How long until we see 3D printed components in boat building?

Well I had hoped that folks like you and me (with friends in the CAD and 3D printing industries) could bring these fantasies down to earth a bit. A DLMS with precision mirror alignment just isn't gonna sit on anything smaller than a Carrier. (ok maybe a cruiser) it just makes no sense.

 

Ancient dude,

Its about time you put your money where your mouth is.
Get something built, use it, and make a report.

Note I'm not asking you to put a machine in your living room.
Nor on your 100' yacht.

The only thing of use on any forum is hard won information. You are ignoring that, and have none of your own.

Show me wrong. Send me a PM if you can. I'm going to drop this thread.

 

Baltic,

I am glad you have friends in the 3d printing business. I work in it. I am in house council for a metal fabrication company that currently uses SLS to print titanium parts. Including parts being used for the marine industry.

I think you would be hard pressed to find a 50' boat these days without at least 1kw of power, heck my old 50' had a 13kw generator installed, and my current 38' had a 8kw until we ripped it out, but we still have a 2kw gas portable.

And you don't need anyone on board who knows how to operate RHINO you just need someone who knows how to load a cad file from the computer library to the printer. It isn't rocket science, it's almost point and click.

You are right that the machine is the tricky part, and finding one that is capable of living thru rough weather, and getting pounded will be a challenge. But this is a solvable issue.

 

1kw is fine, you need more like 3kw dedicated to this device and it has to be 3 phase 240 volts. THAT is unusual in a genset.. And as inhouse counsel I would not want to go up against you in court, but in engineering I think you are out of your depth.

If you are going for a rare part you need someone on board who runs Rhino, cuz the MFG sure aint giving you the toolpaths. So you have to roll your own. You do not get the library of parts.

So not only is the machine the tricky part, so is its size (1m cubed ain't gonna cut it as a DLMS sintering device) nor are you going to get certified parts that your mfg will stand by, nor are you going to really be able to produce the part in less than a few days.

btw by friends, I mean folks I did business with directly in both 3D printing and CAD, some of whom are off creating a whole new parametric engine to integrate solids and NURB style lines.

Its a field I was directly active in for over 6 years technically and still am invited to speak at conferences.

And while it is solvable to accomplish what you see - the OPEx cost is prohibitive - so it won't happen

 

I thought you were going to leave, Baltic. I really am, we're just repeating ourselves. Bet you repeated yourself more than I did . . .

 

While this is true for a simple part, if you are making something that has to interface with something else, you need to look at fits and tolerances with the mating hardware. Just having a scan of a part doesn't give you what you need to know. Between the tolerances of the original part and the tolerance of the process it is easy to make something that won't fit, or is too loose to work. While you can make something that looks like the original part, it may or may not fit and work.

For that reason, you are seriously underestimating what it takes to engineer a part if it is more than a paperweight. The more complex the part and the more mating surfaces you have the longer it takes to do the reverse engineering.

We have done very high technology turbine parts that have demonstrated the limits of additive manufacturing (high pressure turbine nozzle for the Dept of Energy) and the process works, but there are limits as to what you can do with it.

Parts made with the process using certain metals can be "HIP'ed" or hot isostatic pressed and after that process and appropriate heat treatment we were getting very close to forged properties.

To hold very close tolerances (as we do in turbomachinery), you do generally have to add some material on those surfaces and go back and do a skim cut to insure those surfaces are in the right place.

Lastly, the process is, at this point and time, hideously expensive. While it is a neat thing to do, the reality is that, if you have any other way to make a part, it's probably less expensive to make it that way than by additive manufacturing. Perhaps that's a bit of an overstatement, and maybe it isn't that bad, but we just spent over $25K for a 10 inch diameter turbine vane set that had some intricate cooling passages and literally hundreds of tiny holes in the part. It came out beautifully, but if I could have made it any other way it sure as heck would have been less expensive.

 

Yellow jacket,

I really doubt that replacing major engine components is going to be done in the near term for exacally the reasons my mention, but you could have the entire McMaster Carr parts catalog available. Sure if you need a new cam shaft that is going to have to be flown in, but if you need hose clamps, brackets, nuts, bolts, ect...

If that type of thing is enough to justify the purchase or not would of course be a seperate issue.

But the moment I can find a metal printing machine for ~$2,000 or less I plan on buying it to put in Jamaica. Instead of having to bring down hundreds of dollars of stainless parts because the maintenance guy can't figure out what size is needed I can just print off exactly what I need. Not to mention the problems in dealing with non-available parts... Like the 12 1/4 stainless bolts I paid $1,000 for over Christmas, because I had to have a courier fly them in from Florida.

 

If you need new hose clamps you pull them out of stores - its a lot cheaper to have them in stores than to have the CAD jockey on staff. Even fairly simple things like the screws on a hose clamp need to be reasonably precise to get them to work well (ever tried to tighten a cheaper one? PITA).

Non-available parts are the ones that need a CAD jockey to mill. if your SS bolts are going to hold load that cargo or lives depend on, I doubt your insurer will give you coverage for 3D printed parts of unproven load capacity. And given that everything from the actual toolpath being used to the quality of the sintering material and even how stable the DLMS machine's base was in turn affects the quality and hence strength of the part - good luck getting any insurer to buy off on it

 

>We haven't seen a single part put on an aircraft in spite of 5 years development and lots of corporate support.<


Then just call EADS , the article claims they are currently flying with sintered TI parts TODAY!

There is even their photo of the parts being used.

 

All good points. If you were to make nuts and bolts this way you would also have some that the nuts were loose, some tight and some that might not work at all, either not running down, or stripping easily. If you wanted to make a cleat or a fitting, you could and it would work fine, but it will cost 20 times what it would if you simply cast the part by conventional processes.

As I said, if you have to make something that you can't make by any other process, then this is a great technology. But I doubt that you are going to see a machine that costs only $2,000 that will be able to accurately sinter metals any time in the near future. Time may change that, or it may not, and I don't have a crystal ball that will let me look that far into the future.

The reason that most consumer electronics (things like laser DVD players) that were once very expensive have come down in price is the ability to manufacture the key components (the laser stuff) in very high quantities. I'm not sure that every household or even every business needs a machine to make metal parts even if it is relatively inexpensive. Harbor Freight sells a lot of $200 welding machines, but there isn't one in everybody's garage.

A final note is that this technology also requires very very clean raw material and the material has to be specially processed into a powder form. This stuff is not going to be inexpensive, most likely ever.

If there is even a tiny speck of dirt in the material powder, that dirt becomes a flaw in the part and could lead to catastrophic failure if the part is loaded like a perfect part. Those issues can be avoided by HIP'ping the parts, and by performing x ray inspection, but that additional quality control jacks up the cost even more.

We made parts that simply couldn't be made with conventional methods. The parts came out beautifully, we were very happy with the results, and after HIP and X ray we determined that they were acceptable for use in a turbine engine. But the parts cost $25k each. A similar investment casting, if it was possible to make, would have cost about $4k. The premium for using this process was about 6 X that of conventional processes. It was worth it for us, and in the future I am confident that the cost will come down, but I doubt that it's going to ever be competitive with conventional higher production manufacturing methods.

We are looking at making an aluminum housing for a small waterjet by this process. Compared to an aluminum casting the cost for a "one off" prototype is less expensive than tooling and making one or two by this method is probably fine, but if you want to make more than a couple of pieces you are far better off biting the bullet and buying tooling and making the part by conventional processes. For that reason, again, unless you can't make it any other way, you won't see this process used for production hardware.

 

This article from Gizmag is very relevant to some of the posts in this thread:
http://www.gizmag.com/tornado-3d-printed-components-flight/30334/

 

Winner! Winner! Chicken Dinner!

spot on WRT everything. the big cost drivers in this are input materials and the MECHANICS of the controls. And what brings down costs of mass consumer products is the consistent and continual replacement of mechanics with software and integrated silicon.

Where the CD player of the 1980s had discrete circuits for the amps, the noise filters, the LED laser the mirror controls for that LED Laser, today all of those are integrated into a single chip.

But when you need 4 degrees of motion freedom with micrometer accuracy, that is not something you get without cost.

 

This is exactly what I was talking about. Of the crazy amount of bits and pieces it takes to keep things running a large number of them can be produced with 3D printing. And the cost of tooling, low volume production, storage, tracking, expedited delivery and downtime with markups at each step adds up to more than the cost and time to print as needed. The next step is that the military will require all the parts they buy to be evaluated and validation tested for 3D printing and will warehouse stl and slice files instead of parts. Corporations will follow.

 

Be careful as to what you see in places like Gizmag and Popular Science...

These guys love to tweak the imagination and predict that we will all be driving cars like the "Jetsons" had.. Never mind that there's a thing called physics that tends to limit what can really be done...

Nowhere in that article did they say what the part actually being used in flight did. There are a lot of pieces on an airplane that are "non-structural". Pieces like covers and heat shields aren't structural pieces. That means that you can readily make them by printing and if they have flaws in them or aren't perfect then it's no big deal. As I said, for structural pieces you have to HIP them to properly consolidate a structural part. That isn't something that you can do in your garage. It takes a large high temperature oven that can be pressurized to very high pressures. Nobody is saying that these pieces that are used in the aircraft aren't processed further after they are printed either.

This is a wonderful technology that may, in time, become competitive with conventional processes for a select small quantity of pieces that are very difficult to make by any other means. Right now it isn't even close to competitive in terms of cost, and, due to the high cost of the processed raw material, may never be. The advantage of producing a near net shape part, and making parts that are impossible to make otherwise, will continually increase the market penetration of the technology, but when it comes to true mass production conventional casting and machining processes beat it every time.

 

http://www.3ders.org/articles/20130...st-3d-printed-titanium-fighter-component.html

looks machined - maybe for finish - or its just bs.

Either way. I see that it can really be popular for specialty applications - I consider jet fighters pretty special. But printing common hardware? fat chance. People also here seem to dismiss that its not just pieces of metal that we need. A simple plumbing component can be combination of nylon or urethane seal, several metals, maybe steel webbing, crimped connections etc. Going from printed geometric parts to functioning entities is not that simple.
And the idea that one machine can print a piece with all kinds of materials nicely in place (think a faucet or something complex like an electronic devise) is pure sci-fi for several decades and even if possible in the future the physics will dictate it to special situations where cost is not the 1st factor (usually it is).

and fully agree with grain of salt with gizmag etc.

They stretch things so far and take them out of contexts that its not even funny - they also publish any kind of stupid news release for new inventions. Check the wind generator concepts over past years - anyone with an ounce of sens sees that the concepts constantly released are pure nonsense and only exist to collect investments.