wrong places...

<http://www.fpl.fs.fed.us/products/publications/several_pubs.php?grouping_id=100&header_id=p>

What you seem to ask is more to do w/ engineering and is what architects
and structural engineers do. It's not common in the recreational
woodworking literature because there's such a widespread conventional
set of dimensions used for virtually all such pieces that amateurs or
semi- to moderately-advanced woodworkers tend to build that there's
little real need for actual stress or sizing calculations. Really
experienced will have enough through their experience to be able to
extrapolate or will consult w/ knowledgeable folks who can do that.
Otherwise, it's really all about the design for appearance.

OTOH, the other place where there's real engineering for wood products
is w/ the manufacturers who design for production and efficiency and cost.

As for the complaint of joint selection for purpose and the like, I
would recommend Tage Frid's series of books published by Taunton Press
probably 30 yr ago. While somewhat dated in their appearance by today's
standards of publishing, they're very well done and not much better than
I know of yet.

And, of course, for structural design, there are building codes that
provide minimum requirements for structures that include things like
live/dead loads, deflection limits, wind and snow loads and such things
as hurricane/tornado/earthquake enhancements for locations affected.
Don't know the UK equivalent, in the US it's known as the Uniform
Building Code.

--

--

--

[Alexander Lamaison]

dpb <no...@non.net> writes:

> On 5/10/2013 1:36 PM, Alexander Lamaison wrote:
> ...
>
>> Being such an old problem, I don't know why there seems to be little
>> advice. All woodworking books I've got my hands on spend the their time
>> describing the million-and-one ways of making a joint. None give advice
>> on how to structure the pieces you are joining. Most don't even explain
>> when to choose one joint over another.
>>
> ...
>
> You're looking in wrong places...
>
> <http://www.fpl.fs.fed.us/products/publications/several_pubs.php?grouping_id=100&header_id=p>
>
> What you seem to ask is more to do w/ engineering and is what
> architects and structural engineers do. It's not common in the
> recreational woodworking literature because there's such a widespread
> conventional set of dimensions used for virtually all such pieces that
> amateurs or semi- to moderately-advanced woodworkers tend to build
> that there's little real need for actual stress or sizing
> calculations.

I'm not trying to build anything as engineering-critical as a building
... yet. Just every day items: log store, yard gate, kitchen cabinet,
TV table, wash stand, window frame etc.

I'm not aware of a "conventional set of dimensions" so maybe that's
where I'm struggling. I didn't even realise there was such a thing.
Perhaps you could enlighten me. Let's say I wanted to build a simple
table to hold a small television; how would you size the pieces that
make this up?

> As for the complaint of joint selection for purpose and the like, I
> would recommend Tage Frid's series of books published by Taunton Press
> probably 30 yr ago. While somewhat dated in their appearance by
> today's standards of publishing, they're very well done and not much
> better than I know of yet.

It must be good. Even old versions are still expensive on Amazon :P

Scanning the contents list it still reads like the others: let's go
through a load of tools, let's go through a load of joints. Hopefully
its interspersed with higher-level information about good structure as
well.

[Keith Nuttle]

This is shows how half laps are cut

http://en.wikipedia.org/wiki/Lap_joint

The strength come from the fact that the weight of the cross member is
borne by half of the up right. The sides of the cut prevents the joint
from twisting, and keeps it square.

http://www.gazebodepot.com/gazebos/corner.asp

These are a lot fancier that what you need but gives you an idea of what
I am talking about. A brace is a piece of wood placed in the corner to
create a triangle.

I suspect our problem is I am speaking English in America and you are
speaking in English in the UK.

[dpb]

On 5/10/2013 3:07 PM, Alexander Lamaison wrote:
...

> I'm not trying to build anything as engineering-critical as a building
> ... yet. Just every day items: log store, yard gate, kitchen cabinet,
> TV table, wash stand, window frame etc.

Well, it's been experienced carpenters and the like that take the kinds
of work for the more and scale it down to the lesser...I've certainly
seen a number of DIY'er type books on racks at the places like the BORGS
and all in the States--I'd presume they're there, too.

> I'm not aware of a "conventional set of dimensions" so maybe that's
> where I'm struggling. I didn't even realise there was such a thing.
> Perhaps you could enlighten me. Let's say I wanted to build a simple
> table to hold a small television; how would you size the pieces that
> make this up?

...

Well, virtually everything in the US for face frames, rails, things like
table aprons, shelf uprights and shelves, etc., etc., etc., are nominal
3/4" stock w/ widths generally adjusted for appearance on most small
stuff. Larger spans/heavier loads means wider aprons to provide
additional bending moment/load capability. That's about it for
structural design.

After that for joints, furniture is generally a combination of M&T and
possibly some dowels for traditional craftsmanship; commercial stuff
uses many shortcuts or machine-cut joints for efficiency and
cost-savings. OTOH, fences and the like generally are just
nailed/screwed together w/ perhaps a lap or slip tenon or something
again pretty simple.

Again in the US standard lumber sizes are nominal 1x, 2x, 4x and
occasionally 6x timbers for structural members like deck legs and the
like. These are, as said, nominal rough/green dimensions whereas the
actual stock as sold is 3/4, 1-1/2, 3-1/2, etc., ... Also, widths of
construction lumber in the use are nominal less 1/2" pretty consistent.

Ply used to be all english units as well but now has unfortunately
succumbed to the metric of being 19 mm (I believe it is otomh) which is
just under 3/4" enough to be a real pita in thickness compared to actual
lumber. It's still 4x8 sheet size, however, owing to the universal stud
spacings of 16" for 2x4 or 24" on much 2x6 now so that at least sheet
lengths and stud spacings still work.

I am sure there are standards similar in dimension but w/ metric numbers
over there as well. It doesn't take much thought to realize that when
one can by 1x4 or 1x6 off the shelf that 99%+ of projects will use one
or the other but only a very pricey custom piece of work will actually
specify 1x5 because it is exactly what a load calc says is need or the
aesthetics are pleasing since it would require buying the 1x6 anyway and
then onsite cut to fit.

Hardwoods or clear pine/fir/etc. for cabinet work or furniture are _not_
sold in prefinished sizes like construction lumber, however, but as
roughsawn thickness (in 1/4" increments generally) and random
width/length and priced on a board-ft basis. Some retail outlets will
sell dimensioned stock but it will come at a very premium price markup
as compared to market prices.

As for the Frid books, indeed Tage does talk a lot about using what when
altho it's _not_ home repair or yard projects or fencing he's talking
about--he was a furniture maker/highly skilled cabinet maker and that's
the audience he's after, not the DIY'er homeowner.

If that's your target, you might look at the Taunton site again, but at
the Fine Homebuilding site instead of Fine Woodworking. I can't make
recommendations; I've not had interest in the genre so don't know about
the selection.

There are also the trade journals and such like Journal of Light
Construction (or very similar title) and there are, of course, trade
school and college design texts altho my collection there is so dated
that I'm sure none are still in print so won't name any even though the
content is still valid for the most part for what an individual would
find useful. What's missing in them are all the new engineered
materials such as glulam, engineered trusses, joining systems, etc.,
etc., etc., ...

I can't help much on that front other than the generalities--I'm another
who just picked it up from having grown up w/ such things so it just
seems second nature. Doesn't hurt I suppose that am also engineer by
training so structural and mechanics are sorta' routine even though I
was a NucE/Physics guy primarily still had to have the rudiments of
statics/dynamics/strength of materials, etc., ...

--

[Alexander Lamaison]

dpb <no...@non.net> writes:

> On 5/10/2013 3:07 PM, Alexander Lamaison wrote:
> ...
>
>> I'm not trying to build anything as engineering-critical as a building
>> ... yet. Just every day items: log store, yard gate, kitchen cabinet,
>> TV table, wash stand, window frame etc.
>
> Well, it's been experienced carpenters and the like that take the
> kinds of work for the more and scale it down to the lesser...I've
> certainly seen a number of DIY'er type books on racks at the places
> like the BORGS and all in the States--I'd presume they're there, too.
>
>> I'm not aware of a "conventional set of dimensions" so maybe that's
>> where I'm struggling. I didn't even realise there was such a thing.
>> Perhaps you could enlighten me. Let's say I wanted to build a simple
>> table to hold a small television; how would you size the pieces that
>> make this up?
> ...
>
> Well, virtually everything in the US for face frames, rails, things
> like table aprons, shelf uprights and shelves, etc., etc., etc., are
> nominal 3/4" stock w/ widths generally adjusted for appearance on most
> small stuff. Larger spans/heavier loads means wider aprons to provide
> additional bending moment/load capability. That's about it for
> structural design.
>

snip

Thanks. All very useful information.

[Alexander Lamaison]

Actually, we call them the same thing. It's nice to have explained for
once what lap joints are good fore. The books tend to skip that bit.

I'm familiar with braces, but I've never know when they are necessary
and when they aren't. Once you start adding braces in one place, its
tempting to add a few in others and hard to know when to stop.

[Swingman]

Alexander Lamaison <aw...@doc.ic.ac.uk> wrote:
> Are there any basic rules governing the design of a wooden structure?
>
> I'm trying to improve my wood working skills, and my next project is a
> log store for the garden. My design so far is here [1], here [2] and
> here [3].
>
> Usually I decide on the structure by vague 'intuition' about weight and
> load, but I'd much rather apply some tried and tested strategy. For
> instance, how do you decide on the relative size of the members? And
> which members transmit load to others?
>
> Any advice is greatly appreciated.

There is no one path to learning "design", but there is one sure fire
method to get you started on your own designs for just about anything you
want to build out of wood:

Stand on the shoulders of those who came before by carefully researching
the item you wish to build, then incorporate appealing elements and
combinations of form and function, gleaned from your research, into your
own design.

This can be as simple as clipping ideas from magazines, or using technology
like Google searches using the "image" feature and/or websites like Houzz.

Then either sketch, or use a 3D modeling program like SketchUp (with which
you already appear to have some proficiency), to finalize your design,
paying particular attention to incorporating appropriate, traditional
joinery methods that have stood the test of time

The more research you do, the larger your store of design ideas becomes,
for both current and future projects.

--
www.ewoodshop.com (Mobile)

[woodchucker]

I would like to chime in here.
I don't think the structure will be strong enough. I think the boards
for the shelves are not oriented correctly on the bottom. The top has a
skirt that helps but all boards in my estimation should be set on edge
to hold the heavy weight of the logs. That way they won't sag.

Your corner and middle supports can be downsized to conventional stud
lumber. They are vertical and the stresses are less than you think. It's
the shelves that need the most.

I would also put some plywood gussets in the top corners to strengthen
the OPEN structure. tie them into the vertical and horizonal supports
and it will not rack.



--
Jeff

[dpb]

On 5/11/2013 3:05 PM, woodchucker wrote:
> On 5/10/2013 2:24 PM, Alexander Lamaison wrote:

...

>>>> Usually I decide on the structure by vague 'intuition' about weight and
>>>> load, but I'd much rather apply some tried and tested strategy. For
>>>> instance, how do you decide on the relative size of the members? And
>>>> which members transmit load to others?
>>>>
>>>> Any advice is greatly appreciated.
>>>>

..

> I would like to chime in here.
> I don't think the structure will be strong enough. I think the boards
> for the shelves are not oriented correctly on the bottom. The top has a
> skirt that helps but all boards in my estimation should be set on edge
> to hold the heavy weight of the logs. That way they won't sag.
>
> Your corner and middle supports can be downsized to conventional stud
> lumber. They are vertical and the stresses are less than you think. It's
> the shelves that need the most.
>
> I would also put some plywood gussets in the top corners to strengthen
> the OPEN structure. tie them into the vertical and horizonal supports
> and it will not rack.

I hadn't actually looked at the links...I tend to agree. Being US, I
look at the sizes and see roughly 5' width w/ only a couple of supports
in the center at front and rear.

The drawings don't show proposed dimensions of the material for the
shelves; agree doesn't look like it's more than perhaps what would be 1x
stock here which would have finished thickness of 3/4". That'll
certainly sag w/ time on a 30" span even though it'll likely hold as
he's only got 30" maximum shown for vertical opening at the front that
isn't going to hold a whole lot of round firewood--it just won't pack
that densely in the round.

But, lets see what we might get--doesn't have any dimensions on the
structural members as said but the uprights look like might be on the
order of what a tubafor would be in the US so let's subtract 10" from
the overall for interior and half that would be each opening about 25"
also. So let's assume 5" diameter log and straight enough that could
get five on a first row and then alternate 5/4, 5/4, etc. That produces
3 layers of 9 logs in an opening. Ok, now how much is that?

V= 3.14*(5/12)^2 * 2 * 27 ~~> 3*(25/12)*(2/12) * 27
~ 3*~2/6 * 27 --> 1 * 27 ==> 27 cuft/opening

If assume oak at roughly 45 lb/cu-ft, that's 1200 lb on each of those
shelves which translates to an average loading of ~135 psf. That's
pretty healthy load; more than I'd have thought. Of course, it's not
likely that the actual firewood will be so accomodating as to fit so
well, but it is a point...

Now one can go look at deflection tables and so on and make some choices
on sizes.

Just looking at a minimum I'd put a solid piece of 3/4 ply or the like
in the middle between the two supports plus their definitely will need
to be support across the width of the openings under the shelves.

--

[Alexander Lamaison]

dpb <no...@non.net> writes:

> On 5/11/2013 3:05 PM, woodchucker wrote:
>> On 5/10/2013 2:24 PM, Alexander Lamaison wrote:
> ...
>
>>>>> Usually I decide on the structure by vague 'intuition' about weight and
>>>>> load, but I'd much rather apply some tried and tested strategy. For
>>>>> instance, how do you decide on the relative size of the members? And
>>>>> which members transmit load to others?
>>>>>
>>>>> Any advice is greatly appreciated.
>>>>>
> ..
>
>> I would like to chime in here.
>> I don't think the structure will be strong enough. I think the boards
>> for the shelves are not oriented correctly on the bottom. The top has a
>> skirt that helps but all boards in my estimation should be set on edge
>> to hold the heavy weight of the logs. That way they won't sag.

You mean flip the shelves 90 degree to make them stand like joists
rather than floor boards? That would make them strong and let more air
in (both good). How should I join them to the supporting beams without
losing the strength of having them run accross the top, because now
they'll be a bit thick to drive a screw through the top of them, even if
I reduce the vertical dimension a bit?

Or did you mean something else?

>> Your corner and middle supports can be downsized to conventional stud
>> lumber. They are vertical and the stresses are less than you think. It's
>> the shelves that need the most.

Nice. That more than halves their cost.

>> I would also put some plywood gussets in the top corners to strengthen
>> the OPEN structure. tie them into the vertical and horizonal supports
>> and it will not rack.
>
> I hadn't actually looked at the links...I tend to agree. Being US, I
> look at the sizes and see roughly 5' width w/ only a couple of
> supports in the center at front and rear.
>
> The drawings don't show proposed dimensions of the material for the
> shelves; agree doesn't look like it's more than perhaps what would be
> 1x stock here which would have finished thickness of 3/4". That'll
> certainly sag w/ time on a 30" span even though it'll likely hold as
> he's only got 30" maximum shown for vertical opening at the front that
> isn't going to hold a whole lot of round firewood--it just won't pack
> that densely in the round.

Yes, sorry, should have mentioned the dimensions. The shelves are 25 x
100 finished size (1" x 4" ish).

> But, lets see what we might get--doesn't have any dimensions on the
> structural members as said but the uprights look like might be on the
> order of what a tubafor would be in the US so let's subtract 10" from
> the overall for interior and half that would be each opening about 25"
> also. So let's assume 5" diameter log and straight enough that could
> get five on a first row and then alternate 5/4, 5/4, etc. That
> produces 3 layers of 9 logs in an opening. Ok, now how much is that?
>
> V= 3.14*(5/12)^2 * 2 * 27 ~~> 3*(25/12)*(2/12) * 27
> ~ 3*~2/6 * 27 --> 1 * 27 ==> 27 cuft/opening
>
> If assume oak at roughly 45 lb/cu-ft, that's 1200 lb on each of those
> shelves which translates to an average loading of ~135 psf. That's
> pretty healthy load; more than I'd have thought. Of course, it's not
> likely that the actual firewood will be so accomodating as to fit so
> well, but it is a point...

Now this is getting fun :) Some engineering.

> Now one can go look at deflection tables and so on and make some
> choices on sizes.

I've not come across deflection tables before. Do you have an online
reference to the one you're using? The ones I'm finding online are all
for large structural timbers for house building and don't cover the
kinds of pieces we're talking about here.

> Just looking at a minimum I'd put a solid piece of 3/4 ply or the like
> in the middle between the two supports

I'm not sure where you mean. Across the shelves, knitting the 5 pieces
together like another support beam?

> plus their definitely will need to be support across the width of the
> openings under the shelves.

I was avoiding this in order to allow air underneath. Perhaps I can
compromise using a smaller piece that both provides support and leaves
an air gap.


Thanks for all the advice guys. It really is apppreciated.

[dpb]

On 5/12/2013 6:31 AM, Alexander Lamaison wrote:
> dpb<no...@non.net> writes:
>> On 5/11/2013 3:05 PM, woodchucker wrote:

...

>>
>>> I would like to chime in here.
>>> I don't think the structure will be strong enough. I think the boards
>>> for the shelves are not oriented correctly on the bottom. The top has a
>>> skirt that helps but all boards in my estimation should be set on edge
>>> to hold the heavy weight of the logs. That way they won't sag.
>
> You mean flip the shelves 90 degree to make them stand like joists
> rather than floor boards? That would make them strong and let more air
> in (both good). How should I join them to the supporting beams without
> losing the strength of having them run accross the top, because now
> they'll be a bit thick to drive a screw through the top of them, even if
> I reduce the vertical dimension a bit?
>
> Or did you mean something else?

Yes, he means run as joists. That's certainly one way altho it will cut
down the opening size significantly unless you raise overall height.

You could support them simplest by adding a ledger board on the ends to
rest them on and only w/ a little more effort notch them to hide it.

>>> Your corner and middle supports can be downsized to conventional stud
>>> lumber. They are vertical and the stresses are less than you think. It's
>>> the shelves that need the most.
>
> Nice. That more than halves their cost.

Could even go to two 1x joined in tee and for the vertical will still be
plenty stout-enough...

...

> Yes, sorry, should have mentioned the dimensions. The shelves are 25 x
> 100 finished size (1" x 4" ish).
>

...

>>
>> If assume oak at roughly 45 lb/cu-ft, that's 1200 lb on each of those
>> shelves which translates to an average loading of ~135 psf. That's
>> pretty healthy load; more than I'd have thought. Of course, it's not
>> likely that the actual firewood will be so accomodating as to fit so
>> well, but it is a point...
>
> Now this is getting fun :) Some engineering.
>
>> Now one can go look at deflection tables and so on and make some
>> choices on sizes.
>
> I've not come across deflection tables before. Do you have an online
> reference to the one you're using? The ones I'm finding online are all
> for large structural timbers for house building and don't cover the
> kinds of pieces we're talking about here.

Well, all you have to do is set the dimensions correctly and use
appropriate values for the various material and geometric properties...

I use the beam calculator quite a lot...but for stuff like you're doing
the "sagulator" is probably the easiest tool as it hides a lot of the
complexity by making all the assumptions about material properties and
assumes rectangular pieces...

<http://www.woodbin.com/calcs/sagulator.htm>

>> Just looking at a minimum I'd put a solid piece of 3/4 ply or the like
>> in the middle between the two supports
>
> I'm not sure where you mean. Across the shelves, knitting the 5 pieces
> together like another support beam?

Well, if I read your drawing correctly, it's open the full width w/ the
exception of there being a vertical support at the front and rear in the
middle but no support in the middle internally. Since there would be a
full 5-ft span w/o those that's obviously not possible/reasonable to
consider removing it; it will prevent any use except of the two "bins"
so you're not losing anything by putting a divider between them for
vertical support across the full width. Ply would be the easiest way to
accomplish that; solid verticals would also work. You could either use
the full length shelving and fit the verticals or use solid one-piece
verticals and separate shelves--your choice, same result.

>> plus their definitely will need to be support across the width of the
>> openings under the shelves.
>
> I was avoiding this in order to allow air underneath. Perhaps I can
> compromise using a smaller piece that both provides support and leaves
> an air gap.

...

One way would be an on-edge "X" from corner to corner fastened to the
bottom of the shelves w/ adhesive and screws to "stiffen-up" the shelves
just as does an edging or a table apron.

If I take your design to the sagulator I get (using one of the pines
that would be typical lumberyard material here; pick a species that is
somewhat like what you would have for your material obviously)

Shelf Characteristics
Shelf Material
Shelf attachment Fixed X Floating
Shelf load 200 per foot
Load units lbs
Load distribution X Uniform load Center load
Build shelves with less waste
Shelf span 30 in
Depth (front to back)
Thickness
[Optional] Edging Strip
Material None
Width See note # 10
Thickness
New Apply WoodBin lab correction? yes x no

Sag total 0.27 unit
0.107 in per foot
Target sag: 0.02 in per foot

Just for comparison I took to the

<http://www.engineeringtoolbox.com/beam-stress-deflection-d_1312.html>

beam calculation engine I use quite a lot and while it takes a little
more effort, I'm always comfortable in knowing what actually happens...

For it for the same assumptions

Imperial Units

16.666 q - Load (lb/in)

30 L - Length of Beam (in)

.333 I - Moment of Inertia (in4)

2x10^6 E - Modulus of Elasticity (psi)

0.5 y - Perpendicular distance from to neutral axis X (in)

Unit Load - q : 16.7 (lb/in)
Total Load : 500 (lb)
Length of Beam - L : 30 (in)
Moment of Inertia - I : 0.33 (in4)
Modulus of Elasticity - E : 2000000 (psi)
Perp. distance from neutral axis - y : 0.5 (in)
Support Force - R1 : 250 (lb)
Support Force - R2 : 250 (lb)
Maximum Stress - : 2815 (psi)
Maximum Deflection - : 0.26 (in)

I get a max deflection of 0.26 instead of 0.27 -- pretty doggone good
agreement.

Now, that's based on the previously estimated 1200 lb total load divided
out to the average uniform load on a 1x4 laying flat. You can see the
difference if you turn it around on edge or change various other
dimensions and or loadings, etc., etc., etc., ...

As I had presumed initially from just gut feelings, the sag would be
noticeable but a 1x would likely be able to hold the load w/o actually
breaking but it's well under-sized that way.

BTW, the I for a rectangular section is bh^3/12 where b=base and
h=height. In English units it generally has units therefore of in^4.
For your 1x4 flat that gives 4x1^3/12 = 1/3 in^4. You can see why on
edge helps so much if you turn those dimensions around then it is
1x4^3/12 = 16/3 = 5.333. That's 4^3/4 = 4^2 = 16X times the stiffness
for only 4x the thickness.

Note typical E values for wood are roughly 1/10th that of common steel.

--

[woodchucker]

On 5/12/2013 7:31 AM, Alexander Lamaison wrote:
> dpb <no...@non.net> writes:
>
>> On 5/11/2013 3:05 PM, woodchucker wrote:
>>> On 5/10/2013 2:24 PM, Alexander Lamaison wrote:
>> ...
>>
>>>>>> Usually I decide on the structure by vague 'intuition' about weight and
>>>>>> load, but I'd much rather apply some tried and tested strategy. For
>>>>>> instance, how do you decide on the relative size of the members? And
>>>>>> which members transmit load to others?
>>>>>>
>>>>>> Any advice is greatly appreciated.
>>>>>>
>> ..
>>
>>> I would like to chime in here.
>>> I don't think the structure will be strong enough. I think the boards
>>> for the shelves are not oriented correctly on the bottom. The top has a
>>> skirt that helps but all boards in my estimation should be set on edge
>>> to hold the heavy weight of the logs. That way they won't sag.
>
> You mean flip the shelves 90 degree to make them stand like joists
> rather than floor boards? That would make them strong and let more air
> in (both good). How should I join them to the supporting beams without
> losing the strength of having them run accross the top, because now
> they'll be a bit thick to drive a screw through the top of them, even if
> I reduce the vertical dimension a bit?

So what you might want to do is use wood that is capable of notching the
so the shelves would be notched into the support. Now you can toenail
(nail at an angle) into the notched supports, or just notch and drive
the screw from the top into the support. Do this with a pre-drilled hole
half way throug the shelf so only half of the screw is in the shelf...
either way will be fine the notching will keep them on edge, the screw
or toe nailing will keep it tight... you will need less piece for the
shelf so don't make 8 notches, 4 might be enough, 3 maybe.. I don't
know.. I'm not concentrating on the dimensions.

>
> Or did you mean something else?
>
>>> Your corner and middle supports can be downsized to conventional stud
>>> lumber. They are vertical and the stresses are less than you think. It's
>>> the shelves that need the most.
>
> Nice. That more than halves their cost.

I have a lumber rack for my woodworking. Made of 2x4 supports and it
holds an unbelievable amount of wood thousands of pounds. People under
estimate vertical support. Yes a 1x2 or 1x2 would not do it, but a 2x4
would SAE measurements.. A 4x4 is overkill which I assume is close to
what you showed.

Jeff

[dpb]

On 5/12/2013 10:17 AM, woodchucker wrote:
...

> I have a lumber rack for my woodworking. Made of 2x4 supports and it
> holds an unbelievable amount of wood thousands of pounds. People under
> estimate vertical support. Yes a 1x2 or 1x2 would not do it, but a 2x4
> would SAE measurements.. A 4x4 is overkill which I assume is close to
> what you showed.

...

So would two 1x in 'L' configuration.

The key is maintaining verticality and short enough span (height) so
buckling doesn't enter into it.

--

[dpb]

On 5/12/2013 10:15 AM, dpb wrote:
...

> Well, if I read your drawing correctly, it's open the full width w/ the
> exception of there being a vertical support at the front and rear in the
> middle but no support in the middle internally. Since there would be a
> full 5-ft span w/o those that's obviously not possible/reasonable to
> consider removing it; it will prevent any use except of the two "bins"
> so you're not losing anything by putting a divider between them for
> vertical support across the full width. Ply would be the easiest way to
> accomplish that; solid verticals would also work. You could either use
> the full length shelving and fit the verticals or use solid one-piece
> verticals and separate shelves--your choice, same result.

...

Or, you could simply use a horizontal support fastened to the two
uprights w/ "enough beef" to support the load--tubasix would do nicely
and still be thinner than it appears your verticals are now...

I just went w/ the solid center support (or you could replicate the ends
in the middle as well) as it's simple and adds both vertical support as
well as racking by forming a truss member that is pretty much lacking
anywhere else at all (as others have also noted).

--

[dpb]

On 5/11/2013 3:58 PM, dpb wrote:
...

> ... So let's assume 5" diameter log and straight enough that could get

> five on a first row and then alternate 5/4, 5/4, etc. That produces 3
> layers of 9 logs in an opening. Ok, now how much is that?
>
> V= 3.14*(5/12)^2 * 2 * 27 ~~> 3*(25/12)*(2/12) * 27
> ~ 3*~2/6 * 27 --> 1 * 27 ==> 27 cuft/opening
>
> If assume oak at roughly 45 lb/cu-ft, that's 1200 lb on each of those
> shelves which translates to an average loading of ~135 psf. That's

> pretty healthy load; more than I'd have thought....

Of course it's more than I thought/expected--forgot the divisor of 4 for
radius. So the load/bin is more like 300 lb instead of 1200. That's a
lot more in line w/ what gut feeling was; don't know why it didn't
register before now...

V= (3.14/4)*(5/12)^2 *2*27 ~ 3/4*~2/6*27 --> 27/4 ==> ~7 cuft/opening

So, the deflections below are going to be ~1/4th those numbers which is
getting marginally acceptable; if went w/ tubafor instead of 1X probably
be fine w/ only slight modifications. Still this assumes the solid or
near solid center support, though, of course.

--

[Alexander Lamaison]

dpb <no...@non.net> writes:

> On 5/12/2013 6:31 AM, Alexander Lamaison wrote:
>> dpb<no...@non.net> writes:
>>> On 5/11/2013 3:05 PM, woodchucker wrote:
> ...
>

> I use the beam calculator quite a lot...but for stuff like you're
> doing the "sagulator" is probably the easiest tool as it hides a lot
> of the complexity by making all the assumptions about material
> properties and assumes rectangular pieces...
>
> <http://www.woodbin.com/calcs/sagulator.htm>
>
>>> Just looking at a minimum I'd put a solid piece of 3/4 ply or the like
>>> in the middle between the two supports
>>
>> I'm not sure where you mean. Across the shelves, knitting the 5 pieces
>> together like another support beam?
>
> Well, if I read your drawing correctly, it's open the full width w/
> the exception of there being a vertical support at the front and rear
> in the middle but no support in the middle internally. Since there
> would be a full 5-ft span w/o those that's obviously not
> possible/reasonable to consider removing it; it will prevent any use
> except of the two "bins" so you're not losing anything by putting a
> divider between them for vertical support across the full width. Ply
> would be the easiest way to accomplish that; solid verticals would
> also work. You could either use the full length shelving and fit the
> verticals or use solid one-piece verticals and separate shelves--your
> choice, same result.

In the existing design there were also two beams in the middle running
between the front and back verticals, under each set of shelves. The
extra support in the middle that you suggest, is that to provide extra
support for the shelves, or to help keep the frame square?

As it happens, I was planning to tack some thing slats accross the
middle anyway, just so that I could use each side of the bin
independently. But I didn't include that on the diagram as it wasn't
meant to be structural.

Wow, thanks for leading me through all of that. It's really helpful.

I'll adjust my design accordingly and post links again when I'm done.

Thanks,

[Alexander Lamaison]

Ok, new words.

Truss:

from a quick Wikipediaing, am I right in understanding that, to get a
stable structure, you should join the planes together? And this is
called a truss. The brace and gusset previously mentioned, as well as
your suggested solid support, do this by joining two planes together.
Do this in both directions, and you have a truss?

Racking:

Googling was no use here. It just tried to sell me shelves. What is
racking?

Thanks,

[Alexander Lamaison]

I had to read this 3 or 4 times to work out why this version was
correct. Thanks for the update :)

[no...@none.com]

On Tue, 14 May 2013 09:40:08 +0100, Alexander Lamaison

>Googling was no use here. It just tried to sell me shelves. What is
>racking?

Racking is when some type of shelving structure starts to lean over in
a particular direction and then usually collapses. It often ends up in
a cascade failure.
http://www.youtube.com/watch?v=fIP24ywBAdU

[Alexander Lamaison]

Haha, that's spectacular. :D
Thanks

[dpb]

On 5/14/2013 3:40 AM, Alexander Lamaison wrote:
...

> Truss:
>
> from a quick Wikipediaing, am I right in understanding that, to get a
> stable structure, you should join the planes together? And this is
> called a truss. The brace and gusset previously mentioned, as well as
> your suggested solid support, do this by joining two planes together.
> Do this in both directions, and you have a truss?

...

Strictly speaking, a truss is a coplanar system of structural members
joined together at their ends to form a stable framework. The simplest
form of truss is three members in a triangular shape--clearly you can
see that pushing on any given corner is resisted by the compression of
the two opposing sides and if the material is ideal it's perfectly rigid
to loads in plane.

The suggested solid piece serves the function that a strictly truss-like
mechanism would of an 'x' but adds the needed vertical support the
shelves need in the middle so while not strictly a truss in the simple
definition by being solid it provides the resistance to serve the
purpose as well as the support.

--

[dpb]

On 5/14/2013 3:26 AM, Alexander Lamaison wrote:
...

> In the existing design there were also two beams in the middle running
> between the front and back verticals, under each set of shelves. The
> extra support in the middle that you suggest, is that to provide extra
> support for the shelves, or to help keep the frame square?

A) Didn't appear obvious to me they were there; but then again, I only
looked at it very briefly to get the overall rough dimensions...

B) Both. See response above to note on 'truss' and 'racking'...

...

> Wow, thanks for leading me through all of that. It's really helpful.

...

I see you did find the erratum posting that loading is much more like
expected when one correctly includes the 1/2 factor in the diameter when
computing volume so I'll retire unless have further specific
questions...I just _knew_ something didn't make sense but kept
overlooking it.

--

[dpb]

On 5/14/2013 8:23 AM, dpb wrote:
...

> The suggested solid piece serves the function that a strictly truss-like
> mechanism would of an 'x' but adds the needed vertical support the
> shelves need in the middle so while not strictly a truss in the simple
> definition by being solid it provides the resistance to serve the
> purpose as well as the support.

And, while repeating the end design in the middle would provide the
vertical support, unless the attachment is very rigid there's little
resistance forward/back to racking by doing so. A set of solid vertical
1x4s if the shelves were mortised into it very tightly would provide
some resistance by the twisting moment of the shelves, but there's not
an overall strong resisting member overall.

OTOH, either a solid plank or the ply has length on all sides for
connection and ends up, therefore, providing that rigidity.

--

[Alexander Lamaison]

Could I achieve the desired result, instead, by repeating the end design
but having the planks run diagonally?

[dpb]

On 5/14/2013 6:16 PM, Alexander Lamaison wrote:
>> On 5/14/2013 8:23 AM, dpb wrote:
...

>> OTOH, either a solid plank or the ply has length on all sides for
>> connection and ends up, therefore, providing that rigidity.
>
> Could I achieve the desired result, instead, by repeating the end design
> but having the planks run diagonally?

Certainly helps. The question then hinges on how they're fastened to be
as near a rigid connection as possible and how that connection will fare
w/ time and use...

--