How to design electric elements for kilns

Gold Refining Forum

Help Support Gold Refining Forum:

This site may earn a commission from merchant affiliate links, including eBay, Amazon, and others.

rolynd

Member
Joined
Feb 21, 2009
Messages
15
Location
Germany
Hello guys,
I recently became interested in refining and still reading a lot on the process as in this area I am still a newbie.
What I noticed was that some people are trying to build electric kilns and in this area I do have some experience, so I wanted to contribute a little of what I know on this process to the forums. Maybe this is a help for some of you kiln builders.

This comes in two parts, first the build of the furnace, second the theoretical part of how to design the proper elements.

Overall there are three considerations:
1. decide of a Wattage the furnace must have for a given size
2. from this you calculate the element
3. checking for one of the most important factors: surface load of the element so its longlived and does not melt down when in use.


As outer shell I used an old clothing bin which I had laying around.

For insulation I used a porous stone used for insulating houses and such sold under the brand name Ytong here in germany.

pro: insulates well, cheap, readily available, easy to work with.

Con: usable only for temperatures up to 1000-1100°C, will see how it holds up over time.

Heating is accomplished with a coiled nichrome 80/20 wire which is good for temperatures up to 1200°C



For controlling the power I used a 2600W /15A power control (brand:Kemo)



I decided on 2000W of power for the kiln. with 230V this gives me 26,45 Ohms. the wire has 1,4 Ohms/meter so I needed 18,89m of wire to get 2000Watts.

surface load of the wire is ca 3,4 Watts/cm2 which is a little high if the furnace is used longer times at top temperature but for the intended 1000°C it will suffice.

You could also have built a square one but had the round container on hands so it was a round one.

First the clothing bin was cut to size and a bottom was fit in. The Ytong stone is easily shaped with saw and rasp.





ofen6.jpg




Then a cutting jig was built to saw the stones to 22,5 degrees so that eight of them will fit and make a good chamber. The outside of the angled blocks was fitted to inside shape of the shell with a rasp. For a perfect fit the stones were ground against each other and installed in numerical order.



ofen5.jpg




Then the inside of the blocks was grooved to hold the element and painted with a thin coat of fire cement. The coiled element was fitted in the grooves and the ends lead outside to connect power.

In similar way to the bottom the lid was manufactured from two halves cut from Ytong and fit into the lid. The blocks are held in place by a few nails round the outside of the lid.



ofen1.jpg




After drying the kiln was fired and went up to 700°C in 16 minutes after 25 minutes the thermocouple reads 1030°C with the temperature still slowly climbing. the outside of the kiln only gets up to 50-55°C which speaks for the good insulation of the used Ytong blocks.



ofen2.jpg




ofen4.jpg




ofen3.jpg




After 4 firings now the kiln still holds up fine, no cracking on the inside only the lid shows some minor cracks but structural integrity is not compromised so far. The lid was also coated with the fire cement but it fell off right after the first firing.

ofen7.jpg




The total cost of the kiln was:



- 5 blocks of ytong 60cm x 20cm x 10cm at 1,45 each = 7,25

- 25m of Nicrome wire 1mm = 4,99

- Clothing bin = free

- power controller = 8,47



Total = 20,71€



I dont think that the kiln will hold up forever but I did not want to spend so much money. Total time needed to build this thing was 2 days. So at 20 bucks and 2 days I now have a kiln that reaches 1000°C at least and I did learn enough so that the next one can be more sophisticated. At 2000W the cost of running it is much lower than using gas. Over here gas prices go right through the ceiling.


Designing of the elements:

It was built according to the information on tube furnaces from Georg Brauer, Handbuch der präparativen anorganischen chemie, Volume 1.



At first you will need the size of the furnace and the temperature it will have to reach:


>
For every cm2 of surface area of the kiln you need a certain amount of watts to reach specific temperature.

From Brauer:

Temperature in °C , Multiplication factor

300 , 0.2

400 , 0,4

500 , 0,6

600 , 0,8

700 , 1,0

800 , 1,3

900 , 1,6

1000 , 1,9

1100 , 2,2

1200 , 2,6

1300 , 3,0

1400 , 3,5



As this rule of thumb is for tubular furnaces I did neglect top and bottom area.

With a chamber of 14cm x 20cm this gives me a surface area of 879,6 cm2

I wanted it to reach at least 1000°, but nichrome wire is capable of max 1200° so I chose 1100°

880cm2 of surface area x 2,2Watts = 1936 Watts total power needed

To simplify further calcuations = 2000W



The available power supply is 230V AC.

What we want to do now is to calculate the heating element:

To do this you use R = U^2/P, r =(ohms), U =(Volts), p=(watts)

with 230V and 2000W I get a resistance of 26,45 Ohms



Now you look at your wire, mine was 1mm nichrome wire with a resistance of 1,4 Ohms/meter

You just divide 26,45 by 1,4 which gives the length of wire of 18,89m



Now you check if the wound element is fitting in your kiln. Diameter of the coil should be 10-15 times wire diameter, spacing between single windings minimum = wire diameter



If wound on a center rod of 10mm diameter one winding is 3,77cm

with 18,89m I get 501 windings. With wire diameter 1mm unstreched coil is 50,1cm long with minimum stretch of 1mm between windings ca 1m but can be stretched to needed length easily. You just have to watch that the minimum stretched coil is not too long to fit inside your kiln chamber.

It also does not have to be one single coil, it can be divided into smaller coils which are put in series. only overall wattage must be the same.



If the coil gets too long switch to a wire diameter with higher resistance (usually smaller diameter) this will shorten the length of the element but at the same time surface load will increase. You will have to balance the factors a bit to get optimum performance.



Now you are almost done. Another important factor for the longlivety of your heating element is surface load per Area. The brauer tells us that at lower temperatures up to 900 a surface load of 3,5Watts/cm2 and for higher temperatures e.g. 1350° a surface load of 1,5Watts/cm2 is appropriate for continuous use of the furnace.

You now take your watts and divide by total surface area of the needed wire.

18,89m of 1mm wire gives a surface area of 593,44 cm2

2000W/593,44cm2 = 3,37W/cm2



If used permanently at top temperature the surface load is a little too much , you will not have the elements melt but the life of the elements is reduced some hours but at the intended temp of the furnace it fits.



If you have any further questions or additional information please feel free to ask I will try to answer the best I can.



ps: there is a phenomenon called resistivity that means the wire will change its resistance when temperature climbs, this was neglected in calculation but should change the length of the wire only a small amount (usually shorten it) which will drop the wattage a bit. roughly calculated I will have 1900W not 2000w at top temperature. There is also a Method to calculate the element using a formula which uses the resistivity but as this only leads to a few centimeters difference in the calculated wire I still use the simpler above described method. This has worked for me so far, the kiln is still running fine and I did not have any problems so far.
I think this method is not the absolute pro way of kiln designing , you can go into this much further e.g. calculating with resistivity, conductivity and thermal mass of the used refractory to reach certain heating rates but as said before this is what I have done and it works just fine.
The kanthal handbook at kanthal.com is also a very useful source for all who want to build a kiln, it gives plenty of data on the performance of wires, maximum recommended surface load in different applications and so on.

You can adapt the above principles to your available needs and materials as needed.

Regards
Ralph
 
Yes that is a very nice job. The inside looks really close to the kiln I got from freecycle or craigslist. I plan on converting mine to propane.
It is obvious that you did your homework and planned the project well.
I hope it lasts much longer than you expect.
Jim
 
Excellent job, rolynd!! Well designed and constructed.

I do agree with 4metals, though. To me, it would be more interesting to see you come up with a box furnace. Have you ever used the type furnace you built for melting metals?
 
Rolynd,

I really appreciate your post, as I plan to build another furnace (for fire assays) soon enough.

I did built one similar to yours about one year ago. I used a Kanthal A-1 wire for the element, and Inswool (Kaowool equivalent) for insulation. The furnace was PID controlled with a Solid State Relay.

I did not respect the element max surface load and I melted it. But IIRC, the furnace was 1600W and able to reach 1100C in 10 min.

I am still unsure how to hold the element in my next furnace. I will use insulation boards inside. I saw someone coating his heating coil in order to prevent oxidation but I am not sure about the mixture of the refractory material...

Thanks for the maths, it will greatly help.

P.S. Where did you get your power controller ?


Thank you
 

Attachments

  • pic 047.jpg
    pic 047.jpg
    154.8 KB · Views: 768
  • four 054.jpg
    four 054.jpg
    131.5 KB · Views: 768
I know that you squared the voltage in your formula, but the way it is wrote some may think you multiplied by 2 and divided by the watts.

Excellent post on furnaces, I am just end up building one of them.

I was going to use a 10k heat pack for my elements 2 circiuts with sequencing. The heat pack are from old miller furnace.

Jim
 
Noxx,
thanks for editing.
If you also use Kanthal wire for your next furnace it would be good if you place the element in some kind of grooves because when up at temperature it tends to sag if not supported.
I also thought of coating the elements of my next furnace to prevent them oxidizing too fast especially when melting or assaying in a reduced atmopsphere which is no good for elements..

As the protective coating the kanthal wire itsself is producing when heated consists of aluminiumoxide I will use this mixed with a little bit of water to a slurry to coat the elements. You can get calcined alumina at pottery supply stores, be sure to get the calcined stuff not hydrated alumina. Also some blasting grit (you would need very fine grit) is made fom aluminiumoxide, dont use already used grit it contains residues from the blasted stuff.
Pure aluminiumoxide is also sold as polishing powder for lapidary use, good but expensive. Calcined MgO would also work as its used for embedded elements but I dont have a supply for dead burned MgO. The stuff climbers use is MgO but not the dead burned variety.

I dont know if the calcined alumina will stick good to the wire maybe coating will have to be repeated after some time but it will do no harm to your element. Be sure that whatever you use for coating contains no free silica as kanthal recommends the elements not to be in contact with this, it will shorten the element life.
There is some commercial stuff out there for element coating ITC-100 all purpose and ITC-213 specially for element coating. But its very expensive.

As the infinite power switch always needed a good time fiddling to get an exact emperature I decided to use a Pid with thermocouple. This works fine with no hassle.
Here some pics, neatly housed in an old PSU box. I used a regular wall outlet to be able to connect different kilns/other devices easily. Also you can get a reading from the thermocouple when no device is plugged in and just use it as some kind of thermometer.

pid2.jpg


inside:
pid.jpg



I did some melting with my furnace, I had success with silver but when trying to alloy some bronze the furnace was struggling it did get the copper melted but for casting I would have needed a little more overshoot. So the next furnace is already planned, with this one I will use Kanthal A1 which should give me the desired 1200°C easily.
The ingots below are sadly no gold just Al-bronze but i selected this alloy because of the nice golden color. :) First pic right after cast, second after some cleanup. The first ingot would obviusly have needed more stirring. Second one came out much better, still some surface pitting but usable.

image-FC36_4B498BAC.jpg


image-D418_4B498BAC.jpg
 
Rolynd,

Great job. Lots of good information here.



Did you wet the firebrick before you applied the refractory coat?

"After 4 firings now the kiln still holds up fine, no cracking on the inside only the lid shows some minor cracks but structural integrity is not compromised so far. The lid was also coated with the fire cement but it fell off right after the first firing."

Applying it to a dry surface might leave a cold joint.

on a related subject:


I was in Lowes the other day and noticed some premix refractory patch that was rated to 3000 Deg F. (1649 C.) It was about $15, in case anyone is interested.


Magnesium hydroxide can be made with Magnesium Sulfate and Sodium Hydroxide. This is the process used to make the popular antacid.

The hydroxide can then be filtered, washed, then ignited in a furnace to make Anhydrous Magnesium Oxide. Magnesium Hydroxide decomposes to Magnesium Oxide at 350 Deg. C

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

The melting point of Magnesium Oxide is: 2852 °C, 3125 K, 5166 °F
 
Rolynd,

I do not think one should coat the elements with ITC-100, since it offers low thermal conductivity and is made for coating fibers in the furnace. Perhaps ITC-213 is the solution. But it's expensive. Hopefully we can find the patent and know what it's made of. I'm afraid both aluminum oxide and MgO won't stick properly unless we find a good binder.

MgO offers a thermal conductivity of 42 W / (m*K)
and
Alumina (Al2O3) is 30 W / (m*K)
 
Yes, I also think that coatings of pure al2O3 or MgO would not stick to the surface well but I cant think of an appropriate binder . Any Ideas? Maybe small amounts of CaO, Iron oxide or Titanium oxide would help as these are always found as additions in commercial high alumina refractory mixes but I am not sure if there would be harm to the elements. Nevertheless, when I have finished my next furnace I will experiment with some alumina and see if it works and how it holds up as I think this will ,even if not working, also not damage the element.

If someone could find patents for the ITC coatings it would be greatly appreciated! I have not been able to find some so far. I refrain from buying the stuff because of price... and its always more rewarding to figure something out yourself and then really know why and how this works.

What i could find was the composition of satanite, an high temperature furnace cement:

SIO2 : 34,8%
AL2O3 : 59,4%
TiO2 : 2,6%
Fe2O3 : 1,5%
CaO : 0,3%
MgO : 0,4%
NaO2-KO2 : 1,1%

The Alumina/silica ratio would lead to mullite formation on sufficient heating and the other ingredients must be helping in green strength or sintering somehow.
 
If you look at a commercial parts site, you can get elements that are already coated with minerial insulation and a SS coating that is very resistant for about 25.00.

These are the radiant heating elements used hot food holding stations,
you can bend them to shape and they are very robust.

Jim
 
According to my calculations, my last element had a surface load of 17,68 W / cm2

Now I know why it melted :p

are you sure about Brauer's table ?

According to him a 1 ft3 (5581 cm2 surface) box would need 12 279W to reach 1100C...
 
NoXX
this sounds a bit too much for me, but as I never did build such a big furnace it never ocurred to me to check this rule for bigger furnaces. He does not say up to which size of furnace this rule can be applied. This rule is for tubular furnaces used in preparative chemistry which have neither lid nor bottom, so this area will have to be neglected in calculation. But even if I do this I still get around 8kw for a 1 square foot furnace. But we are using lids and bottom so one can get away with less I think. Here is the orginal quote:

Faustregel: Bis 300 °C benötigt man 0,2 Watt je cm2
, bis 700 °C zusätzlich 0,2 Watt/100 °Cr
zwischen 700 °C und 1100 °C müssen für je 100 °C 0,3 Watt mehr aufgewendet werden,
zwischen 1100 und 1300 °C 0,4 Watt, zwischen 1300 und 1600 °C 0,5 Watt und über 1600 °C
0,6 Watt je 100 °C .

translated:
Up tp 300°C you need 0,2W/cm².........................................this makes 0,2W/cm² =0,2
up to 700°C an additional 0,2w/cm² for every 100° more.............this is 400° more so plus 4x0,2W/cm² =0,8
between 700-1100°C for every 100° an additional 0,3W/cm².........this is 400°more so add 4x0,3W/cm² =1,2 which makes a total of 2,2W/cm² to reach 1100°C
between 1100-1300°C add 0,4W per 100°
between 1300-1600° 0,5W per 100°
over 1600° 0,6W per 100° more

you can get the brauer at: http://gigapedia.com/items/376582/ but its in german of cause.



commercial laboratory (Nabertherm,Rohde etc) and pottery kilns in the 1200°C max temp range use around 0,9-1,4W/cm² so some less some a little more. I could not find data on insulation or heating times. unexpectedly the smaller the furnace is the more W/cm² are used , big furnaces of several ft³ use less W/cm² than the smaller ones.

If you calculate with the upper range of 1,4W/cm² x 5581cm²(surface of 1ft³) i still get 7813,6W = 7,8kw
if calculated in the lower range you will need around 5kw. Which sounds a lot better than 8kw

Maybe you can get away with still less than that but you will need very good insulation and have to be prepared for long heating times to get up to temperature.


What do you want to use the furnace for? If you need the 1ft³ size why not build a propane powered one, burners are easily made and medium sized propane burner has much more BTUs output compared to electric. But if you need exact temperature control or are limited to indoor use there is no way around electric.

I am limited by my household current of 230V 16A max which gives me around 3600W max but if you dont want to pop the fuse every time you turn on a lightswitch when the furnace is running my maximum furnace wattage is limited to around 3000W so i never tried to build some bigger. my personal opinion is to build a furnace with a little power to spare so you get quick heating times and you dont have to run the furnace at the upper limit of its capacity which will prolong life.

Its nice to talk to some fellow kilnbuilder and share some ideas and opinions, if you build your furnace please keep us posted!
 
This is very impressive. I love seeing Georg Brauer's collection mentioned, and in the original German (my idea of retirement is to go through his book and do the experiments)!

How much power you put into your furnace depends on the insulation.

The only thing I would have changed would have been using the YTong as a backing material over conventional kiln brick--granted, you saved much money by not using the typical lightweight kiln brick, but it won't last as long.

I wish they had Ytong here in the United States!!! I would use it in an instant for a box oven, then put heavy duty kiln brick over it as a facer. The heating elements I buy are not free wire for the winding--I buy them as a module that inserts into the furnace. It costs more but less down time when an element eventually goes down.
 
Hello.
I am new in this area of electric ovens, and would like some help.
My idea would be to build a cylindrical electric oven to reach temperature of up to 1800th.
In case it would be for melting glass discs telescope.
I imagine the maximum size of the glass melting is 20 "in diameter.
Whereas the height of the internal parts of the furnace was 20 cm and the width would be 53cm. (20 "x 2.54 + 22 cm = 50,8cm lateriais coating)
In this case, and from what I understand the above, the calculation of the furnace area would be:
20cm X 53cm square area X Pi = 3.328cm2.
how many need wats?
which size of resisdencias for that power?
 
Back
Top