search  current discussion  categories  kilns & firing - construction 

arches, sprung and otherwise

updated tue 10 feb 98

 

Gavin Stairs on tue 3 feb 98

Hi All,

I've been puzzling over this term "sprung arch". People seem to use it as
a synonym for circular or Roman arch. I've only come across it in the
context of an arched kiln roof. I think I know where it arises, but I'd
like to hear what you think.

In a pottery kiln fired to 1200C or so, the material of the kiln expands
from the heat. At the inner wall of a kiln 1.5m (~5ft) wide, the expansion
is of the order of 10mm (0.4in). At the outer wall, it is much less,
perhaps 1mm (0.04in). This means that the arch has a tendency to expand,
and flatten, while the kiln is being fired. The forces generated in a flat
arch are sufficient to pull apart any restraining structure around the
kiln, including large buttresses. On cooling, the kiln will contract
again, but the buttresses, or any stiff structures that have been forced
apart, will not regain their former position, so the kiln will gradually
collapse, at a rate of 10mm (0.4in) per firing.

The solution is springs. The lateral reaction forces on the arch must be
restrained by compliant (springy) frames which will spread under the force
of the heat expansion, but regain their shape when the kiln cools. With
the right force in the spring, the forces balance, and the kiln remains
stable, even though it breathes in and out with each firing.

If you look at some of the old noborigamas, including the Tozan kilns, you
will see large beams being used as springs in this way. There are tales of
the kiln masters vigorously firing their kiln while the spring beams catch
fire and have to be watered.

There are two current methods, to my knowledge, of arranging these springs
in modern kilns. One is to use the spring of a beam. In this method, the
cross beam that bears against the arch foot is attached to verticals in
mid-span, so that the vertical acts like a spring. The other is to use
valve springs from a car or truck engine, with a stiff cross piece. Valve
springs are good and stiff, and are available at any scrap yard for
pennies. In both cases, the foot beam has to be very stiff, so a large
section angle or channel is indicated. The heavier the better, within reason.

So much for flat, or "sprung", arches. What about deep arches, like
catenaries? In these arches, the whole kiln is arch, and the vertical part
is carries the arch force down more vertically than in a flat arch.
Consequently, the arch expansion can be taken up in a flexure of the arch
itself. If you look very closely at such a kiln during firing, you will
see that the cracks in the kiln plastering will open ever so little. This
is enough to adjust for the expansion.

So much for the roof. What about the floor? Floors also get hot, and
expand and contract. The floor of a kiln will eventually creep wider and
wider, especially if dust and litter fills in the cracks in between
firings. Same thing with the longitudinal dimension of the walls and arch.
If the kiln lining lasts long enough, the kiln may succumb to these
gradual expansions. One could spring the floor as well as the arch, but
in practice, this does not usually happen. Floors apparently recover their
shape better than do arches.

Gavin

Vince Pitelka on fri 6 feb 98

>In a pottery kiln fired to 1200C or so, the material of the kiln expands
>from the heat. At the inner wall of a kiln 1.5m (~5ft) wide, the expansion
>is of the order of 10mm (0.4in). At the outer wall, it is much less,
>perhaps 1mm (0.04in). This means that the arch has a tendency to expand,
>and flatten, while the kiln is being fired. The forces generated in a flat
>arch are sufficient to pull apart any restraining structure around the
>kiln, including large buttresses. On cooling, the kiln will contract
>again, but the buttresses, or any stiff structures that have been forced
>apart, will not regain their former position, so the kiln will gradually
>collapse, at a rate of 10mm (0.4in) per firing.

With all due respect Gavin, this is not the case in an arch with an adequate
steel frame, and with a reasonable arch curvature. I like at least 2" rise
per foot of spand. Certainly the arch expands more on the inside surface,
but this does not make it flatten out. On the contrary, any sturdy steel
framework, without springs, will confine it, and it will rise slightly in
the center. The gaps between the bricks will open up slightly on the
outside of the inner course, as the inside expands (as you stated in regards
to catenary arches). After the firing the arch settles back to where it
started out (USUALLY!). The only real exception I have encountered is in
salt kilns, where the arch tends to flux itself together into a monolithic
mass with time. In that case some very interesting things can happen over
time as the arch rises and settles with successive firings.
- Vince

Vince Pitelka - vpitelka@DeKalb.net
Home 615/597-5376, work 615/597-6801, fax 615/597-6803
Appalachian Center for Crafts
Tennessee Technological University
1560 Craft Center Drive, Smithville TN 37166

Gavin Stairs on sat 7 feb 98

At 09:45 AM 2/6/98 EST, Vince wrote:
>>... This means that the arch has a tendency to expand,
>>and flatten, while the kiln is being fired.
....
>With all due respect Gavin, this is not the case in an arch with an adequate
>steel frame, and with a reasonable arch curvature. I like at least 2" rise
>per foot of spand. Certainly the arch expands more on the inside surface,
>but this does not make it flatten out. On the contrary, any sturdy steel
>framework, without springs, will confine it, and it will rise slightly in
>the center. The gaps between the bricks will open up slightly on the
>outside of the inner course, as the inside expands (as you stated in regards
>to catenary arches). After the firing the arch settles back to where it
>started out (USUALLY!). The only real exception I have encountered is in
>salt kilns, where the arch tends to flux itself together into a monolithic
>mass with time. In that case some very interesting things can happen over
>time as the arch rises and settles with successive firings.
....

Thanks, Vince for this clarification. I also had a message from Gil
Stengel about this.

What I was talking about was the change in shape of the arch owing to
temperature gradients alone. This does indeed expand and flatten the arch.
But no practical arch will permit this to happen; instead, the arch rises
while the ends are kept fixed. This does indeed open up cracks on the
outside, while the inside edge is severely stressed. This may lead to
spalling of the inner edge, similar to shivering in a glaze. This is one
of the mechanisms of failure of refractories in a kiln.

Refractory brick also undergoes heat induced changes over its useful life.
In general, these shrink the material. So inner layer bricks will tend to
shrink on the inner face, over time, which in some degree compensates for
the spalling stress noted above. However, it also sets up internal
stresses in the bricks themselves, which lead to delamination or spalling.
It's a bit of a race to see which mechanism wins in the end. Old kilns
tend to be eroded at the cracks, and sometimes the face spalling is quite
deep, so I guess it's a question of what goes first.

The intention of the springs in the arch foot restraints is to lessen this
tendency of the arch to rise against a firm footing. I mentioned two
methods in current use. Gil pointed out that in actual use, valve springs
are rarely strong enough to provide sufficient force. In practice, he has
observed few (or no) cases where the springs were not pulled up solid,
which means they are not acting as springs. I also observe that many
so-called valve spring restraints are actually built as beam springs, with
the valve springs on a cross-tie well above the arch foot, with a springy
column-beam reaching down to the foot. This may work, but the spring is
the beam, not the valve spring.

I also note that several people referred to the problem of the floor
expansion. Any thermally cycled system like this will have a tendency to
walk or creep, and will tend to expand over time. A suitable restraint
system, be it a buttress or a frame, will counteract this to some extent.
It is important to recognize that steel cannot take up the expansions
simply by stretching: a solidly built (no flexure) restraint system will
eventually break, or will break the kiln bricks. Some kind of expansion
pads or springs are necessary.

Most of the final failure mechanisms are fatigue, creep or progressive
collapse. Fatigue in the steel, or more commonly in the refractory. Creep
or progressive collapse in the gradual expansion of the structure.
Refractory fatigue is of course exacerbated by the thermal gradient across
the brick itself, particularly in IFB.

Fibre removes most of these concerns, while of course raising a few others.
I think that the most durable kiln construction may well be semi-rigid
fiber with a refractory coating, like ITC. The fibre is compressible
enough to take up all these motions , permiting a solid framework. Fibre
will, of course, break down over time, but perhaps not as quickly as
stressed firebrick. I'm intrigued by fast, economical,
fibre/refractory-coat kilns. There is a problem with cooling rates,
perhaps, but that should not be insurmountable. And there is the problem
of fibre dust, which must be dealt with by suitable precautions,
particularly during construction. Particularly for use in cities, I think
such a kiln, electrically heated, with an auxiliary propane reduction
flame, might be an effective solution to many pollution concerns. But we
potter/pyromaniacs are still in love with the old craft, and with mosnters
like wood and salt kilns. And as a practical matter of building a kiln out
of stout, available materials that have a wide acceptance, the hard-brick
or IFB kiln is probably going to be the first choice for many kiln
builders. In which case, an understanding of the problems of expansion is
a first need.

Bottom line: heat it and it will expand. Put solid steel bands around it,
and something will give, sooner or later.

Gavin

Gari Whelon on sat 7 feb 98

Having just bought a house, my plans are for a gas kiln, hopefully a gas
/salt-soda. It's a ways off, but I've been following this thread with
interest, thanks to all for the info and ideas.

Vince, your comment on "interesting things happening" in the rising and
settling of arched salt kilns piques my interest. Is this interesting as in
"may you be cursed to live in interesting times" or something better?

On the flat top vs arch question, what advice/opinion is there specific to
salt/soda firings?

Thanx to all for the great wealth of generously shared knowledge.

Gari Whelon

Nanaimo B.C.

At 09:45 AM 2/6/98 EST, Vince Pitelka wrote:
>----------------------------Original message----------------------------
>>In a pottery kiln fired to 1200C or so, the material of the kiln expands
>>from the heat. At the inner wall of a kiln 1.5m (~5ft) wide, the expansion
>>is of the order of 10mm (0.4in). At the outer wall, it is much less,
>>perhaps 1mm (0.04in). This means that the arch has a tendency to expand,
>>and flatten, while the kiln is being fired. The forces generated in a flat
>>arch are sufficient to pull apart any restraining structure around the
>>kiln, including large buttresses. On cooling, the kiln will contract
>>again, but the buttresses, or any stiff structures that have been forced
>>apart, will not regain their former position, so the kiln will gradually
>>collapse, at a rate of 10mm (0.4in) per firing.
>
>With all due respect Gavin, this is not the case in an arch with an adequate
>steel frame, and with a reasonable arch curvature. I like at least 2" rise
>per foot of spand. Certainly the arch expands more on the inside surface,
>but this does not make it flatten out. On the contrary, any sturdy steel
>framework, without springs, will confine it, and it will rise slightly in
>the center. The gaps between the bricks will open up slightly on the
>outside of the inner course, as the inside expands (as you stated in regards
>to catenary arches). After the firing the arch settles back to where it
>started out (USUALLY!). The only real exception I have encountered is in
>salt kilns, where the arch tends to flux itself together into a monolithic
>mass with time. In that case some very interesting things can happen over
>time as the arch rises and settles with successive firings.
>- Vince
>
>Vince Pitelka - vpitelka@DeKalb.net
>Home 615/597-5376, work 615/597-6801, fax 615/597-6803
>Appalachian Center for Crafts
>Tennessee Technological University
>1560 Craft Center Drive, Smithville TN 37166
>
>

Vince Pitelka on sun 8 feb 98

>Vince, your comment on "interesting things happening" in the rising and
>settling of arched salt kilns piques my interest. Is this interesting as in
>"may you be cursed to live in interesting times" or something better?

Gari -
Life is interesting. The action of salt over time tends to cement the
bricks into a monolithic mass, which tends to shift in large pieces. This
is not necessarily a problem at all, but one must keep a close watch on the
arch in salt/soda kilns. I hope some other people respond, because I am not
the person to give any kind of expert opinion on this. I have always viewed
salt kilns as short-life kilns, but perhaps that is because of the frequency
with which we fire them in the academic environment. The salt kiln we built
last May has been fired forty or fifty times, and already looks old.
- Vince

Vince Pitelka - vpitelka@DeKalb.net
Home 615/597-5376, work 615/597-6801, fax 615/597-6803
Appalachian Center for Crafts
Tennessee Technological University
1560 Craft Center Drive, Smithville TN 37166

Gavin Stairs on mon 9 feb 98

For a picture of hard brick vitrification and flow in a wood kiln, see
on my web
site. This is a picture of the interior of the doge (fire box) of a large,
Tozan style, noborigama kiln at Nanaimo, B.C. At this stage, the cracks
have not been bridged. If the kiln is fired more, the bricks may be
"glazed" together into one mass. In fact, this may have happened in this
kiln by now. This picture was taken about a year ago.

shows the spring
beams for this kiln. Note that the design is empirical: there are two
cross beams, one high and one low. The low beam makes the spring very
stiff. The springs are the vertical beams at each side. The arch foot
rests against wedge stones pressed between the spanning steel angle and the
bricks at the arch foot. This same design can be seen executed in wood in
many old kilns.

Gavin

At 10:37 PM 2/8/98 EST, you wrote:
>----------------------------Original message----------------------------
>>Vince, your comment on "interesting things happening" in the rising and
>>settling of arched salt kilns piques my interest. Is this interesting as in
>>"may you be cursed to live in interesting times" or something better?
>
>Gari -
>Life is interesting. The action of salt over time tends to cement the
>bricks into a monolithic mass, which tends to shift in large pieces. This
>is not necessarily a problem at all, but one must keep a close watch on the
>arch in salt/soda kilns. I hope some other people respond, because I am not
>the person to give any kind of expert opinion on this. I have always viewed
>salt kilns as short-life kilns, but perhaps that is because of the frequency
>with which we fire them in the academic environment. The salt kiln we built
>last May has been fired forty or fifty times, and already looks old.
>- Vince
>
>Vince Pitelka - vpitelka@DeKalb.net
>Home 615/597-5376, work 615/597-6801, fax 615/597-6803
>Appalachian Center for Crafts
>Tennessee Technological University
>1560 Craft Center Drive, Smithville TN 37166
>
>