Large Metal Additive with Weld Wire

[aarongough]

Very very cool! Would a simple cartesian system be easier to program/more economical than the UR robot?

Lots of fairly low cost CNC router table kits out there that seem like they'd be a great fit for this!

What about enclosing the machine and flooding the enclosure with shielding gas rather than constantly wasting it through the nozzle?

Very cool that you've just dived in and given it a shot!

 

[Strostkovy]

What would be the best heat treating process after this, and how much distortion do you think would result from stress relieving?

 

[Strostkovy]

Very very cool! Would a simple cartesian system be easier to program/more economical than the UR robot?

Lots of fairly low cost CNC router table kits out there that seem like they'd be a great fit for this!

What about enclosing the machine and flooding the enclosure with shielding gas rather than constantly wasting it through the nozzle?

Very cool that you've just dived in and given it a shot!

My only concern with flooded shielding gas is that steel doesn't weld very well with straight argon, and tends to benefit from some amount of CO2. I don't know enough of the details to know whether the CO2 gets consumed and would need to be repleneshed.

 

[JNieman]

Very very cool! Would a simple cartesian system be easier to program/more economical than the UR robot?

Lots of fairly low cost CNC router table kits out there that seem like they'd be a great fit for this!

For sure there would be cheaper ways to do this. The robot is part of a unit we sell, here, as we do automation solutions, with our biggest expertise being welding. We have a unit we keep around just for training and demonstration. So it was already available. That's as far as my thinking went before using it. It took me all over 10 minutes of being shown how to drive it from the controller to get to the point where I could make simple welding programs. Adding a z-offset to the Reference Frame meant that every loop was like starting a new layer.

After that got to be limiting, I began using Fusion 360 and now it's whatever I want.

What about enclosing the machine and flooding the enclosure with shielding gas rather than constantly wasting it through the nozzle?

Very cool that you've just dived in and given it a shot!

Some do that, flooding a sealed enclosure. I'd have to do more math on the economics of doing that vs the convenience of the nozzle. That's a lot of volume to replace every time you need to cycle. I kind of like the simplicity and accessibility of just using a welding torch as one would be used to. Given that this process lends itself towards large parts (much larger than my test pieces) I think a build chamber may be a bit limiting. I don't know though - I'd have to do the math on that. If anything, I thought about having a manifold of gas dump even more, trailing behind the torch, to keep the heated metal inert until it got much cooler, and thus less reactive.

Thanks. I've been reading about this kind of thing for a while. Didn't want to sit on the sidelines anymore.

 

[JNieman]

What would be the best heat treating process after this, and how much distortion do you think would result from stress relieving?

I haven't gotten far enough to dabble with the stress relief steps.

We have a Romer Arm with a laser scanner among our inspection tools, so once I do, I will DEFINITELY be taking some before-and-after scans of the part, to compare, though. Seems silly not to, since we have one. Once I do a large-enough part, I'll be sure to give it a shot. I don't think it would cost us too much. We have pretty regular traffic to our local heat treater for tool steel hardening and weldment stress relief. This'd just be another weldment, of sorts.

 

[JNieman]

My only concern with flooded shielding gas is that steel doesn't weld very well with straight argon, and tends to benefit from some amount of CO2. I don't know enough of the details to know whether the CO2 gets consumed and would need to be repleneshed.

I've read a fair few papers on the results of different gas mixes in Wire-DED. I have been using 90/10 lately but I've also used 85/15. Only mixes, no straight Argon. That's one variable I haven't personally experimented with, yet.

 

[Orange Vise]

A huge benefit of machining from solid is the ease of performing ultrasonic testing (UT) prior to final machining. Not only is it effective, but it's quite fast and cheap as long as the geometry is simple, e.g. round bar or flat plate material.

Once the part is made, you're left with X-ray and various surface scanning methods that won't catch internal defects the way UT scanning would/did.

Not a problem if the material was UT scanned prior to machining. Defects are unlikely to form internally during machining. Any defects created after UT scanning will likely show up in surface scanning, e.g. dry particle inspection.

Additive manufacturing generally precludes the ability to UT scan the material due to the complex geometry and lack of smooth surfaces - a big problem for mission critical components. Even if additive technology drastically improves in the next decade, it'll be awhile before the inspection techniques catch up.

 

[JNieman]

Additive manufacturing generally precludes the ability to UT scan the material due to the complex geometry and lack of smooth surfaces - a big problem for mission critical components. Even if additive technology drastically improves in the next decade, it'll be awhile before the inspection techniques catch up.

Right, for what I've seen, it looks like many organizations are choosing to certify a process, and then use that process for different shapes and final parts. Destructive testing of those final parts is still used to certify that process for /that part/ in a couple examples I've seen of metal additive flight parts.

I'll be honest, I haven't been diving into the guts of NDT of additive parts, but there's gobs written about it, and tons of lectures I haven't attended about it, that are geared towards metal additive parts. Typically this is related to powder sintered/fused/melted parts but obviously wire deposition / other DED is catching up, with folks like Relativity Space and Norsk Titanium making some pretty critical parts in such ways.

The customer base where I work now is not typically of that nature, so I haven't spent much time on it yet. We do a lot of automotive, agricultural, and some heavy equipment welding automation. As of now, I'd have no idea where we'd apply this technology, among existing or new customers, if we chose to pursue it.

That would certainly drive how deeply I need to look into the most thorough of NDT methods.