Direct Metal Laser Sintering

[JakeAdkins]

Has anyone used DMLS for production parts? I think I may have found an ideal application for this technology.

We do a lot of hydraulic manifolds, so they are low load (about 5000 PSI MAX) and have a complicated geometry. It would also change the way the manifolds are designed. It would make them more efficient and smaller. (because the connecting ports don't need to be drilled out, they connect where they need to)

I have not been able to find anyone that can support production runs. Everyone I have spoken to concentrates on prototyping but says the product is production ready.

 

[Jabberwocky]

We certainly don't use the technology for production, but it's pretty great for form and fit in metal. One thing to keep in mind is the material properties, even if they tell you it can do 316 it's not going to be necessarily equivalent to a machined part. Another thing to keep in mind is surface roughness - by it's nature the parts will never be particularly smooth - they often require lots of secondary operations.

 

[ajack1]

I would agree with Jabberwocky in that the finished result is slightly porous and the surface finish is not great, however for manifolds I have seen a process where they basically force an abrasive gel through the part that gives good cheap results.

I would have thought the biggest problem for production was the cost; this is still an expensive process for any production parts. Having said all that the cost and quality keep improving with every new generation of machine.

I have no idea where in the world you are Jake but in the UK I doubt you would have any problems getting production runs done, basically they just want the machines running 24/7, maybe you should spread you search wider?

 

[HDS]

the prototyping emphasis probably comes from costs. We had the same thoughts you did until we realized the part prices would be 100X a machined part from the one vendor we spoke with. The vendor did say they were doing production work for aerospace applications that could not be made any other way.

Have no idea if the cost would be lower if the machine was brought in house.

 

[tripleZ]

Is this something that could be powder metal injection molded (MIM or PIM depending on who you talk to)? The MIM process is capable of fairly complex geometry, although the size of the part can put limitations on this manufacturing methodology. You do run into some problems with materials as well as we generally aren't keen on MIMing tool steel type material (D2, M2, etc.). It all really depends.

If you're looking at 316, I'll second Jabberwocky's comment that it will not be the same. Specifically your corrosion resistance will be significantly lower than wrought grade. There will be remnant porosity because you're not liquid phase sintering the product...I believe it's more akin to high temp sintering.

The cost to the process is prohibitive just because of the time it takes to "print" the prototype. It's used primarily for prototyping as it simulates the material and doesn't require us to build the mold tooling. I'm guessing that's why it hasn't gone mainstream yet. It's more of a sales tool than a production tool. Once prototyped, a multi-cavity mold and sintering furnace will outproduce the "printing" machine (for production quantity parts that is).

 

[IRstuff]

Aside from the issue of comparing efficacies of the treatments, anytime you're doing a single part at a time, you're going to run into cost issues. So cost cannot be a consideration here.


Performance, however, particularly something that cannot be achieved otherwise, is a different matter. However, one can reasonably argue that someone who designs their part to not require the unique processing might have a cost advantage with little or no performance penalty.

TTFN

FAQ731-376

 

[JakeAdkins]

The problem with MIM or PIM is that the complicated geometry is internal. The external geometry of all most all or our manifolds is very simple (a rectangular solid) An additive process like DMLS would eliminate the need to have intersecting bores that need to be plugged later.

The idea was to take advantage of the build properties. Since it is build in layers you could have internal twists, turns, spirals, or what ever else you wanted but could never actually build using traditional machining methods.

I am under the impression that the DMLS process produces fully dense parts, unlike other metal RP prcesses.

I have spoken with several people that have used this process to build manifolds (prototype only, and did not take advantage of the unique build properties)

I got a couple of quotes as well as talked with a guy at my office that bought one of these machines for the last company he worked for and the only way it would work is if we actually bought a machine for ourselves. It would cost us the same as we spend now for machining and it would only work on the complicated manifolds. Simple manifolds would be cheaper to machine.

The problem is that hydraulic manifolds tend to be very large for this process. Maybe when the build time speeds up it would be more reasonable.

 

[mighoser]

I've used DMLS to make gears. The OD spur teeth were surpising good as far as profile tolerance. I opted to wire EDM the internal spline. They were made from 300M in the as produced condition. I've had parts made from 17-4 alloy. Try Linear Mold Inc. 734-422-6060.

They have an EOS DMLS machine.

 

[bklauba]

Another DMLS is RPM & Associates, Rapid City, SD, phone number (605) 348-0555. They have an Optomec DMLS that can manufacture parts of at least 4 feet (perhaps 6 ft) high.

BK