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 Historical content remains viewable.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

Monitoring electricity consumption of a device?

7 views
Skip to first unread message

Jf...@maui.net

unread,
Sep 13, 2005, 9:43:25 PM9/13/05
to
I'm building what I believe will be a super high efficiency electric
furnace utilizing engineering technology overlooked by studio glass
makers. But I need a method to prove my energy usage so I can present
my findings. Besides thorough documentation of my charge and idle
cycles, I need a meter that will work with a phase-angle SCR. I'm not
familiar with any device that can monitor my usage through an SCR. With
an inductive ammeter, one would need a constant graph since current
draw flucuates wildly from moment to moment. Any usenet folks out there
have any ideas?

Mike Firth

unread,
Sep 13, 2005, 10:26:24 PM9/13/05
to
Go to www.sensors.com and ask the same question.
You will almost certainly have to use a data collection system sampling at
a fairly high rate. I don't believe there exists any simple meter that will
integrate the usage, so you need the digital equivalent of a continuous
chart that will do the integration for you. Of course, the power company's
meter does that, if you want to spend the money and record the readings by
hand.

--
Mike Firth
No more levees
Bury old Orleans
Raise New Orleans up if it is worth saving
--
<Jf...@maui.net> wrote in message
news:1126662205.2...@g47g2000cwa.googlegroups.com...

Moonraker

unread,
Sep 13, 2005, 10:39:20 PM9/13/05
to
So, does this overlooked technology make Ohm's law obsolete? Why can't you
calculate the energy usage? You know the resistance of the heating elements
and the amount of time you need to achieve the temperature rise in the
furnace and the line-in voltage. What else do you need?

<Jf...@maui.net> wrote in message
news:1126662205.2...@g47g2000cwa.googlegroups.com...

Ron Parker

unread,
Sep 13, 2005, 10:19:07 PM9/13/05
to
On 13 Sep 2005 18:43:25 -0700, Jf...@maui.net wrote:
> cycles, I need a meter that will work with a phase-angle SCR. I'm not
> familiar with any device that can monitor my usage through an SCR. With
> an inductive ammeter, one would need a constant graph since current
> draw flucuates wildly from moment to moment. Any usenet folks out there
> have any ideas?

Why won't a regular watthour meter like the electric company uses work?

And if it won't work, aren't you concerned that the electric company will
come after you for using power in a way they can't measure?

nJb

unread,
Sep 14, 2005, 12:22:40 AM9/14/05
to

Unless I'm missing something, Kilowatt Hours over a given time will give
you the bottom line. Buy a KWH meter. If you're correct, it will pay for
itself. IIRC, you are off the grid? If your place was ever on the grid,
"borrow" their meter for a while.

--
Jack

Plonked by Native American

bobo1148atxmissiondotcom


http://photos.yahoo.com/bc/xmissionbobo/

nJb

unread,
Sep 14, 2005, 12:31:01 AM9/14/05
to
Moonraker wrote:

> So, does this overlooked technology make Ohm's law obsolete? Why can't you
> calculate the energy usage? You know the resistance of the heating elements
> and the amount of time you need to achieve the temperature rise in the
> furnace and the line-in voltage. What else do you need?
>

Ohm's law works fine but he's only firing at a fraction of the time once
he's up to temp. This varies by the surrounding conditions. A KWH meter
will tell the real story. My kiln draws 56A when it's on full bore, but
seldom is it doing that.

On most controllers we can see what percentage of full power is being
applied. He could see what percentage each furnace uses to maintain a
given temp (2100F?) at like ambient conditions.

Jf...@maui.net

unread,
Sep 14, 2005, 10:11:10 AM9/14/05
to
A couple things: My controller is always set at 100% output signal;
it's the P.A. scr that varies the output. If you put an ammeter on it,
even while the furnace is at equilibrium and totally thermally stable
and idleing, the current jumps all over the place, and that's normal. I
do think a regular watt hour meter is what's required but like Mike
Firth says I'm not sure if I can get one that will work only on the
furnace circuit and also be compatible with the SCR. Yes of course the
power company meters the usage, and I'm certainly not trying to cheat
them. But it is noteworthy that at my other studio in Hawaii the meter
always had trouble and they never really figured it out.

Jack my house in Hawaii and coldwork shop is off the grid (solar and
bio-diesel) but my former hotshop over there never was. Here in Oregon
I have grid power. Also

The technology I'm using has to do with insulation, not the energy
input. The reason I need to monitor the power is that I don't believe
surface area and heat loss equations are going to provide a good real
world analysis. Plus, I suck at math. I have a way to vary these
insulation qualities while keeping the other parameters (mostly) the
same. So if I can accurately measure usage...

Javahut

unread,
Sep 14, 2005, 12:51:09 PM9/14/05
to
Kilowatt Hour meter, yup, believe it or not, I have one. But I don't know
if it will work for your application.
I was in a building that we shared the electric bill, thought I was getting
screwed over, so I put a sub meter on the pieces of equipment that was
running so I could tell what was "my share".
No longer need it, still have the meter, you would have to get your own hook
ups for it, but I know it was pricey, we can negotiate if you need it,
anyone.

"nJb" <no...@nowhere.com> wrote in message
news:dg889v$5i9$1...@news.xmission.com...

Mike Firth

unread,
Sep 14, 2005, 3:48:49 PM9/14/05
to
Moonraker
Only part of each cycle is being used in a phase shift SCR controller.
http://users.ticnet.com/mikefirth/control.htm#PHASE
If he were using zero crossing trigger SCR or SSR, and they were fired not
very often, then simply adding a small analog clock to the circuit would let
your method work- the number of minutes the clock was on gives a good idea
of the actual power used (% of day) [or a digital timer with electronic
on-off. I have analog, I don't own a digital timer.]

Jfuse
If you are interested in the details, you are going to need to do data
collection on a millisecond by millisecond basis, which is considered slow
in the Sensors world. And you are going to want to feed it into a file so
it can be loaded into Excel for averaging, totalling and graphing.
However, I do not follow what you say about your controller being at 100%
output. Something has to be controlling the Phase Angle SCR and that is
usually called a controller.

--
Mike Firth
No more levees
Bury old Orleans
Raise New Orleans up if it is worth saving
--

"Moonraker" <moon...@NOSPAMbellsouth.net> wrote in message
news:l2MVe.3083$Ve1....@bignews4.bellsouth.net...

Jf...@maui.net

unread,
Sep 14, 2005, 3:53:29 PM9/14/05
to
Sure, how much did it cost? I'll try and find out if it will work with
the scr, or what modifications might be needed. What can you tell me
about brand, specs, etc.

Javahut

unread,
Sep 14, 2005, 5:25:17 PM9/14/05
to
I have had it, un-hooked and on the shelf, for 15 years, looks new and
worked fine when it was attached.

You have to bear with me a bit here, because a friend that is an electrical
engineer hooked it up for me.
I had a 440 air conditioning unit, the other tenant also had one, they were
connected to the same meter, we wanted to know if the bill was being split
according to use or just split 50/50. Enough of that, there were 3 coil
like donuts that encircled the hot wires and produced a reading.

I will attempt to get a model number and details for you, if this address is
your email, I'll send a note.


<Jf...@maui.net> wrote in message
news:1126727609.4...@g14g2000cwa.googlegroups.com...

Jf...@maui.net

unread,
Sep 15, 2005, 10:23:27 AM9/15/05
to
Hi Mike. As Jack pointed out, many of us have controllers that are
sophisticated enough to allow percentage-wise control of the output
signal. This can be used for various purposes, notably in our case, to
reduce watt-loading of elements. For example if your heater draws 60a,
you might set your output signal at 50% so that the elements are not
taxed at their maximum. Simply, a way of current limiting from the
controller. I don't use that feature however, because my scr has a
built in current limiter that can be set.

The analog clock will work with a mechanical relay or a mercury relay.
I would be very skeptical about it working with a burst-fire ssr under
any circumstances, but wtfdik.

I really have to avoid a very complicated system, Mike. If it is as you
say, and a particular type of current meter won't work, I'll just run
the furnace and know it is very efficient but not quantifiable.
Because, I don't build furnaces for a living, and I'm not writing a
dissertation. I have to make glass for a living.

Jf...@maui.net

unread,
Sep 15, 2005, 10:27:19 AM9/15/05
to
PS. Errata in reference to the clock I meant an *SSR or a burst fire
SCR.*

Joe

unread,
Sep 15, 2005, 10:03:28 AM9/15/05
to Jf...@maui.net
This may be of interest to you:
http://www.brandelectronics.com/powermeter.htm
I've not used one, but I had bookmarked the page, as it had gotten pretty
good reviews in another NG I read regularly.

Alternately, find a rundown, abandoned house with a meter still in the box,
and borrow it (just kidding, of course). If it's really old, the power co.
would replace it anyway. Maybe even ask a power co employee if they can get
you an obsolete meter. Sometimes the rural electric coops are easier to
deal with. They are all changing over to the meters that can be read from
the truck via RF.

HTH,
Joe

Mike Firth

unread,
Sep 15, 2005, 12:49:54 PM9/15/05
to
>The analog clock will work with a mechanical relay or a mercury relay.
I would be very skeptical about it working with a burst-fire ssr under
any circumstances, but wtfdik.<
Works fine on my rig. The signal coming off an SSR with zero crossing
(not phase control) is exactly what goes in. If the not too radical
assumption is made that it starts pretty quick at power on and only coasts a
bit on power off, then it records the length of the on-time. My controller
tends to stay on till the controller sees how the power affects the temp (it
"learns") and then at temp, it stays off for many seconds and on for a few
seconds, so the analog clock works. At one time, I stood there with a stop
watch and timed the indicator light and found the watch and the clock were
matched close enough for my purposes.

--
Mike Firth
No more levees
Bury old Orleans
Raise New Orleans up if it is worth saving
--

<Jf...@maui.net> wrote in message
news:1126794207.1...@g43g2000cwa.googlegroups.com...

den...@debrady.com

unread,
Sep 18, 2005, 8:24:13 PM9/18/05
to
If your goal is to develop a kiln that holds the heat more efficiently,
I suspect you'll find only the hobbyists are interested. Production
kilnformers usually want kilns that lose heat faster. Many build kilns
with fibreboard instead of bricks for specifically that reason. The
objective is to get the heat out as quickly as is safely possible so
you can fire the kiln more frequently. Most existing glass kilns lose
heat too slowly. Why would somebody want one that's even slower? The
cost of electricity is too small to be of consequence. My calculations
are it's about 1% of the value of each load.

I'd be happy to pay for 10 times as much electricity if I could cool
the kiln twice as fast.

Moonraker

unread,
Sep 18, 2005, 8:55:55 PM9/18/05
to

<den...@debrady.com> wrote in message
news:1127089453.5...@g43g2000cwa.googlegroups.com...

>
> I'd be happy to pay for 10 times as much electricity if I could cool
> the kiln twice as fast.

Ummmm....that's what fans are made for.


Jf...@maui.net

unread,
Sep 19, 2005, 1:01:17 AM9/19/05
to
This is a glass furnace Dennis. Its a completely different animal than
a kiln. But, In a world of runaway fuel costs even kiln formers should
be looking at ways ways to increase efficiency. There are ways to
increase efficiency vastly in kilnforming. Heating up and cooling down
for a few pieces is why most kilnformers will never make it pay. And
beyond that you'll have to figure out the secret yourself.

By way of background, glass blowers are generally told that furnace
efficiency has a defined limit. You can only insulate so much, beyond
which you start progressing backward in efficiency, due to the
increased radiant surface area of the furnace.

I have always had a problem with that fatalism. It has seemed to me
very uninspired thinking. My new furnace takes advantage of technology
my brother (He's a phd in materials science) and I have discussed over
the years. It has to do with strategizing to limit both conductive and
radiant heat loss, and emmissivity of metal surfaces. Beyond that...

Also, I have been very interested in cogeneration from glass furnaces.
Some of our learned scholars in the glass world believe profit in this
regard violates the second law of thermodynamics; I plan to grow
lucious avocados in Oregon in January AND have a super efficient
furnace--which is proof enough for me. My new furnace will lend itself
to co-generation, when I have time to do it. Or when gas hits six bucks.

Mike Firth

unread,
Sep 19, 2005, 11:07:52 AM9/19/05
to
Well, your reason most people think adding more insulation not cost
effective is wrong. Most people who have done measurements feel that the
gain in efficiency after a certain point (about 11") is not worth the money
spent on the insulation. Why spend $100 on insulation that saves 1% over
the previous savings? And you have to live with a really bulky furnance
with a hole in the side that is really deep to get into the glass.
I have asked the question: Since Ceramic Fiber's special insulation
mechanism basically works at higher temps (according to the makers) and it
works by air entrainment at lower temps, why not use ceramic fiber ($$$) at
inner walls and when you get the inside the wall temp down below glass
melting temp, use cheap fluffy fiberglass insulation?
And Henry Havens has mentioned publicly online that he uses a stainless
steel tube with holes drilled in it and low pressure air to increase the
evenness of heat in a kiln during annealling but if too much air is put in,
it cools the kiln, which is sometimes desireable.

--
Mike Firth
No more levees
Bury old Orleans
Raise New Orleans up if it is worth saving
--

<Jf...@maui.net> wrote in message
news:1127106077....@g47g2000cwa.googlegroups.com...

Jf...@maui.net

unread,
Sep 19, 2005, 11:34:06 PM9/19/05
to
Right Mike,,, Henry Havens? You're drunk, right?

At a given thickness (generally agreed to be 7-8" by longtime furnace
builders and engineers) Conductive heat loss through packed fiber is
not reduced in proportion to the radiant heat loss from the surface.
Efficiency proceeds in reverse. That's the theoretical point.The
practical point is, yes at some point additional fiber cost begins to
outweigh energy savings. But even if fiber were free, it still wouldn't
make sense. Remember that fiber is an insulator, but it also conducts
heat energy. In general conductive heat transfer is the most efficient
form of heat transfer. But don't get the wrong idea, obviously the
sheer volume of radiant heat loss per unit area of surface surpasses
the conductive heat loss of an insulating material, at some minimum
level. Ask a guy with an under-insulated furnace. So, you see we aren't
dealing with absolutes, but defined equations. Too much either way, and
efficiency is reduced.

My point is, the fact that you can't insulate further with a conductive
material is not the end of the story. Efficiency can and has been
increased by engineering techniques that are used outside the studio
glass arena.The trick is to limit heat transfer via conduction, while
also limiting the corresponding increase in radiant heat loss per unit
area. A hint, air is one of our best insulators, and it is free. It
insulates well because of its low heat conductivity. However, it is a
great medium for radiant heat loss. Which also fiber is not... and so
on, and so on...

As a side note, efficiency can also be increased by investing some
additional energy in stopping loss from the system.So that, while the
overall energy used by the system is increased, the proportion of
energy within the system put to work, not lost to the surroundings, has
increased. This activity can take different forms.

nJb

unread,
Sep 20, 2005, 12:05:41 AM9/20/05
to
Mike Firth wrote:

> Well, your reason most people think adding more insulation not cost
> effective is wrong. Most people who have done measurements feel that the
> gain in efficiency after a certain point (about 11") is not worth the money
> spent on the insulation. Why spend $100 on insulation that saves 1% over
> the previous savings? And you have to live with a really bulky furnance
> with a hole in the side that is really deep to get into the glass.
> I have asked the question: Since Ceramic Fiber's special insulation
> mechanism basically works at higher temps (according to the makers) and it
> works by air entrainment at lower temps, why not use ceramic fiber ($$$) at
> inner walls and when you get the inside the wall temp down below glass
> melting temp, use cheap fluffy fiberglass insulation?
> And Henry Havens has mentioned publicly online that he uses a stainless
> steel tube with holes drilled in it and low pressure air to increase the
> evenness of heat in a kiln during annealling but if too much air is put in,
> it cools the kiln, which is sometimes desireable.
>

That's Henry Halem. Henry Havens was that guy that sang at Woodstock. ;)

Jf...@maui.net

unread,
Sep 20, 2005, 12:47:22 AM9/20/05
to
Right Henry H. did have an idea (or reported an idea) to achieve very
even heating by using compressed air through stainless tubes to set up
convective currents in a fusing/casting kiln. Henry is also not a kiln
former. In any case, the idea was not to cool down the kiln, but to
achieve more even heating.

The problem with that idea was the bane of every kilnworker producing
gallery quality work--crap landing on your work. Even a slight breeze
can stir up refractory dust from the primer or brick and shitify the
piece. Also, convective currents already exist in a kiln naturally when
there is an air gap under the shelf. Uneven heating is more a problem
when firing fast or when the elements or kiln height are configured
wrong.

Joe

unread,
Sep 20, 2005, 9:52:20 AM9/20/05
to

Nah, that was Ritchie Havens. I believe he was the first performer of the whole
weekend.

Joe

nJb

unread,
Sep 20, 2005, 7:02:14 PM9/20/05
to
Jf...@maui.net wrote:

The breeze isn't a problem as it would only be used during annealing.
IIRC, it was a proposal to correct annealing problems for large kilns
and large pieces.

Jf...@maui.net

unread,
Sep 20, 2005, 9:20:39 PM9/20/05
to
Good point. I hadn't even considered that. It would be a good idea for
annealing. I thought he had mentioned it for heat processes. It seems
like we discussed it at one point and he didn't mention it was for
annealing.

nJb

unread,
Sep 21, 2005, 12:03:17 AM9/21/05
to
Joe wrote:
>
> Nah, that was Ritchie Havens. I believe he was the first performer of the whole
> weekend.
>
> Joe

You're thinking of Richie Valens, the guy that was killed in the skiing
accident with Sonny Holly. Richie Havens is a gated community in Florida.

nJb

unread,
Sep 21, 2005, 12:29:09 AM9/21/05
to
Jf...@maui.net wrote:


I hadn't either until you mentioned it today. It was only after reading
your post that I put the annealing connection to it. I was concerned
about the dust also but it dawned on me that in kiln forming, evenivity
(that's what they named it) would only be important during annealing
larger pieces and maybe certain slumps. All of this is of no concern for
the firing of small pieces but could be a real pain in the ass for large
slabs.

I think Henry lurks here and can point out if I have this all wrong.

--
Jack

bobo1148atxmissiondotcom


http://photos.yahoo.com/bc/xmissionbobo/

Jf...@maui.net

unread,
Sep 21, 2005, 6:58:40 AM9/21/05
to
"Evenivity" is a Blanthornism. Brian Blanthorn of the Isle of Wight, he
of pebbles and a pioneer in devitrification as design. It actually
refers to even firing in a bunch of different ways. Because,
kilnformers are very concerned with uniform temperature across the
whole range of process temps, not only annealing. Uneven temperature at
fusing and slump temps can cause more problems actually than uneven
temperature at annealing. For example if you have a 30f difference from
the center to the margin of a flat piece, it will still anneal fine.
But that difference at fusing temperatures could present big problems
in the outcome of the piece. Evenivity at fuseing temperatures is
extremely important.

Joe

unread,
Sep 21, 2005, 9:29:25 AM9/21/05
to

nJb wrote:

>
>
> Richie Havens is a gated community in Florida.
>

Nuh-uh, the Gated Community is somewhere near Microsoft HQ.

I've noticed *everybody's* English languageing abilitythings have gone to hell in the
past 5 years. Might be due to the effects of new-kew-luhr radiationism.

Joe

Mike Firth

unread,
Sep 21, 2005, 7:58:48 PM9/21/05
to
Well, at lower temps, one entrains air and keeps it from moving, so you
want a fluffy blanket or other stuff on the outside. If you consider fiber
a conductor, I have to say: relative to what? I hope nobody is considering
packed fiber all the way out. Packed fiber is supposed to reflect heat, as
the specific mechanism, but below the temps that works at (1700F as I
recall) volume of non-convective air will work well.
The pipe was originally suggested by me because the person wanted to cool
the kiln easier. Since kiln work is crash cooled, I assumed the request was
for bringing an extra insulated box down through the annealing range.

--
Mike Firth
No more levees
Bury old Orleans
Raise New Orleans up if it is worth saving
--
<Jf...@maui.net> wrote in message

news:1127187246....@f14g2000cwb.googlegroups.com...

Jf...@maui.net

unread,
Sep 22, 2005, 8:48:09 AM9/22/05
to
It would be misleading to say that fiber is a "conductor". But, as a
solid it does conduct heat energy. In general we can state that the
more dense a material, the better in conducts heat. (Don't ask me if
that goes for all material or what exceptions there are.) I.e. a solid
is more heat conductive than a gas.

However a denser material may be much more effective in stopping
radiant heat loss. Which is why as you (sort of) point out, a furnace
maker packs fiber insulation: Radiant heat loss is the *primary issue*
at higher temperatures. Packed fiber is more effective reflecting heat
energy, and limiting radiant heat loss.

So Mike in short the question for an engineer is, how does one take
advantage of the best qualities of packed fiber (stopping radiant heat
loss) while ammeliorating its undesirable qualities (heat conduction).

Wrapping your furnace in an outer layer of less dense home insulation
is a simple attempt on the right track. But not the best solution.
Because, you are increasing surface area dramatically, while still
using a (far less dense) solid material. As I said, a gas is going to
be the least heat conductive material. So, the least amount of material
thickness and resulting increase in outside surface area. And at lower
temperatures (near the outside of your furnace) there are also things
one can do with emmisivity of materials to combat the tendency of a gas
to pass radiant heat energy.

Moonraker

unread,
Sep 22, 2005, 9:02:40 AM9/22/05
to

<Jf...@maui.net> wrote in message
news:1127393289.1...@f14g2000cwb.googlegroups.com...

So, why wouldn't you want to design a "dead air space" just outside the
denser packed fiber insulation and inside the outer wall of the furnace? A
la a "thermopane" window?


Javahut

unread,
Sep 22, 2005, 9:32:37 AM9/22/05
to

"Moonraker" <moon...@NOSPAMbellsouth.net> wrote in message
news:JWxYe.1417$a1....@bignews4.bellsouth.net...

Heat convection would not allow it to be a "dead" air space, the air would
constantly be moving due to convective forces bringing hot air up and cold
air down. NO insulation value unless the air doesn't do that. When a seal
breaks in a thermal pane, fogging occurs, cold air /hot air exchange....


Moonraker

unread,
Sep 22, 2005, 10:12:16 AM9/22/05
to

"Javahut" <not...@nowhere.net> wrote in message
news:dgubp...@enews2.newsguy.com...
Well, I was talking about "sealing" the dead air space. As long as we are
"engineering", why not? Don't you think your "squiggle" will hold up to
the temperatures? ;>)


Jf...@maui.net

unread,
Sep 22, 2005, 4:51:06 PM9/22/05
to
Right, that's the basic idea.

Mike Firth

unread,
Sep 22, 2005, 9:27:40 PM9/22/05
to
> Wrapping your furnace in an outer layer of less dense home insulation
> is a simple attempt on the right track. But not the best solution.
> Because, you are increasing surface area dramatically, while still
> using a (far less dense) solid material. As I said, a gas is going to
> be the least heat conductive material. So, the least amount of material
> thickness and resulting increase in outside surface area.
Excuse me?
Tiny strands of glass, most of which do not connect through the space and
where they make contact to transfer heat to another we are talking point
contact, a "solid"? While the air, entrained to keep it from going into
convection, is certainly a gas. One can not, at the temperatures just below
the melting point of glass, use a more inexpensive insulator (ie plastic
foam) nor justify the use of more expensive insulating foams (the new glass
foams.)
If a furnace is at 2000F and 3" of insulating castable bring the outside
temperature of a glory hole to 400F when measured with a thermocouple beside
still air, then adding a couple of inches of fiberglass will bring the
interface up to about 700F (reducing the loss through the castable because
the temperature difference is less.) and the loss from 700 to 70F (air temp)
across an insulator with a higher R factor than the castable reduces losses
considerably.
If the furnace/glory hole/etc. is originally 24" inside diameter, then
the OD with 3" castable is 30" and with 3" of fiberglass is 36" OD. If it
is considered a cylinder with a lot of insulation on the ends (for
convenience in calculation) then the area of the wall of the larger is only
19.57% higher (30" OD = 94.25; 36" OD = 113.10 circumference, walls assumed
same height.) You are making quite a reach to say that doubling the
thickness of the insulation with higher R is going to be overwhelmed by a
19.6% increase in area. [And if you treat it as a sphere, going up by R^2,
the increase in area is 44% with more than a doubling of insulation value.]
You weren't cubing numbers were you?

--
Mike Firth
No more levees
Bury old Orleans
Raise New Orleans up if it is worth saving
--
<Jf...@maui.net> wrote in message

news:1127393289.1...@f14g2000cwb.googlegroups.com...

nJb

unread,
Sep 22, 2005, 9:44:54 PM9/22/05
to
Moonraker wrote:

You have a real good idea here if it was cost effective. I have no idea
if it would be. When I was a mere youth. I remember a Thermos (TM) being
referred to as a "vacuum bottle". Little or no air in the dead space to
enable convection or conduction. It works great when dealing In lab
jargon I think they call them a Dewars. So we build a furnace inside a
borosilicate dewars. If it's not cost effective now, just wait a few years.

Jf...@maui.net

unread,
Sep 22, 2005, 10:16:18 PM9/22/05
to
Mike your way of writing is very hard for me to follow. Plus, I'm not
inclined to do any math checking unless my nuts are in a vice. Maybe if
you could just, start with a simple thesis or topic statement, and then
bring out the math arsenal. Then I'd know at least where you are trying
to go.

It does seem that you have been arguing some kind of simplification of
the engineering, or perhaps more emphasis on cost/benefit. But really
I'm just guessing. In any case, you aren't going to save any money even
in the very short term by wrapping your insulating castable in
fiberglass insulation. Frax isn't that expensive.

This is an interesting discussion, and certainly efficiency is
important these days. But for me its more of just a way to noodle
around and keep stimulated while I'm equipment building. Afterall, no
one in this business is going to keep their head above water by being a
glass techno-nerd. One good design will far outway all the efficiency
you could ever hope to design into your equipment. One's energy is way
better spent creating.

Mike Firth

unread,
Sep 23, 2005, 2:49:41 PM9/23/05
to
Well, how about this: When you increase the volume of a hot thing, the
outside area, which determines the heat loss, goes up a lot less than the
volume of insulation. Air space insulation, such as fiberglass, has almost
no conductive paths through the volume so speaking of increased solids in
the volume is wrong. Fiberfrax is a lot more expensive than fiberglass and
at the temperatures involved gains almost nothing because they are both
restraining air from convection and the air is the actual insulator, frax
being reflective of heat at higher temperatures. One problem to deal with
is that if the inner insulation is too thin or ineffective, then the
temperature in the wall where the inner insulation meets the fluffy stuff
will get high enough to melt the fluffy stuff if it is glass. A second
problem is that if the fluffy stuff is exposed to furnace heat escaping
through the door (i.e. poor containment of fluffy stuff) it will melt along
the edges if it is glass.

--
Mike Firth
No more levees
Bury old Orleans
Raise New Orleans up if it is worth saving
--
<Jf...@maui.net> wrote in message

news:1127441778....@g47g2000cwa.googlegroups.com...

Jf...@maui.net

unread,
Sep 23, 2005, 10:38:49 PM9/23/05
to
Okay, I understand your point about the geometry of a cylinder. It
doesn't apply exlusively to "increasing the volume of hot things". It
applies to all cylinders. But your conclusion doesn't follow. Simply
the fact that volume increases much more than the corresponding
increase in surface area, does NOT mean that a hot cylinder will never
reach a point such that further insulation is inefficient--regardless
of cost.

Frax reflects heat in proportion to its density. That's why we pay more
for a higher weight frax of the same thickness. As density increases,
so does heat conduction. Why not just use a lighter weight insulator?
Because its not effective at stopping radiant heat loss at high
temperatures. There's a reason why we don't wrap our furnaces in
fiberglass insulation. As I said before you can wrap the outside of
your furnace in home insulation after the frax. But THAT is a
questionable use of funds, more than the frax. Because, like I said,
air is free.

nJb

unread,
Sep 23, 2005, 11:19:20 PM9/23/05
to
den...@debrady.com wrote:

How much would you pay for the kiln? I've thought of building one with
tubes that would use forced air during the cooling cycle. It wouldn't
add more than a few hundred dollars to a large(16 sqft bed)kiln.

nJb

unread,
Sep 23, 2005, 11:25:31 PM9/23/05
to
Jf...@maui.net wrote:

he sent out pebbles for the magless project a few years ago. Too bad I
wasn't in.

Lack of evenivity at fusing temps would be detected right away. At
annealing temps it might show up too late. If a kiln is evenifite at
fusing temps do you think it would be the same during annealing?

nJb

unread,
Sep 23, 2005, 11:27:07 PM9/23/05
to
Joe wrote:

The main offender should have spent more time in the libery.

Mike Firth

unread,
Sep 24, 2005, 5:31:27 AM9/24/05
to
> questionable use of funds, more than the frax. Because, like I said,
> air is free.
but some thing, cheap or expensive, has to keep the air from moving,
otherwise you get convection which moves heat very nicely.

> increase in surface area, does NOT mean that a hot cylinder will never
> reach a point such that further insulation is inefficient--regardless
> of cost.
As I read this double negative, you are arguing my point: after a certain
amount of insulation, there is no point in adding more, but not because the
heat loss starts increasing with greater area.

--
Mike Firth
No more levees
Bury old Orleans
Raise New Orleans up if it is worth saving
--
<Jf...@maui.net> wrote in message
news:1127529529.4...@z14g2000cwz.googlegroups.com...

Mike Firth

unread,
Sep 24, 2005, 5:34:15 AM9/24/05
to
And I want to extend my apology to both Henry's. I meant Halem and his
book. Henry Havens was a co-worker of mine in the electrical department at
Elliott's Hardware who began working there after suffering a high voltage
accident and who has since died.

--
Mike Firth
No more levees
Bury old Orleans
Raise New Orleans up if it is worth saving
--

"nJb" <no...@nowhere.com> wrote in message
news:dh2glk$q39$2...@news.xmission.com...

Jf...@maui.net

unread,
Sep 25, 2005, 11:58:05 AM9/25/05
to
No not neccesarily. There are many factors that contribute to uneven
firing at fusing temps. Firing too fast, such that there is a large
difference between the thermocouple temp and the surrounding
refractories, is probably the main culprit. And then you have bad
element configuration relative to the firing surface. Like if the
elements are too close to the work, or badly spaced. And then there's
unfortunate geometry, like square kilns, where the corners are cold.

At annealing some of those things can come into play, but more in the
case of long annealing times. Evenivity is somewhat easier at annealing
temps because the refractories are reradiating their heat out to
equalize. So that, initially and for a period of time every surface is
hotter than the annealing temp, so every surface is radiating, not just
the elements. The biggest issue in evenivity for annealing for fusers
IMO is firing directly on IFB like some guys (Bert) do. Since you have
no flow under the work, and since heat has to pass through the glass to
get to the IFB, you have a dramatic gradient between the top and bottom
surfaces. Of course that's not as bad as a gradient from side to side
in a flat piece, but its still not good, especially for thick pieces.

Right on Jack about the thermos analogy. We are cold here in the
coastal range of Oregon. My kid had never SEEN a thermos before, and
when we were driving back from town I poured some joe from mine that
was hours old, and it was great to see his amazement. And, this
technology is used to great benefit with low-emmisive coatings. If you
remember when you broke your brady bunch thermos in the third grade,
how the glass that came out was mirrored and coated with some weird
powder....

Jf...@maui.net

unread,
Sep 25, 2005, 12:25:01 PM9/25/05
to
Right, you are arguing a practical/financial point, and I'm arguing a
theoretical one.

I do concede that you're right, there is a huge difference in the
extent of diminishing returns for lightweight materials like fiberglass
insulation, and packed frax. I really have no idea when--or even
if--those diminishing returns eventually proceed in reverse with the
lightweight materials. I haven't thought about it much because I see it
as largely irrelevant to my situation. It is an article of faith among
professional, long time engineers and furnace builders that this
(efficiency proceeding in reverse) is the case with packed frax, at
somewhere between six inches and a foot. I'm done with the furnace and
its cooking off the refractories. It really wasn't a problem to create
a dead air space, although obviously you have to have a door.

But lightweight insulation is a solid nevertheless. Do ten shoulder
presses with a roll of r-30 and see how you feel about the issue.

nJb

unread,
Sep 25, 2005, 6:50:12 PM9/25/05
to


I'm picking out a thermos for you.
Not an ordinary thermos for you.
But the extra best thermos that you can buy,
With vinyl and stripes and a cup built right in!
I'm picking out a thermos for you,
And maybe a barometer too,
And what else can I buy so on me you'll rely,
A rear end thermometer too.

Henry Halem

unread,
Oct 3, 2005, 6:45:59 PM10/3/05
to
> Jf...@maui.net wrote:
>
>> Right Henry H. did have an idea (or reported an idea) to achieve very
>> even heating by using compressed air through stainless tubes to set up
>> convective currents in a fusing/casting kiln. Henry is also not a kiln
>> former. In any case, the idea was not to cool down the kiln, but to
>> achieve more even heating.
>>
>> The problem with that idea was the bane of every kilnworker producing
>> gallery quality work--crap landing on your work. Even a slight breeze
>> can stir up refractory dust from the primer or brick and shitify the
>> piece. Also, convective currents already exist in a kiln naturally when
>> there is an air gap under the shelf. Uneven heating is more a problem
>> when firing fast or when the elements or kiln height are configured
>> wrong.
>>
>
> The breeze isn't a problem as it would only be used during annealing.
> IIRC, it was a proposal to correct annealing problems for large kilns
> and large pieces.

We have used this convection system in our large kilns that were not
configured all that well and had serious thermal gradation. The system we
employed is a common system that has been used long before I found out about
it. We never found any particles blowing around the kiln as the work never
had any deposits of anything on their surfaces. The word "breeze" is really
an overstatement in relation to what is actually going on in the oven. The
caveat is that I would imagine if you put too much pressure into the system
you could create problems. The air movement is ever so slight and does not
aim directly at the work but is aimed toward the walls or crown depending on
where you place the tubes. We employed the system from the moment we turned
the oven on until the temperature was around 400 deg. F.

nJb

unread,
Oct 3, 2005, 11:21:10 PM10/3/05
to

Henry,

Metal or ceramic tubes? What composition?

Henry Halem

unread,
Oct 3, 2005, 11:50:22 PM10/3/05
to
nJb

Stainless Steel

Jf...@maui.net

unread,
Oct 4, 2005, 8:30:37 AM10/4/05
to
It does sound like a useful system for big ovens. I've always been
skeptical about 'zone control' in these big fiber fusers; its
problematic for a bunch of reasons. This is a simple modification and
it would be easy to test with a couple three or four pyrometers. Jack
you might want to ask Ron Coleman at warmglass, I think he was
considering using this for an evenivity-controlled fusing project.
Still, I use Bullseye wash--which is special because of its extremely
fine particle size--and I am totally paranoid about any whiff of a poof
of a shoosh. If you try this, do a test and run your fingernails over
the surface of the piece, blackboard style. You'll be able to feel if
there is too much crap on the surface.

Mike Firth

unread,
Oct 4, 2005, 1:20:01 PM10/4/05
to
> where you place the tubes. We employed the system from the moment we
> turned
> the oven on until the temperature was around 400 deg. F.

400F going back down, right?

--
Mike Firth


0 new messages