Rules with Straightening Titanium

Gut feel tells me it would be better to bend it cold. 4mm is a rather small bend. I use to straighten motorcycle fork tubes with much greater bends to within .1 mm of being straight. We would straighten them cold to avoid changing the metallurgy with heat. New camshafts for automobile engines are often straightened in automated straightening presses after heat treating but prior to final polishing.

Of course you need to bend the shaft beyond straight to cure the bend, but how far? For things like crankshafts, I would put dial indicators on the crank at several points, and keep track of how far I was bending it when it would just yield. (This would require approaching the yield point in increments -- the several or more elastic bends are nerve wracking, but do not fatigue the metal in any serious way) Then, once the yield point was found, I'd add the desired (remaining) correction to the deflection at yield and press onward to that total, always fearing that I'd press too far (and have to press it back the other way, work hardening the piece in the process).

Some titanium alloys are pretty soft, so you'd want to make sure that the v blocks and ram fitting are softer. The press needs to be pretty hefty relative to the thing straightened, or else its own flex makes measuring the bending action more difficult.

That's about all I can think of -- and it is probably not enough, given the probable cost of a replacement shaft. Hopefully others will reply with more specifics. If you poked around a shipyard or two, you might find someone who is good at straightening propeller shafts, rudder shafts, etc.

Hi, I work in an aircraft maintenance facility and I am used to titanium parts repair processes.

I agree with the coleague. Just do not heat titanium. It is too much reactive to do so without a high vacuum or highly controlled atmosphere.

In your case, it seems that it doesn't pay to do so, just go cold forming. And, of course, if you are supposed to have a high reliability component, it's always good to perform a crack inspection after the final work - I am pretty sure you ca find someone to do it for you near your home.

If, after all the work you need a more precise shaft alignment, consider machining the work diameters and/or flanges (only the specific geometry, not all the shaft of course!). And, do not forget: if it is suposed to be installed in a rotor, balance it!

DO NOT TRY HEAT out side of a controlled environment an environment with NO Oxygen!! First the thermal conductivity of Titanium is close to zero. Second Titanium's favor element in the whole universe is oxygen, above 750°C titanium will absorb all the oxygen it comes in contact with. Titanium oxide is very brittle and initiates cracks.

To bend you need to establish where to apply the load or loads and how to hold the part where the reacting loads will be applied. If you can give the design to a stress engineer. show the engineer the shape and the desired shape you can get some help.

The other thing you need to do is get a load cell or force gage of some kind to know how much load you are applying. You will need a series of dial indicators so you can watch the part move as you load it. As you apply the load the part will bend elastically. Elastic bending will fully recover to the original position (less any initial offset in the supports). You will want to find the elastic limit by plotting the applied load against the deflection, hence the reason for the dial indicators. Once you reach elastic limit the part will bend plasticly. The dial indicators will move but the load will remain near constant. Once you start plastic deformation (bending) continue until you have moved the amount of the initial amount you desire to straighten.

Be patient!!! If possible sent the part out to have it stress relieved.

Forming May be hot or cold formed using hydropress, power brake, stretch or drop hammer methods. Similar in characteristics to 300 series stainless steels.

Heat Treatment Solution Treating--heat to 871-927 C(1600-1700 F), hold for 15-20 minutes and water quench

Cold Working The cold work characteristics of this material is similar to that of a moderately tempered austenitic stainless steel. In multiple cold forming operations, intermediate stress relief is recommended to prevent tearing or other material damage. Post-work annealing is required to reattain optimum performance characteristics.

Good luck

Hey thank you all for your help! Sometimes when you're heading into a job (especially one of the cost and possible nightmarish after effect of this one) it pays to spend a little time doing your homework first and you guys have helped me a lot. So thanks again!!

Cheers everybody.

Where is this shaft used?

Its used as an agitator shaft for the autoclave in a local gold mine here in Otago NZ.

This is a dumb question but I am curious, how the hell did the bend the darn thing in the first place?

PS I am interested in the ultimate outcome of you straightening attempts so pleas post your results successful or other wise.

I'm not sure but these shafts are under a lot of load and are constantly pounded inside the autoclave. They get a lot of build up around them with chunks of hard ore going through there as well. The shafts are 2.5 to 3m long, so maybe one too many big build ups and the whip action of the shaft? I'm guessing here!

Thanks for the interest I'll try and get some photos up here once it's back up and running.

Cheers

I'm guessing that the shaft is like this, fully cantilevered from the flange. In that case, you would want to make sure you put the shaft between centers and indicate the flange. A small deflection there could make a relatively large one out at the impeller. But in any case, I'd assume the speed is 100 rpm or less, so imbalance is not likely to be a large issue. Also, if the shaft ends up a little s-shaped, but with the impeller center close to in line with the extended centerline of the gearbox, you are probably fine. The uneven torques induced by the varying consistency of the slurry being mixed are probably much higher than variations caused by being a millimeter or two out of line.

(The picture is from mixingsolutions.com and is not used with permission, but I suspect they'd be happy to have the exposure.)