I'm not sure if this is the right place to post this, but since this is about 3d printing in metal, I thought I might put it here. I was wondering if anyone has ever come up with a method of producing forged quality parts with a 3d printer (besides printing and then forging them, which produces rough parts anyway.) I had an idea a little while ago, but I thought I would see if anyone has already come up with a solution to this problem first.
... I've made some small parts in overall sizes of millimeters with resolutions/accuracies of some microns with laser melting metal wires (Platinum and Nickel) with diameters of 10 to 50 microns.
With thicker wires you'll get bigger building volumes in the same fabbing time, but for bigger parts the internal stresses are more a problem too, so you have to temper/heat-cure the part while stacking ...
I always thought it would be cool to make an inductive levitating forge 3D printer. Maybe cut a small piece of wire, levitate until molten, drop accordingly. Maybe you could print and chill an encasing support material to help channel the droplets. i.e. could ABS hold the channels long enough for the droplets to cool, and then an acetone bath removes the ABS at the end?
I don't think you would want to mill after each drop. I don't see why you couldn't drop a 0.01mm droplet? The droplet would get the heat sucked out of it as soon as it hit the rest of the part. You might end up with holes and it wouldn't fuse together.
I was thinking that you would drop a blobby shape with the approximate area/volume and mill away to subtract. It is actually a subtractive process that deposits its own feedstock vs using plates and blocks. You might mill after depositing so many 'layers' or to create channels to guide additional material into the general shape.
Edited 1 time(s). Last edit at 09/04/2013 10:55AM by jason.fisher.
I've been thinking about another way that you could do metal forging in another new topic (I probably should have put it here). It involves using a semi porous "blank" that would be heated and then hammered into shape. Hell, you could probably attach a file to the toolhead and use that to shape the metal as well, yielding a full robotic blacksmith. [forums.reprap.org]
What's your idea AJ?
Edited 1 time(s). Last edit at 09/08/2013 09:56AM by dudesom.
Here's the idea that I wrote up. It's really long, since I spent some time on it:
3d Printer Forging:
The ongoing surge of interest in 3d printers has increased applications considerably for 3d printing, but this interest is held back by the fact that 3d printing, while cheaper than it was, is still limited to parts that are lower quality than the best parts being produced by conventional means. The best example is that of forged metals, which are not only considerably stronger than their 3d printed (milled-quality) counterparts, but can also be tempered, quenched, and treated in other ways to increase the strength of the basic metal even further.
The main factor in these limitations is that the basic processes for 3d printing have remained the same for the last 20 years: either extrude material in the cross-section of a part, and repeat the process in layers until the part is finished(video link), or lay down a bed of powder and melt (partially or fully) a cross-section of the part in a similar manner (video link). Another method also exists that solidifies the surface of a translucent liquid using light, again in the cross-section of a part (video link).
All of these processes make parts that are, by design, limited in molecular structure. They only melt or otherwise change the state of the building material just enough to produce a solid part; they do not do anything to enhance the intrinsic strength of the material. If this continues, certain things that are mass-produced will never be made on a 3d printer, regardless of cost, because the technology to 3d print these simply does not exist.
To overcome this problem, I think an ability needs to be added to 3d printing to strengthen or treat the building material for parts during the printing process. In layman’s terms, 3d printers need a way to forge, temper, quench, and otherwise treat metal parts. This would allow a 3d printer to have every advantage over conventional manufacturing except speed. More importantly, making forged-quality parts on a 3d printer might be cheaper than forging and later milling them, and they might even replace weaker parts for certain tasks. This becomes more important as costs come down. Metalicarap, for instance, is intended to cost under 10,000 euro and makes milled-quality metal parts.
To understand how my idea works, it’s best to understand forging, quenching, and tempering first. Forging is the act of applying pressure in a variety of ways on a very hot part (near its melting temperature) so that the internal structure (or grain) of the part follows its contours. Quenching and tempering is simply heating and cooling a part at a certain speed so that the metal forms a specific structure (or that a certain percent of it is a specific structure) at the molecular level. Both make the part a good deal stronger. Forging is usually done on a simple shape, and involves either:
(images for each of these)
2 dies that close just like a casting or mold, or:
One die that is flat and the other a certain shape, or:
2 flat dies, or:
2 dies that are in a shape, but don’t fully close the part.
Both put pressure on the part, so that the part’s internal structure deforms. To do this on a 3d printer such as metalicarap is hard, considering it is an electron beam melter (link). However, the printer’s method of operation looks a lot like casting (solidifying metal in a mold) at a tiny scale, blob of metal at a time. Can a similar thing be done with forging, tempering, or quenching?
With tempering, this looks like a certain “yes.” The process of tempering is just like melting, but with less power. After the system finishes melting a certain layer of powder into a cross-section, it will heat up the cross-section and temper it before laying down another layer. One problem with this is the possible heat dissipation into the layer below it, or worse, the layer separating from the layer beneath it because it is being tempered. One of these problems can solve the other. If the heat dissipation can be calculated (and predicted), then the electron beam gun can just heat one layer until the layer beneath it can be tempered at the proper temperature. This dissipation of heat helps bond the layer to the layers above and below it, and hopefully avoids destroying the tempering of these layers. Any layers above the layer being tempered will be tempered later. Long cooling times will be addressed in the later paragraphs.
When it comes to forging, it seems very hard to do on such a small scale, but it might be practical. Forging small “blobs” of metal one at a time is probably most similar to the old method of forging by hand, with many hammer blows on a part. The forging would probably involve a press that looks like a letter punch, but attached to a movable head, just like a dot matrix printer. In fact, any forging attachments would probably work like a dot matrix printer, pressing down (and deforming the structure of) metal in the form of the cross-section. After a 3d printer melts a cross-section of a part, the forging attachment would press down on each segment of the cross-section, forging it. Of course this would not have nearly the accuracy or tiny dot size of the electron beam gun, but it wouldn’t need to. It would just need to do what ordinary forging machines can do, which is producing (somewhat) forged quality parts of a given forging direction out of a 3d printer.
As for quenching, we can approach this problem by looking at the other solutions discussed. Since these, just like 3d printing itself, are done at a tiny scale one layer at a time, this is what I propose for quenching and cooling. A press like the forging press, but with liquid cooling, could draw off heat on one tiny area of the part at a time, for controlled quenching to go with the rest of the processes on a cross-section of a 3d printed part.
With this, a hypothetical 3d printer could melt, temper, forge, or quench any part, in any combination, layer by layer. It would, if it can be done, allow 3d printing to completely replace ordinary manufacturing, including forging. Basically, the idea would be to use something like a dot matrix printer to press down on and cold work the heated part of each layer after it has been melted by the electron beam gun. A similar system, but with a cooled plate that can move around, could do quenching, and the electron beam gun itself could temper metals.
That's all of it.
I'm not quite sure I understand how the other methods would produce a forged part. If a blob of metal was levitated, do you mean that you could drop the blob so fast that it essentially cold or hot works it instantly?
Could you do this with a solenoid-powered hammer and a blast of dry-ice vapor? I am picturing something that is like a 4-part multihead?
Does it make sense to print the outer shell of the object and temper it completely before depositing a solid infill? Would tempering the shell allow it to withstand higher temperatures than the material being deposited?
Could you print a thin layer of higher-temperature metal using weaker bonding non-vacuum processes (steel) as a shell that lets you deposit contiguous lower-temperature but higher strength (aluminum) material? i.e. printing a bathtub so you can fill it with jello.
Anyway, is there any existing system to produce forged quality parts on a 3d printer? If such a system exists today, then I would not need to develop this kind of system. (except for the quenching and tempering part, I'm pretty sure no one has done that yet on a 3d printer.)