Info on home made power supply wanted

[megalomania]

Does anybody out there have information on building a power supply to run an oven or furnace? I am looking for something that can run off of 110 AC current and put out more than 10 amps. I have been searching the net for plans, but I have not found anything conclusive. I just want something simple, like the electronics of an oven or hotplate. If anyone can steer me in the right direction I would appreciate it.

I am trying to build an inexpensive improvised tube furnace modeled somewhat after a kiln furnace. Fortunately kiln users have provided considerable do-it-yourself information, except where it comes to the heater. I have acquired an inexpensive temperature controller on Ebay and I found some wiring diagrams to hook it up. Add a couple of relays, some switches, and an outlet and this will be a quite spiffy thermometer.

The temperature controller is also a digital thermometer. You wire up a thermocouple to it and set it for a specific temperature. The addition of a relay controls a plug outlet into which your furnace is plugged in. This plug stays on when the temperature is below the limit and off when the temperature gets too high. The actual temperature controller is way more complicated than that, but that is the jist. It can be programmed to ramp up the temperature at a certain rate, it has dual alarms in case your temp goes too high or too low, and some other stuff I’ll never use.

I was thinking if I can’t build my own power supply that I will buy some cheap hot plates and rip off the elements and just use the electronics. 3 hotplates should give me enough power to run 3 separate heating elements giving me 3 zones on my tube furnace. These would be cranked up to high and plugged into my controller which turns them on and off as necessary. An old electric stove might work as well.

I will likely be using some kanthal wire as my heating element. I will coat a 12 inch stainless steel pipe with some fireplace cement and let that harden. Next I will wrap the kanthal wire around the pipe in spiral fashon, probably in 3 separate sections if I use 3 different hotplate controls, or if I can build a big enough PS myself. Finally another coating of cement is applied to seal in the wire and affix it to the pipe. The first coat of cement is necessary to prevent an electrical contact between the pipe and the wire.

I will leave a small hole for my thermocouple in the center of the tube. I found a cheap supplier where I can get a 2 or 3 inch length of grounded stainless steel enclosed thermocouple with 12 inches of lead wire all for $11. This arrangement actually measures the temperature of the pipes exterior, so I will have to calibrate it with my other digital thermometer to see what the difference between the surface temp and interior temp is. I doubt it will be off by more than 15-30 degrees C.

Total cost may be pushing $100 by the end of all this. That is damn cheap considering the cheapest I have ever seen a tube furnace is $700. The most expensive part is the temperature controller, but I got a damn good deal on Ebay for one that retails at $120. I got mine for ¼ of that. Buying the relays that can handle line current will likely run me another $20. Wire, switches and terminals will cost a bit more, and I may want to get an enclosure, so that could be another $10. The wildcard is the price of the power supply, three new hotplates or an old oven might add up, that’s why I figure it may be cheaper to build one myself.

[Ollie Snowie]

What is it for, exactly? Does the temperature need to be very tightly controlled? How big does it need to be?
I have taken a temp. controller from an electric cooker apart, and there is no temperature sensor that goes to the main heating element; just a small heating element in the temp. regulator. I suppose they are in series so they both get hotter and colder in time with each other, and this is good enough for an electric cooker.
If it doesn't need to have a large diameter, bar heaters are a possibility because they are hollow if you remove the end caps. They are a bit brittle though, but as long as the wire stays on, it tends to stop them falling apart even if they do fracture. I have found that the disk shaped heating elements in hotplates do give a fairly uniform temp; I have managed to anneal some pieces of glass up to 4cm from end to end and some 3cm circles without them cracking. I suppose disk shaped elements are of no use to you though.
A fairly whacky idea I've never tried myself would be to get several bar heater elements, make parabolic reflectors for them and put the tube slightly off the focal point so as to get more uniform heating.
I've also found that bar heating elements burn out very quickly when used in a furnace where the heat is trapped and the temp. is much higher, but electric hob elements are very resistant, probably due to the coating, so I would advise you to put another layer of fire cement or something over the wire.

[Tuatara]

I don't understand what you mean by 'power supply'. Can you not set your elements up to run directly from 110VAC ? Thats how most ovens are set up - simple on/off thermostat and a mains relay connected to an element.

Do you need to use stainless? If you were to use quartz tube you could wind your heater directly onto the tube, then insulate with mineral wool. The only fireplace cement I've used had fairly poor thermal conductivity. In fact, if you were to obtain a quartz heater element (the 'glass' type with a wire inside) you can remove the wire (which will be nichrome) then re wrap it around the outside of the tube. The wire size and length should be about right for direct connection to 110V. You don't need Kanthal unless you're planning to go well over 1000C, and stainless wont hold up at that temp (neither will quartz).

If you were to provide some specifications for your tube furnace we might have some more ideas to help you (like size, temperature, power etc)

[megalomania]

I was laying in bed last night thinking I could just run the heating elements directly from line current, but it can’t be that easy can it? I am thinking there must be more to heater controls than that, of course I don’t know. If someone could provide an example way to set that up this is exactly what I am looking for. My temperature controller will turn it on and off, I just need it to run.

I don’t want to use a porcelain (quartz or ceramic) tube for matters of convenience. It is easier to connect a pipe with threaded fittings. Although I will have to consider it because a ceramic tube would offer some chemical resistence as well as a way to swap out ruined heating elements if I just wrap them around the tube and cover them with some fire brick. There is just the problem of sealing the ends up without permanently sealing it.

Indeed I don’t need kanthal, but I haven’t really found any better wire. Kanthal sells many different kinds of wire and the prices are not that high.

I am looking to operate the tube furnace from 400-800 degrees C. Most of the time it will be running around 500 C, with the occasional need for 800 C, but I would like it to go up to 1000 if need be. I would only rarely run it that high if ever, even 800 C would be pushing it.

The size is as I mentioned, 12 inches long, maybe a few inches less since I have to leave the threaded ends open. I will probably use ½ or ¾ inch diameter pipe, I forget what I have.

I will use a thin layer of cement to protect the pipe surface and prevent electrical contact, so its low thermal conductivity should not matter. I will pack a thicker layer on the outside to keep the heat in. If someone can suggest a cement with better thermal conductivity I would like to use it for the inner layer.

[Tuatara]

Yes, it really is that easy! For 110VAC, 10A you need a heater resistance of about 11 Ohms - just simple Ohm's law (V= I x R). That will give you about 1kW. As the element heats up the resistance will increase somewhat, so it will self-regulate to some extent.

While I was out digging potatoes today another problem occured to me. The thermal expansion coefficients of the SS and the cement wont match, so after a couple of cycles the cement is going to come loose - not what you want with a live element. You may need to buy straight, sheathed elements and bend them to the shape you want. This is how hob and oven elements are made. The elements will take a fairly tight bend, but not 3/4". So you would need a number of short straign elements which could then be bound to the tube using steel wire. The elements themselves consist of a thin incoloy tube, packed with refractory powder (MgO IIRC), with the wire heater held firmly in the centre and nicely insulated from the sheath.

How about a ceramic tube with standard ground taper joints? ($$$$!)

[McGuyver]

There is a company that sells flexible heating elements that you can wrap around w/e you want. I believe it works off of 110 VAC too.

Well, did a little search and here is the site: heating element (http://www.sciplus.com/category.cfm?subsection=16&category=155)

Now if you need something to control the heat you'll need some kinda potentimeter (sp?) or I guess you could just take one off of some stove.

I don't know if it will be powerful enough for you though because it only draws 90 watts. Yeah I doubt it but I don't think it's possible to have it flexible with more power than that.

Ah, farther down that page there is a iron element that gets much hotter too bad it doesn't wrap around. Maybe if you bought a bunch though it would work.

[megalomania]

I am pretty sure I will be using Kanthal AF wire, which is good for a variety of reasons. First of all it is quite bendable, it is just wire after all. It has known properties of resistence, thermal expansion, and max temperature handling. It also has quite a long use life, especially at the temperatures I will be running it.

I will be hitting the junkyard sometime soon to acquire myself an old oven. I should be able to acquire it for a few dollars.

My current trouble is figuring out the amperage of household current. I have learned most house fuses are rated at 20 amp. I also now know my furnace should be able to handle up to 3000 watts. At 110 V and 20 amps I need a resistence wire of 5.5 ohms (110V/20A = 5.5, V/I=R). Kanthal AF is 0.649 ohm/foot at 19 gauge, so I will need about 8.5 feet of wire. That assumes I know the amps, which I don't.

Hopefully once I get the oven taken apart I will be able to see how it works. I would prefer to just plug the resistence wire in to line current, but I still need to find out the ampage to get the resistence wire length right. Anybody know how to regulate the number of amps?

I don't know where the watts fit in exactly, but as far as I can gleen from my kiln books a small kiln needs about 3000 W for 3-4 cubic feet. Since mine is a fraction of that my 2200 W system should be fine (110V X 20amp = 2200W, V x I = W).

[grendel23]

You seem to have a handle on Ohm's law, that will give you all the info you need to select the correct impedance for your heater. If you want 2000W, more than enough IMHO, P/V=I, so 2000/120=16.66A. Then V/I =R, so 120/16.66=7.2ohms

That part is easy, and luckily, so is the rest. Use a solid state relay to control the heater. A unit rated at 25A and 240V can be had new for about $30 and often scrounged for much less. I have several and would be willing to send you one if you like.

Most of these use a control voltage of 3-32 volts DC to control them, easy to interface to your thermostat.

If you really wanted to get fancy you could use a phase control regulator, but that would be overkill for your application. You are better off selecting a heater, or the correct length of nichrome wire to get the wattage you want, then using a simple on/off controller for temperature control.

[Jacks Complete]

Megalomania,

I can see from your post that you are a little confused about the whole mains powered thing. Might I suggest that you be really, really careful with your oven, as touching the elements, etc. will probably kill you with the circuit you describe! (You know this, but it bears repeating.) A liquid or metal powder spill could cause a conduction path back to the tongs you are holding, and to you. Try to get a well-sealed element from an oven, or make yours sealed. I wouldn't want to trust loose type glass fibre with my life.

The issue with household mains is that, circuitwise, you treat it like a constant voltage source. From V=IR you can see that as the resistance drops to zero, the current reaches infinity. This is what we have fuses for, to avoid destroying the wires!

Wire a slow-blow fuse into the circuit on the live side, rated at whatever you decide to run your oven at.

Basically, you want to pick any arbitary current below the fuse rating, plug it into the formula with whatever the line voltage is where you are, and then from the resistance you determine the length of wire you want.

You can work backwards from the power, or forwards. Use P=IV (= I^2 R) and determine the power you are going to be putting out, or work from the power you want to determine the current draw you want, and hence the wire length, as described by grendel23.

Anyway, try not to kill yourself, as I really like your forum!

Edit: Sorry, some of this might seem a bit condesending. It isn't meant to be.

[Tuatara]

A couple of extra notes: 110VAC is a nominal rating. Line voltage is usually controlled to +10% -15%. So work everything our based on 122V and you should be fine.

Also if you are going to use a metal pipe it would be adviseable to wire it to safety earth. Ask your local sparky about the line supply in your area, as you may or may not have an earth connection available in your house. Here in NZ we have three-pin outlets - live, neutral and earth.

Choose your power according to your heat needs - how much heat energy to you need to supply to your reaction, how much are you going to lose through the insulation. The power/size guide for kilns is just that - a guide. The better your insulation the less power you need, although warm-up time also becomes an issue with a kiln.

If you can find some heat resistant fibre or cloth to wrap your tube in first, that should keep your element clear and electrically insulated. I initially thought of fibreglass, but that would melt at 800C. Asbestos would work, but I doubt you could get any these days. Have you a local kiln supplier, whose brains you could pick?

[megalomania]

I have cracked open an old heater and probed its inner reaches. There is nothing to it really. There is a 3-prong cord that ends in 3 wires, one of which is grounded to the metal case. The second attaches to the heaters variable control dial. The dial is set to off at the lowest and high at the highest with a 180-degree turn between them. This knob controls the temperature. The third wire connects to a switch with goes from 1300 watts to 1500 watts (says a label on the front). The other end of the knob connects to a fan that circulates air, and then the wire exits the fan and connects to a series of heating elements. The third wire also connects to the elements at the other end of the circuit.

Obviously the only important piece of hardware here is the dial. Does this control the voltage or the amount of amps? Anyone know anything about knobs? Or maybe that watt switch controls the current somehow?

[Bert]

The knob is probably a thermostat. The high/low wattage switch works by controling the current. This is done by changing the resistance of the circuit with the heating element- Could be done by switching a resistor into series with the element, or switching part of the resistance element into/out of the circuit. Raising the resistance of the series with the element would cut down the wattage-
power(watts) = volts X current (amps)

volts = current X resistance (ohms)

current = volts / resistance

[Tuatara]

If its a thermostat you will probably hear it 'click' at some point as you turn the knob through its range. The click will be at a slightly different point turning one way, compared to the other. The thermostat does not control either current or voltage, as such, merely turning the elements on or off according to the temperature it measures.


The other kind of control, commonly found on electric hobs, is called an 'energy regulator'. All it does is switch on and off at a reasonably constant rate, the energy throughput being controlled by the ratio of 'on' time to 'off' time - which is adjusted with the control knob.

[udtst]

A cool loking little generator which you might be able to use is Tom E. Bearden's MEG (Motionless Electromagnetic Generator) project.
he is a link to it: http://jnaudin.free.fr/meg/megv21.htm

to me it looks like it should work. I am going to try and build it within the next few weeks.

[Ollie Snowie]

Nawwwwww,,,
Don't want to go down the road of "free energy" when all you want is a furnace power supply.
Cool, yes (if it's not a hoax - we'll be delighted if your model does work!).
Relevant, practical, no.

[matjaz]

Mega, your knowledge on chemistry being so intimidating, it's a profound pleasure to see you less confident in another area, hehe! :)

I agree with everything Tuatara said. A 1000-2000 W regulated power supply would be too big, too expensive and too dangerous for mass production in kitchen appliances. All this things just use line juice and a form of a smart on-off switch in series. The regulation is always acomplished with time management of on/off and not with a continuous regulation of current.

One sleak option would be a light dimmer with a stronger triac. This would work, but at the cost of a lot of EM smog in your powerline, since switching 20A 120 times per second is sort of dirty. And there's no need for such a rapid response, because the heat capacity of the kiln takes a lot of time to heat/chill anyway.

As for electrical insulation, ceramic beads could be used on the resistive wire. I have a couple of those in a huge old soldering iron. They are tiny cylinders, 5mm (.2") long and 2mm (.1") in diameter, with a hole drilled down the axis. They would fit very loosely on the wire, even rattle a bit actually, and the wire can be then dressed in these and wrapped around whatever.

If you need a high temperature in your kiln, remember that the thin heating wire will be much hotter than the temperature you need inside the chamber. I just has to squeeze all that power through its very limited surface. An oven heating element is designed to dissipate its 1500W inside a 200C cabinet. In your kiln, it will have to dissipate the same power into a place which will already be at 400C-800C or whatever. So, it would have a much higher temperature itself. Any additional cement on it, and it gets even worse, because you thermally insulate the heater wire.

Bert, switching on and off some of the heater element resistance is feasible, so is changing the configuration from serial to parallel. But inserting a resistor in series is completely out. The heat load on this resistor would be similar to that on the heating element itself.

McGuyver, this flexible heat wraps are very handy, but usually can't tolerate extremely high temperatures. (We sometimes use them to bake beamlines to improve the vacuum.)

Udtst, this MEG is a fancy high voltage generator and the circuit seemes reasonable, but it doesn't produce more power than what is fed in. So no free lunch. Surprisingly? ;)

[megalomania]

I am actually going to build a device similar to the one described here: http://toblerglasscreations.com/fuji/ I already have the temperature controller and the rest is easily enough obtained. Since this device will do the actual on/off switching all I need then is something to deliver maximum power. A dimmer switch is actually a good idea that I shall consider.

I doubt I will find oven or heater components to supply enough power in one unit, but I can always get more than one. I think three 20 amp heating coils would provide effective heating. Most kilns or tube furnaces use such three zone heating systems anyway.

Now I just need to know how to consistently provide 20 amps of power. I do know that the thickness of the heating element does affect the temperature, that's why I need to know how many amps I can get to choose the proper thickness for my temperature needs.

I was thinking about making a concrete mold with grooves that I can lay the wire in, like a kiln, and then cover that with a thick layer of alumina. I could lay in the pipe, cover that with more alumina, and close a top portion made like the bottom. This would give me two heating zones, a top and bottom, the alumina would provide electrical insulation as well as being heat resistent, the grooved cement molds would allow the heating element to expand, and the element would be easily replaceable if it ever burned out. Such a setup would also allow me to easily swap different catalyst pipes instead of cleaning the thing out each time as would have to be done in an embedded system. This design is actually how modern tube furnaces work.

I wonder if I could apply the alumina as a slightly moistened paste and burn it in gently to drive off the moisture, but stay a solid mass? I suppose I could also use a bit of modeling clay. I mention alumina simply because I happen to have a bunch not doing anything in my lab. I could also likely use dry concrete, mortar, or lime. In fact using lime may actually be better because it would be a nice byproduct to produce CaO after each furnace run (from my Ca(OH)2 and CaCO2 mix).

[matjaz]

I feel that you still want to have some sort of a power supply somewhere? That's not what I wanted to suggest - the dimmer was my inferior alternative to a temperature controller, not an addition. Temperature controller is way better. You will feed the heaters directly from line power. That's just the way it goes most reliably. The maximum power will be set by the resistance of the heater wire.

Can you tell me the following:
- what will be the dimensions of the kiln (pipe length, thickness, door design)
- will you be heating rods, chunks, dust, liquids?
We need that info to calculate the maximum power demands and to think chamber design. After we have established that, we can go into electrical design, since that depends on the current that we'll be switching. At that point, we'll need to know:
- what kind of temp. controller you have (either a model number or detailed specs (power demand, switching specs))
- what line voltage you have (110, 220?) and what is the rated amperage of your line supply (with heaters it is easy to exceed the rated power as allowed by your electricity provider)

A mold with grooves is a good idea if you plan to build your own heating element. But expect your cement to fracture and fall away from the metal pipe after a few uses. We'll need to think a way to keep the wire from touching the pipe.

[Bert]

matjaz-
(We sometimes use them to bake beamlines to improve the vacuum.)
I cut my teeth in electronics fabrication, machineing and a number of other technical endeavors on the Aladdin synchrotron ring project (http://www.src.wisc.edu/Outreach/materials/Brochure.pdf) at the U. of Wisconsin's Physical Science Labs. Where are you baking your beam lines? (This is OT, of course- Please contact me off list if you want to!)

[CommonScientist]

Mega- I have looked up to you as a father figure in chemistry and its related occupations.I am going to learn as much as possible from you, and hopefully, become a well respected member of The Forum.

Has anyone thought of an oven as a source of the heating element? Some can go up to 650 degrees Fahrenheit. I dont know if that will work for you but you could give it a chance. As for everything else, I cant help you much. Just tryin to help



[edit: grammer problem]

[matjaz]

Because you admire his chemistry you expect to become a respected member? :p
Sorry if I've misinterpreted you, just couldn't resist...:) No offense meant.

[CommonScientist]

Ignore matjaz's last post , I had to edit something that he so wonderfully pointed out to me. Sorry. No offense taken.

[hodio]

A good power supplier i use for metal casting is an electric welter (i have one up to 100 amp)
and you can move from 20A to 100A turnig an handgrip, it has also (very handy) a overheating switch in case you are using too much power.You just have to plug the two wires to your resistence and regulate the intensity as you wish.The only problem?..like always the price!
I paid mine about 60$ but i was lucky,you dont need a new or very good one so maybe a second hand will do the trick.Anyway it is always a nice tool for building your own stuff.

[megalomania]

While an oven element may be an initial first choice, the lack of flexibility of the elements rather limits their usefulness. They can be quite brittle if you try to unwind them, if that can even be done without some access to tools. They are so thick that they really don't provide the temperature necessary for many applications.

Now then, if one were going to make a round bottom type heater of a large size I would say they would be useful. I plan to make a rather large heater for a metal reactor that only needs to reach 120 degrees C thereabouts. An oven heating element in this case should work fine.