Progress on (library) research on the metal deposition head/the other metal printer

Okay so it look like the problem at this point boils down to this:
We need to find a way to:
1. Deposit metal onto a metal object without changing the ultimate shape of be object in ways that are too difficult to predict to achieve the desired tolerance (IT 5 I think probably though that needs more research. We want enough to print hydrostatic and hydrodynamic bearings basically that is the most demanding thing on the wanted object list to print I think in terms of tolerance.) It's "not too hard to predict" rather than "does not change the shape beyond tolerance" since if the shape changes can be predicted practically, they can be compensated for so they are not a problem.
AND
1a. The material has reasonably close to the desired properties. That means reasonably low internal stresses and reasonable metallurgical properties (like ductility strength and porosity etc.). This is ideally achieved by allowing your choice of alloy to be used and getting it's usual standard properties as those tend to be the best.
OR
2. The ability to form the metal all at once in a highly precise casting operation with e.g. a water soluble cast.


Basically right now it looks like the only practical methods are welding or "microcasting" with layer by layer treatment or even as the bead is being laid down to release the stress produced by the thermal expansion, OR a form of highly precise casting might be doable. So either use a welding method like tig and just put one bead on top of another basically. Microcasting is just dropping droplets (like 1/8 of an inch diameter or smaller) onto the metal object, and it turns out it builds up good metal. Some sort of post treatment of the metal you just laid down is a must though. The problem is that the metal is deposited as a liquid, then it cools to the melting point. One, the thermal expansion of nearby solid metal can cause enough stress in the rest of the object to cause permanent distortion. Two, and worse, the metal after cooling to the melting point obviously keeps cooling and when it contracts pulls on the nearby metal, leaving tensile stresses. These build up in the object additively.

Treatment options include:
1. Ultrasonic vibrations, "ultrasonic peening" or "ultrasonic impact treatment". This can relieve only like 30% of the stress probably.
2. Vibrational treatment using lower frequencies. This is interesting and is about as good as annealing, but takes half an hour. Still could work maybe.
3. Electromagnetic peening. Basically a powerful electromagnet, discharge a capacitor through it and it's like a hammer hitting the surface of the metal. This compresses the metal you just deposited, reducing the tensile stresses (the material was being stretched) by a lot, could be 90% or more, even leaving compressive stress instead with a powerful enough electro whack.
4. Thermal tensioning. It's supposedly now known exactly how this works but basically it uses a thermal gradient to cause distortion in the opposite way, so that when the deposited metal shrinks it can shrink freely as the nearby metal returns to it's normal shape it ends up accommodating the stress.


The next most interesting is nickel vapor deposition which as the name implies only works with nickel. What happens here is that nickel actually converts to nickel carbonyl, a gas the same as in the mond process [en.wikipedia.org] and then deposits back as nickel metal at like 250 degrees C. So you could print a mold in the exact shape of your part, then deposit nickel on the mold, either until you have a solid nickel part, or just a shell of nickel which could then be filled with another metal alloy like a cast. Then you would get a sort of nickel plated metal part out of it. The main problem is the shape changes which occur when you do the casting thing. If they are too hard to predict due to thermal distortion or phase transition (when the metal solidifies there are several different crystal forms it can take so it may take some work to predict the final shape or it may not be doable to a very good precision) then you would not be able to make precise parts.

The third most interesting is to make a sintered metal object, then infiltrate it with nickel using NVD, or electrochemically (just like electroplating) which means you get more choice of metals and even some alloys. The problem with this is doing something about the voids that might be left.

no interest?

What would probably drive interest is to include articles.