Fail Of The Week: This Inanimate Titanium Rod

You saw [Chris] cast aluminium on the cheap using Kinetic Sand a few weeks ago, didn’t you? He recently got his meaty hands on some titanium through the magic of modern transactional methods and was bowled over by its strength, hardness, and poor heat transfer.

He thought he would cast it into a nice, strong bottle opener. As you can probably guess, that didn’t go so well. First off, it wasn’t easy to saw through the thin rod. Once he did get it split in twain, it was surprisingly cool to the touch except at the tip. This is nasty foreshadowing, no?

[Chris] takes a moment to help us absorb the gravity of what he’s about to do, which of course is to send several hundred amps through that poor rod using a DC arc welder. Special precautions are necessary due to the reaction between oxygen and heated titanium. His trusty graphite crucible is grounded to the bottom of a big aluminium tub, and a cozy blanket of argon from a TIG welder will shield the titanium from burnination.

Well . . . the titanium didn’t melt. Furthermore, the crucible is toast. On the up side, vise-enabled cross-sectional examination of the crucible proved that there was still gold in them there walls.

Do you have any (constructive, on-topic) suggestions for [Chris]? Let him know below.


2013-09-05-Hackaday-Fail-tips-tileFail of the Week is a Hackaday column which runs every Wednesday. Help keep the fun rolling by writing about your past failures and sending us a link to the story — or sending in links to fail write ups you find in your Internet travels.

69 thoughts on “Fail Of The Week: This Inanimate Titanium Rod

  1. Wow, there are actually some constructive comments on youtube!?!? For those too lazy to go and look, his main problem was the titanium reacting with the graphite to form titanium carbide. I think induction heating will be the way to go. Pretty impressive fail though – who tries to cast titanium in their garage?

    1. fwiw titanium forges fine at room temp, although it has limited annealing cycles (and hammers just bounce off of it.) It doesn’t actually forge much better at heat, either, or at least not at heats that make sense for hand or small-process trip/steam hammer forging.
      Vacuum (or inert gas) melted and cast, yes.

    1. Thermite can come in a variety of qualities. A quick search shows vastly different temperatures from 3000 to 6000. Titanium gives it up at 4000, so it would require quite good thermite in large amounts to melt it. Problem is, then all you would get is a gross mix of iron and titanium. Nevermind the crappiness of that test, mind the gap!

  2. Awesome video. It only took Lockheed a year or so and a ton of double-secret CIA money to figure out how to work with Titanium. :) I’ve heard the same things as you, probably, that to cast Titanium you just need a very evenly hot environment and you need to keep it hot.

        1. Military spending in general was a problem for the Soviets. Peak US spending between the end of WW2 and the fall of the USSR was in 1985ish, amounting to about 7% of the GDP. Soviet spending peaked in about 1982 and was about *14%* of their GDP.

          Suffice it to say the second world war did weird things to the country’s psyche.

  3. Consulting various internet sites I have a theory that a major part of the problem was the carbon crucible. Titanium forms titanium carbide aka TiC, if it is heated with carbon. Thus using a graphite crucible could be a problem unless you want TiC a brittle material with nearly 2x the melting point.

    Consulting the materiel properties:
    Ti – melting 1668c
    TiC – melting 3160 c
    Carbon sublimates at 3642c

    Reference sites :
    http://www.chemguide.co.uk/inorganic/extraction/titanium.html
    Wikipedia

    Just my 2 cents

    1. The only metal not reacting with titanium is copper. But copper melts before Titanium. Bad luck. Dont’t start with water cooling, it is extremely dangerous. (they have meter thic concrete walls in there!)
      Small quantities of Ti are melted in graphite crucibles, but by induction heating within some seconds to prevent too much reaction wit the carbon. High vaccum (10^-5 milibars) is often used. The Ti is then instantly poured in the mold, wich also has to withstand 1500°C and the reactive titanium, otherwise you have the same effect as pouring molten metal in a wet mold. No Fun!
      One other solution for the crucible is a yttria coated one. Expensive and one time use.
      Industries tries to find econmic solutions for titanium casting for quite a while now.

    1. Ahh, the wonders of the internets. Twenty years ago we would have all given up at the first sign of trouble, shrugged in ignorance, and *shudder* gone outside to do something else.

      To think the barbaric times we once lived in.

  4. The problem with this particular experiment is noted in comments, but I’ll add that as much as I have enjoyed [Chris]’s commentary over the years, I’m glad that he and Colin Furze are separated by an ocean.

  5. It sucks when you bring all you have to bear on a project, and come up empty on it. I have to admit that dross slug he showed at the end was certainly an interesting result though. Makes me respect the people who managed to make the titanium stock he started out with much more. As an interesting aside Canada has had far better luck working with titanium than the USA has. Canada was making a titanium fighter jet until we talked them out of it.

    http://en.wikipedia.org/wiki/Avro_Canada_CF-105_Arrow

    Notice how quickly the project came to an end. It was going so well for a while there too.

    Oh, and Miller owns Lincoln Electric today too. Or at least they did the last time I checked. Companies change hands so fast these days it is hard to keep up.

    1. Miller does not, and has never owned Lincoln. itw welding owns miller and ITW welding and Lincoln are competitors.

      sources: my father is a senior executive of itw welding, and i’m a welding engineer that has worked for miller in the past.

    2. Want to clarify, since pcf11 (unintentionally, I assume) makes it sound like they were pressured into dropping their plans for political reasons.

      What actually happened as per that wiki article was Canada joined with NORAD and the US started sharing its intel, which led Canada to reallocate priorities away from aircraft designed to defend against bombers and in favor of nuke-tipped anti-air missiles. Now, some of the intel was wrong (most notably that whole “missile gap” thing) but it was, at least going off the available evidence, wrong in good faith.

  6. the method for heating sounds pretty archaic. how does he expect to control the temperature? magic? a torch isn’t gonna cut it. a temperature controlled furnace might just do it if he melts to 3300º. if it reaches 3200 isn’t gonna melt. if it passes 3600 it happens what we saw. ofc induction is the way to go. kinda puzzles me why didn’t he saw that before. maybe he’s just busy trying to get youtube views instead of doing proper stuff.

  7. Here’s a method. Seems to be several, but I’ve read that even Argon can cause problems with cast Titanium. I haven’t tried this, but the method seems valid.
    This site (I have a serious jonesing going on for their hammers) shows hhow to do it using lost wax investment casts. They then coat that in a silicate slurry, put that in a kiln to cure the mold and burn out the wax. Then they put that mold in a vacuum and use an arc furnance to melt the Titanium an pour it in the mold.Titanium Investment Casting: http://youtu.be/WXFRRg8YMT0

  8. I’ve been trying to melt titanium safely for a while and haven’t had any more luck. It’s interesting stuff to work with. I built a spotwelder and thought it’d be fun to make welded ti chainmail. That didn’t work so well: the combination of thermal expansion and lack of thermal conductivity exceeded the metal’s yield strength and it would burst in layers like an onion. (Only much smaller.) I also found out that it’s a bad idea to use someone else’s expensive bandsaw to try to cut it up, and then got to buy my friend a new blade.

    1. I was wondering why he didn’t take the pipe direct from the gas cylinder regulator and just let it free flow, ideally near/over the crucible rather than dribbling out of a TIG torch in the corner of the box, away from the action.

      Anyone who’s welded in a slight draught will know how quickly gas dissipates in moving air, just from the end of the torch shroud to the tip, having the gas 12″ away from the subject seems very optimistic when you’re moving stuff around.

  9. Nice experiment. Ceramic crucible and induction heating would be the cheapest way in trying to solve the carbide problem. It quite hard to cast titanium even in industry. Casting a chunk and machining it is prefered way

        1. Touchy much? Calm down kemosabe.

          Are these two examples financially feasable? Can we replicate this without spending hundreds of thousands of whatever currency you deal in?

          In the first video the info link is dead. I would love to know what energies they had to apply to keep the molten Ti levitated.

          In the second video the machinery looks enormous (and expensive). This would put this kind of melting out of reach for most hobbiest (remember this is what we are trying to discuss).

  10. I use an argon arc furnace on a regular basis (research scientist in materials engineering) to melt Ti and so I can confidently say that yes, you can melt Ti with an electrode setup and no, it does not react with argon.

    Just briefly for those interested, the setup we use is composed of a sealed chamber with an electrode inserted inside which you can then manipulate externally.The chamber is first pumped down to a vacuum and then filled with argon under low pressure. The Ti is located on a water cooled Cu ‘hearth’ used as a ‘getter’ to react with any remaining oxygen in the chamber before we move on to melt our samples. The molten metal does not stick to the hearth due to the difference in temperature. The electrodes are water cooled tungsten rods and can operate at currents up to 800A, although only a fraction of that is required to melt a small Ti sample. The device I use is something similar to the following link, but about 40 years old and hand made.

    http://www.hhv.in/argon-arc-melting-furnace,

    Just to be clear, the above setup is great for melting and mix high temperature alloys, but not much good for creating a large amount of molten metal, which is what Chris is trying to achieve. This is because the Cu hearth cools the sample very quickly and so make the current requirements for melting a large amount of sample enormous.

    1. I had found one study that looked at the porosity of Ti cast in an Argon atmosphere vs vacuum, there were some mentions of problems with Argon listed at other sites. Atmospheric. Pressure in an Argon atmosphere would be easier to do, unless you’re on the space shuttle.
      What is your opinion on the “silicate slurry” the hammer company uses to make a mold? Do you think that is glass or a simple ceramic slip. It stays white opaque after firing and looks like a vitrified ceramic so my vote is with the ceramic slip, maybe a porcelain slip.

  11. From a fuel forging stand point:
    The REAL issue is the sheer, raw, HEAT required for this.
    Titanium melts (and therefore becomes malleable) at temperatures far surpassing most readily available materials.

    Most well built forges/crucibles are constructed from a very dense metal (iron/machine steel).

    I’m not very familiar with temperature related attributes of ceramics such as firebricks, but most containers will melt before the titanium exhibits any significant change to its physical properties.

    In addition, not many fuels are able to burn with such intensity, even oxygenated, and those that ARE capable of producing that heat start to decline in cost effectiveness.

    Lastly, the unique chemical properties of titanium under extreme heat make this quite a feat and balancing act.

    Bringing it to heat at all is difficult, let alone bringing titanium to heat under the necessary conditions.

    I suspect that a heating method which does not rely on physical consumption of a fuel or direct contact with the work piece will be your best bet. Induction heating makes sense here.

    The second major hurdle is the container. It must retain its form and structure when subjected to differences in temperature in excess of 3k degf. over less than a second. Tungsten might be ok… diamond? who has a DIAMOND CRUCIBLE?!?!? still… maybe kiln fired ceramic? or would that shatter… dunno…

    1. Alumina (AlO2), a component of firebrick, melts at 2072 C. Ti at 1668 C. Alumina is also used as a “kiln wash” to protect shelves from glaze and keep glass from sticking to surfaces. It’s available from most ceramic supply stores. I wonder if that could be used to coat the investment cast.

  12. Use a molybdenum crucible & electrode, melt Ti with a low Mo content, some of the Mo crucible will get into the Ti, it could be possible to target an already known Ti Alloy which has more Mo (kind of “contamination engineered alloy”).

    On cutting Ti, it’s react damn fast with Air (O2/N2) and it’s bad thermal conductivity get problem with heat build up, soldering chips on the saw’s tooth and other…
    That why machining Ti, high pressure (35bar) and high flow coolant system are used, the part is completely over flood, covered with coolant. (reminds me of a coworker getting drenched in seconds)

    1. Molybdenum =~44$ for 1kg of material. that’s going to be a expensive for a crucible.
      Can’t find prices on graphite… wkipedia says that forms of graphite are thermally conductive, while others are used for high temp thermal insulation. not sure about melting temp…

  13. I worked with a guy who installed an industrial Ti smelter in Japan (I think it processed ore into base metal). It used induction. 2000V @ 2000A! Conveniently built next to a nuclear power station.

Leave a Reply

Please be kind and respectful to help make the comments section excellent. (Comment Policy)

This site uses Akismet to reduce spam. Learn how your comment data is processed.