Sciencemadness Discussion Board - Electrical furnace elements : Pt or Ta wire instead of Kanthal ?

Electrical furnace elements : Pt or Ta wire instead of Kanthal ?

metalresearcher - 5-10-2010 at 12:48

For some purposes I have an electrical resistance furnace which uses Kanthal wires and it can heat up to 1200oC. I have build the furnace myself, but I see cheap Ta wire on ebay.

Tantalum has a very high melting point (2996oC) and seems to have a good resistance to oxidation at high temp. Is it feasible to make an electrical furnace using Ta wires ?

dann2 - 5-10-2010 at 14:45

I think you need a reducing atmosphere or at least inert if you are going to put the temperature up above red heat to stop oxidation.
I often wondered would Propane or Butane do as a non oxidizing gas in very high temp. ovens. You would have to vent the gas carefully and make sure the oven was purged of Oxygen before you started turning on the electrical supply.

One (a bit exotic but OTC) element for an extremly high temp. oven that I have pondered on was Silicon Carbide in the form of sharpening stones (from hardware store). They are totally non conducting when cold. When they go to red heat they are very conductive. You can pass a massive current through them with a welder if you heat them to red heat first. They can achieve extreme temperatures.

Dann2

not_important - 5-10-2010 at 16:11

Tantalum starts absorbing oxygen around 500 C, the oxygen diffusing into as well as reacting with the metal. This affects the bulk physical properties, and may cause the wire to weaken and/or change resistance. The reaction rate goes up with increasing temperature, and eventually volatility of the oxide will lead to excessive mass loss.

Problem with using hydrocarbons as non-oxidising atmospheres is that they are not inert. They crack to carbon, which reacts with many metals when hot resulting in changes in volume, conductivity, and strength.

SiC is used to make resistance elements for heaters. When quite hot it reacts with oxygen, the SiO2 skin formed protects the bulk of the heater to some extent. The high resistance when cold, the negative Tc of resistance (conducts better as it gets hot, meaning it pulls more current), and the relative fragility restrict the use of SiC.

watson.fawkes - 5-10-2010 at 16:45

Quote: Originally posted by not_important

Tantalum starts absorbing oxygen around 500 C, the oxygen diffusing into as well as reacting with the metal.
[...]
SiC is used to make resistance elements for heaters. When quite hot it reacts with oxygen, the SiO2 skin formed protects the bulk of the heater to some extent. The high resistance when cold, the negative Tc of resistance (conducts better as it gets hot, meaning it pulls more current), and the relative fragility restrict the use of SiC.

To elaborate: The engineering behind metal heater wires is an alloy whose oxide passivates the metal surface. Only certain metals and alloys do this, and it's well researched. The metal also has to be refractory, which is why aluminum is unsuitable. Indeed, using aluminum as a heater element is how you create aluminum vapor in a sputtering chamber.

SiC has all the practical difficulties mentioned above, plus some others. It has be run with a transformer, as a rule, since it's mechanically stable aren't wire-gauge size and have relatively low resistance. So you need a big and heavy and thus expensive transformer (lots of copper). SiC also loses a fair bit of mechanical strength as it heats, which mean that it requires mechanical support. After a certain point you just can't use it on the ceiling. Wall mount can require embedding it into a relief in a refractory. Third, thermal cycling isn't very kind to them, as the oxide layer grows inward along crystal boundaries, which means it's much better suited to a continuous industrial process than a short-batch oven.

Another material, better than SiC in many respects except for, of course, price, is molybdenum disilicide MoSi2. It's got much better aging characteristics than SiC. They also require a transformer. Duralite sells them, which whom I've got no connection other than as a customer of their resistance wires.

metalresearcher - 5-10-2010 at 22:47

Quote: Originally posted by watson.fawkes

Duralite sells them, which whom I've got no connection other than as a customer of their resistance wires.

This looks very similar to Kanthal Super.

dann2 - 6-10-2010 at 04:14

Hello,

Guess I've been stone walled on my sharpening stone idea

Some info. from Procedures in Experimental Physics. This book should be available somewhere.

Dann2

Attachment: piep.zip (178kB)
This file has been downloaded 477 times

peach - 6-10-2010 at 10:15

I Squared R, for all your direct to mains, industrial SiC and MolyD element requirements.

They're not super duper cheap. And you'll need to read the PDF's on using them. The Moly D's higher temperature comes with a price beyond the initial upfront cost.

I'd suggest you look at Nichrome, Kanthal and the other nickel / chrome wires first.

Some of these elements / wires don't like being in direct contact with the refractory.

Two many posts asking the same thing, note the splattering effect in the replies.

[Edited on 6-10-2010 by peach]

metalresearcher - 6-10-2010 at 12:40

I did already several experiments with Kanthal wire with varying success.
www.metallab.net/kanthal

peach - 6-10-2010 at 23:16

Nice page!

With the Ta and Pt wires, I think strength may be a problem, as I'm guessing they're thin if they're cheap. Also, if they're thin, they won't be able to output as much heat energy so things may take a while to warm up if you're doing anything above small test melts.

How much is cheap? That's not a joke either. I mean, if they're still quite expensive by regular wire terms, you may be better going with one of those I^2R elements. Those are designed specifically for high temperature oven and resistance furnaces and to be very easy to implement, both mechanically and electrically.

There are other companies selling them who make them in coils that'd be the ideal size to fit a cup sized crucible (or slightly bigger) in.

I^2R does similar helical elements, but they're designed more for tube furnaces and tend to be pretty long for their diameter, a few feet for an ID of an inch or three.

The guides on the I^2R site are very good. They talk you through all the details in a readable way, how to mount them, what temperatures they can and can't work at, inert atmospheres, currents, methods of powering and controlling them and so on.

I think it's the MolyD elements that, whilst they can work at very high temperatures, do not like being run below 800C, which creates a massive gap in the spectrum where a lot of chemistry works. They're more for specifically 1000C+ work; melting steels, titanium, other metals, looks at ores, minerals and the likes.

The silicon carbides don't go as hot, but they're also much more forgiving, cheaper and will produce a continuous spectrum in the temperatures a lot of other chemistry is interested in.

[Edited on 7-10-2010 by peach]

metalresearcher - 7-10-2010 at 03:55

@peach:

I checked I2R's page they sell Kanthal (cloned?) elements the MoSi2 are very similar to Kanthal Super (which also use MoSi2). But can I order from them as an individual outside the US ? They sell nice quartz tubes as well.

On ebay I found Kanthal A1 wire for $35 for 30 meters @ 1mm thickness and Ta wire for $10 per meter @ 0.5mm thickness. So I think heating 0.5mm Ta wire to , say, 2000oC will let it oxidize quickly.
Kanthal wire can be heated to 1350oC for a longer time and resists cycling.

The Starbars of I2R look nice.

watson.fawkes - 7-10-2010 at 07:18

Quote: Originally posted by dann2

Guess I've been stone walled on my sharpening stone idea

It's worth trying, if you're interested. Just know what the limitations are, and be aware when you're pressing the boundaries of the material.

peach - 7-10-2010 at 23:37

They do.

I squared R is like porn for anyone who likes fires and melting things.

As to getting them, I'm not sure. They MUST have distributors. And there will be foundries ordering them. If they won't sell direct to you, one way is to ask them if they can put you on to one of their bulk customers, who might be able to pass one on to you.

That's the same for quite a few things, like metals. Quite often only a small bit is needed, but the yards only sell it in huge sticks or plates, for hundreds a go. If you ask about who's buying it, they'll almost always have an offcuts bin that's going back to the foundry for recycling, which can yield the bits you're after.

I raided a skip at one point outside a place doing duct and conduit work with aluminium, for some bits to weld. There can be quite big lumps in there that they can't make use of. The scrap value for them isn't a lot off what they paid for them, because it's still pure material (not mixed with coolant from machining).

I asked first. The first guy said probably not, since it's still worth quite a lot. But then he asked the guy who owned the company and he said it was fine to take some for free.

I've also had machine shops do bits of work for free, provided it's very simple, won't take them long and they don't think you're going to sell it.

Actually, there is an unspoken price, passing on the attitude.

There are a huge number of guys producing SiC elements. A lot of them are in China or India, where a lot of foundry related gear is produced. But getting a single unit out of those guys may be tricky, as they tend to deal in numbers measured by the freight container load.

Here's a link to one site I found that did coil elements which look great for small crucibles or things like your sodium work. They sell in single units, and produce just about every combination of element / insulation, you'd ever want. From tiny fibre optic furnaces the size of a coin, to elements prewrapped for lab use, to industrial scaled things. But I think the prices may be higher than normal.

I also found these on ebay. You could make something similar yourself, but that's a prepacked, tested, simple option. Just twirl a potentiometer to change the power through the element.

Made in Germany too, so hopefully that means it's tricky to break and built reasonably well.

In terms of the Ta, I'm not entirely sure how that's going to behave. I have a suspicion you'd be better off putting the money towards a SiC or MolyD, especially if it fails and then you end up buying one anyway.

Coiled MolyD elements from MHI

[Edited on 8-10-2010 by peach]

metalresearcher - 8-10-2010 at 05:09

@peach:
Thanks !
Such a power controlling device I have which has a max load of 2kVa I bought it for $30 in a local electronics shop.
Here the picture:

http://www.metallab.net/kanthal/index.php?index=9

It works fine even when I plug a welder into it and use it as a transformer for Kanthal wire (600W) and it also works in a larger (1800W) Kanthal furnace directly connected to the 220V mains.

And the mhi-inc.com site looks interesting !

peach - 8-10-2010 at 07:08

That looks super duper for this kind of stuff. So easy and small, yet lots of controllable juice with the slide of a dial.

The one I linked to goes on a heat sink to the do the really high wattages. Which is understandable, that's a lot of power.

I wonder how those elements are made. They're obviously silicon carbide, which isn't too hard to get. I did a quick patent dig;

Here's one talking about making strips of silicon carbide to increase surface area and decrease wasted volume

Does anyone know the fusing temperatures, or if someone has designed a method to fuse it at lower temperatures? The only mention of heating I've seen is the use of a "laser bean", all letters as per the patent.

It decomposes on melting, right up around 2700C. So it can't be a melt process without an appropriate atmosphere.

I'd be surprised if someone hadn't come up with an easier method, as melting under special atmospheres at those temperatures is not going to be easy at all in terms of making them cost effective.

[Edited on 8-10-2010 by peach]

Contrabasso - 11-10-2010 at 11:19

For a different job I've been looking at Hot Surface Igniters from www.crystal-technica.com
mail@crystal-technica.com
t +44 (0) 1626 214780
They have USA dealers too!

Look at their website and look at some of the big ones.

These things draw a high current from ambient to 1200c then self regulate, they are robust enough to live in flames permanently and they can be used as glow plugs in the cylinders of Diesel engines so they should be robust enough and some go to 200w.

peach - 12-10-2010 at 00:52

They're nifty looking.

The one I'm looking at is 400-500W.

Only thing that bothers me about them is how well they spread the heat out. If you're heating a flask or crucible, it may be a bit annoying to have one side considerably closer to the element than the other.

Have you got any prices for them though?

I think this thread could do with some investigation of that being done.

If those coily elements are a few hundred each, that's pushing it too far for at home use, for me anyway. In that case, I'd probably just use nichrome for sub 1000C chemistry and a flame for melting crucibles of metal.

Metal Researcher is in the unlucky position of needing the control those element provide above 1000C to get his melts hot enough but not also melt the containers

{edit}I'll email MHI and do the investigations on them. If any of the US guys could call the dealers for the other elements that'd be good. Virgin signed me up to a "super do anything!" package. There's a picture of a fiber on the leaflet, and they told me I'd be getting one. No fiber. And no 20MB speed from the test servers, let alone actual websites. The phone bills have been gigantic as well, and I don't even use the bastard.

[Edited on 12-10-2010 by peach]

metalresearcher - 15-10-2010 at 05:39

What about this ?

http://cgi.ebay.com/Platinum-Iridium-20-002-wire-x-10-Ten-Fe...

This is (too) thin 0.05mm Pt-Ir wire. I am actually looking for Ir wire which can be used as a (relatively) "cheap" alternative to MoSi2. I mean "cheap" as Ir costs less than half of Pt per gram and MoSi2 is

- brutally expensive
- VERY vulnerable (particularly under 1000oC).

peach - 16-10-2010 at 14:06

I don't think it's that vulnerable provided you use an inter atmosphere or vacuum. You could really be doing with some of that going on given what you're interested in making and the temperatures involved, reasonable tight control over the atmosphere.

That wire is very thin.

The trick to making use of such wire would be to make the coil very dense and long, so a low amp supply will still output a lot of heat, just at a higher voltage. Breaking it would be very easy as well, though.

{edit}MHI still haven't replied, but I did email them towards the end of the week. I have emailed them before, and they never replied. "Yeah! Lets make a website! Then not answer the emails!". Teases.

[Edited on 16-10-2010 by peach]

Wnat about molybdenum or tungsten wire ?

metalresearcher - 29-10-2010 at 11:06

For making a really hot electric furnace I found Mo and W wire on ebay:

Mo: http://cgi.ebay.com/Pure-Molybdenum-Wire-0-020-5mm-x-10ft-3m...

W: http://cgi.ebay.com/Tungsten-Wire-019-48mm-x-10-3m-99-95-Pur...

And prices are reasonable.
How does this metal behave @ 1600-2000oC ?

dann2 - 29-10-2010 at 12:56

Would the following electrodes be any use in a garage furnace?

The materials are reasonable easy to get but the final temperatures not that high and you might need fancy control gear.

Dann2

Attachment: US2490825.pdf (191kB)
This file has been downloaded 630 times

peach - 30-10-2010 at 06:22

Tungsten will disappear fairly soon, that's why light bulbs, thermionic valves and TIG welders have inert gas shields and vacuums.

Molybdenum will oxidize at high temperatures.

I have to have a bit of a complain here, after you saying you knew what you were doing with molten sodium. It isn't at all hard to check these things and tungsten is an obvious no go for roasting things in the open atmosphere, as is sodium. This is why I kept complaining about trying to produce molten sodium without the oil or inert gas. That is a hugely reactive metal, even at room temperature.

There is a reason the guys producing these commercially use things like silicides or carbides, it's to try and avoid the element burning out and to make it structurally strong enough that it won't turn into cooked pasta over 1000C.

You will seriously be better off buying one that has already been designed to do what you want it to. If there was a better way to do it, the guys making thousands of tons of steel and the others would be using it, as they need every penny back that they can get to make any money on it.

If you want to use these elemental metals at huge temperatures, you need inert gas or a vacuum. And, fine current control & support for hair thin wires. That's just unavoidable fact given the physics and chemistry at work, and has nothing much to do with design ability.

{edit}Predictably, MHI have not replied. They did this last time as well. It needs someone in the US to give them a phone.

[Edited on 30-10-2010 by peach]

metalresearcher - 30-10-2010 at 07:12

@peach: I prefer the MoSi2 elements like MHI offers. A look at Mo and W was just by 'lack of better' (MoSi2 or SiC elements) and I already expected it that these elements are unsuitable.

One guy told me on this forum that a SiC garden grinder tool can be used after preheating it making it conductive.
I tried it to heat to red hot ut did not conduct.
So MoSi2 is the only option.

watson.fawkes - 30-10-2010 at 07:35

Quote: Originally posted by dann2

Would the following electrodes be any use in a garage furnace?

The materials are reasonable easy to get but the final temperatures not that high and you might need fancy control gear.

These electrodes seems specifically designed for electrical resistance heating of glass melts. You run AC directly through molten glass, which itself forms the resistive element. The problem they're solving is the corrosive action of molten glass on the electrode.

watson.fawkes - 30-10-2010 at 07:39

Quote: Originally posted by peach

Tungsten will disappear fairly soon, that's why light bulbs, thermionic valves and TIG welders have inert gas shields and vacuums.

Molybdenum will oxidize at high temperatures.

Check and check. Both of these metals are used for resistance heating in inert atmospheres (generally argon), reducing ones (generally hydrogen), and vacuum. The book on vacuum tube construction I've got specifically mentions molybdenum wire as a resistive heating element for hydrogen brazing of vacuum parts. If there's oxygen around, it's treating your resistive element as a consumable.

peach - 30-10-2010 at 07:54

Also keep in mind that MoS2 elements get picky about their atmosphere and the way you run them.

They have a glaze of oxide over the element. If they're run at low temperatures too early on, they'll fail. If the atmosphere is neutral or reducing, the coating may also fail.

The silicon carbides are the weapon of choice for ease. So you really need to need those MoS2 temperatures for it to be worth their cost and the additional care needed with them.

The maximum temperatures of the elements are given for inert / vacuum use. The temperatures for use in air are lower. If the atmosphere is reducing, that'll strip the glaze off them.

I Squared R has pdf's all about their use and are very easy on the eye and understandable compared with most datasheets.

I expect the grinding grade of carbides is likely coated, contaminated with something or a binder of some description is used in some of them, so it may be brand specific as to which ones will function as elements. The grainyness of grinding discs will also massively increase their resistance, meaning not a lot of current will flow for the same voltage, and it's the current that heats the element.

For really high temperatures, you're basically going with a flame for none / low reactive metals or elements for precision, reactivity and control. But you'll want vacuums or inert gas for those.

A vacuum furnace really needs casting to make the lining continuous, so it's inert gases. Nitrogen will react with a fair few things at high temperatures, so that means Argon.

[Edited on 30-10-2010 by peach]

peach - 30-10-2010 at 08:11

Can't remember if I posted this, but it looks amazing for the lower temperature chemistry. 480C inside a thermos that'll go on your desk

Stick a coil of borosilicate or quartz in there that spirals round and round and you have a tiny gas reactor. Or an appropriately sized test tube / beaker would work for solid phases.

[Edited on 30-10-2010 by peach]