_GOTOBOTTOM
General Aircraft
This forum is for general aircraft modelling discussions.
Opinions on 3D printing
BigfootV
_VISITCOMMUNITY
Colorado, United States
Joined: December 24, 2005
KitMaker: 1,624 posts
AeroScale: 385 posts
Posted: Sunday, December 04, 2016 - 12:25 PM UTC
Hey gang,

My daughter has offered to buy me a 3d printer for a "X-mas" present.
I was wondering if anybody uses one for aircraft and which one?
Thoughts/opinions? Thanks.

See ya in the funnies..................
drabslab
_VISITCOMMUNITY
European Union
Joined: September 28, 2004
KitMaker: 2,186 posts
AeroScale: 1,587 posts
Posted: Tuesday, December 06, 2016 - 12:08 AM UTC
To own a 3-D printer is one thing, to know how to use the software to create the models needed to print something is a whole other story.

Next to that is quality. From an affordable printer don't expect anything near resin casting, or injection moulding. Very often the surface looks pretty rough.

If I would find the time to follow the necessary trainings to learn how to develop the 3d models I needed, I would prefer using one of the professional 3d printshops that are around. You send these your model, and they print and ship the final product to you. Some of these even give pretty good advice and have a webshop where other people can buy prints of your designs.

Finally, having a daughter that is prepared to fork out the money to buy you such X-mas present; man, that is the real gift, I am jealous.
pigsty
_VISITCOMMUNITY
United Kingdom
Joined: January 16, 2007
KitMaker: 1,226 posts
AeroScale: 640 posts
Posted: Wednesday, December 07, 2016 - 12:36 AM UTC
At the moment the technology is too immature to make affordable 3D printers really worth having. As drabslab points out, the mass market ones are a bit rough. This is because the resolution is too coarse for modelling standards. You can get printers that produce items that are indistinguishable from resin castings, but they cost £100,000.

My mate Tom has been slowly building a 1/32 Shackleton (I know, I know ...). A friend of his produced wheels and engines for it by scanning the 1/72 Airfix parts, enlarging them, and cleaning up the edges before going from bytes to resin. Put through the £100,000 printer, they're brilliant. They would also, he estimated, cost £500 if he charged what he should have done. And the market for 1/32 Shackleton wheels is a bit small to allow the cost to come down through volume.

It's like a lot of new gubbins: there's a lot of potential, but it's easy to believe the hype prematurely. Unfortunately, also like a lot of new gubbins, you have to wait for the early adopters to pave the way. Without those people, there will never be enough of a market to take them out of the machine shop. But they have to put up with high prices and first-generation quality.
BigfootV
_VISITCOMMUNITY
Colorado, United States
Joined: December 24, 2005
KitMaker: 1,624 posts
AeroScale: 385 posts
Posted: Wednesday, December 07, 2016 - 02:29 AM UTC
Thanks for replies guys.

Drab,

No need to be jealous Drab. LOL. I've talked her out of it for the moment until I do some more research on what's out there, also because of space issues.

I'm looking more towards the scanner/printer combo's. I know it's cheating. I'd like to start doing things like canopy frames, no glass, for starters. I do have prior CAD experience, but it's been years since I've done anything. The software back in the day was nothing like today.

Sean,

I agree that the printer's haven't caught up with the software and the one that have kept up with the software are way over the budget we've set. Now, when I hit the lottery, that will be a different story.
I've look at what Shapeways is doing with the 3D printing and some of the 1/35th armor stuff looks darn impressive. Again, their using high dollar gear and the prices aren't too high. About the same as if you were to buy it at your LHS. Would I trust them to do a simple thing like canopy frame and get charged $10.00 for it? No. Not worth it.

Oh god, I'm sounding like my Dad more and more.........

See ya in the funnies.........
KitKan
_VISITCOMMUNITY
Bangkok, Thailand / ไทย
Joined: November 20, 2016
KitMaker: 18 posts
AeroScale: 15 posts
Posted: Wednesday, December 07, 2016 - 11:42 AM UTC
Firstly, I would advise you to ask your daughter what type of 3D printer she is planning to buy. Without that info it will be very difficult to judge the qualitative aspects of the printed products.

Most laymen referring to 3D printer are describing a 3 axis robot with an extruder head for melted plastic (typically PLA or ABS) acting as a 4th axis. These typically cost between USD 400 to USD 5000. A good example of this type would be the very ubiquitous RepRap based devices. Personally I have found that realistically you cannot get a better resolution than 0.05mm on the horizontal plane and 0.17 for the vertical plane on these, even with a highly optimised system (temperature controlled room, fixed frame, good timing belt). For market systems the resolution is about 0.1mm/0.3mm for reasonable print speed and ease of maintenance. Vertical plane accuracy is very poor due to the fact you are printing in layers, and these layers act as a limiting factor in vertical resolution. Printed product will generally need to be filed/acetone sprayed to generate a smooth finish..

Resin printers are much pricier (the one I played with while studying at imperial cost GBP 200k iirc). The optimal resolution of these (as far as my inexperienced model maker eyes can see) is comparable to many of the molded plastic parts that come with the model kits. Resolution a typical resin printers are realistically in the hundreds of microns. They are much harder to maintain than the plastic printers however, and their print substrates can be rather expensive. I am also not 100% sure how compatible these resin parts are to common modelling chemicals, though I think thinner should be fine (iirc the main content is Isopropyl Alcohol).

While this varies depending on person, I find making 3D objects is not too hard. Solidwork I find is a very good program for beginners (It took me 2 months of self-education to make 3D objects proficiently using solidwork, starting with only very basic knowledge of 2D CAD).
xrz100
_VISITCOMMUNITY
Berlin, Germany
Joined: May 28, 2010
KitMaker: 193 posts
AeroScale: 2 posts
Posted: Wednesday, December 07, 2016 - 12:23 PM UTC
Hello Brian,

Modelling quality can be achieved only with sintering printers. Those melt acrylic powder to generate the profile of the model, than another 0.1 mm layer of powder is given and the laser starts melting again. This is the current best to buy quality in print shops, but seldom affordable for consumers (I don’t know your daughter). Further quality and material is not mature yet, therefore by outsourcing the printing job to a print shop you leverage the current technological developments with variable cost.


For my 5 cent I am using Shapeways, for my modellship models mainly in 1:35 scale, as you build planes this is and information not an advertisement ;-): http://www.shapeways.com/shops/iron-models

I have at shapeways also my shop where I sell my designed product to other people building the same model. This shop is free of charge. When you sell, shapeways get the production cost and you get the mark up (whatever you have defined). The minimum thickness is 0,3 mm with the Utra Frosted Material. Shapeways offers a wide portfolio on extra materials including steel, silver, brass porcelain.. Gold (just something you can design for your daughter).

Learning how to draw in 2D took me 4 hours of watching youtube tutorials for Rhino, the program I am using. That was 3 years ago and now I can almost handle all shapes that are coming across my ship models. It is important that you learn by the challenges you face when realizing your model – no challenge, no learning effect and the knowledge is gone within a few weeks.

3D printing enables you to realize every model you though about and where no kit is existing currently. I am currently buildinga Torpoedoot in 1:35 sclae where nearly no parts are available in the market (besides figures) but the plain superstructure will be done through milling parts, all others like 105 mm gun, 3 tube torpedolafette, Rangefinders, funnels are generated through 3D printing making this a unique and precise model – one of its kind.

Hope this helps

Bis denne

Christian
KitKan
_VISITCOMMUNITY
Bangkok, Thailand / ไทย
Joined: November 20, 2016
KitMaker: 18 posts
AeroScale: 15 posts
Posted: Wednesday, December 07, 2016 - 12:37 PM UTC
Christian actually raises a really good point, why not CnC milling?

Resolution for these are generally excellent with crisp/brittle plastics like PVC, and both easier to maintain and generally cheaper than resin 3D printers. It is also much easier to acclimatise into for people with experience in 2D CAD.

CnC milling cant generate 3D shapes easily, if at all though (and where it can, this is achieved with very finicky rotations and readjustments).
BigfootV
_VISITCOMMUNITY
Colorado, United States
Joined: December 24, 2005
KitMaker: 1,624 posts
AeroScale: 385 posts
Posted: Wednesday, December 07, 2016 - 12:37 PM UTC
Hello Kittichan,

The printer would be a gift from my daughter to myself for X-mas.
You do bring very good points to this discussion. One printer/scanner that caught my eye is AIO Robotics, Zeus. However, this is a little above the budget. Cost retail is about $2,500 US.
So we keep looking and researching.
The goal would be to make some extra cash on the side with it by offering canopy frames in 1/72, 48, 32 scales. Scratch building them is a bit of pain.

Watch this, now someone is going to run with idea. Just my luck. Me and my big mouth.

See ya in the funnies.............
BigfootV
_VISITCOMMUNITY
Colorado, United States
Joined: December 24, 2005
KitMaker: 1,624 posts
AeroScale: 385 posts
Posted: Wednesday, December 07, 2016 - 12:52 PM UTC
Hello Christian,

Just caught your post.
CNC would be an option, but I think the material loss would be high just to do a glass free canopy frame. I'll have into that before committing to anything.
Thanks for the reply.

See ya in the funnies...............
KitKan
_VISITCOMMUNITY
Bangkok, Thailand / ไทย
Joined: November 20, 2016
KitMaker: 18 posts
AeroScale: 15 posts
Posted: Wednesday, December 07, 2016 - 01:33 PM UTC
Greetings Brian,

I took a quick look at the 3D printer you are interested in, and I see 2 key plusses. 1) It has a built in scanner 2) It can use generic filaments (this is going to massively decrease cost of operation on the long run).

On the flip side this platform is otherwise very expensive for what it does. The resolution really isnt very good, its print areas are small, no temperature controlled print chamber, and it has no heated bed (no ABS printing). In terms of raw printing ability I would suspect this machine is rather less good than a USD 700 or so customized printing machine.

Do not be fooled by the claim that it contains an internalized control processor is actually a plus. Sure, this is more powerful than the Atmel 32 you get in a generic reprap, but that is why you have a USB attached computer to do the heavy lifting in such setup. The quad core 1.7 of the control system on the 3D printer is actually pretty weak, and no match whatsoever compared to any modern home PC, and definitely massively less flexible. Remember that you will need a PC to do 3D design using CAD anyway. In any case, the vast majority of people I have met who 3D print to any moderate scale uses a PC-3D printer setup.

Honestly, consider building a 3D printer from a kit, someone who is skilled enough to build a model plane would find the construction trivial, and such kits tend to be highly flexible for modification and optimization. Making the machine yourself also gives you a good understanding of machine maintenance, which is essential for maintaining any degree of print quality.

Go for this one, I used it in gradschool and it was great:
https://reprappro.com/shop/reprap-kits/mendel-full-kit/

Its very easy to mod, and can generally be fixed with off the shelf parts. Also only costs about USD 500
xrz100
_VISITCOMMUNITY
Berlin, Germany
Joined: May 28, 2010
KitMaker: 193 posts
AeroScale: 2 posts
Posted: Wednesday, December 07, 2016 - 02:21 PM UTC
Brian, this is a Z-Axis printer, which cannot print unsupported structures. Easy example, I f you scan yourself standing on the junction and stretching your arms left and right and you look like a T. The printer would start printing support material starting from the hip upward. Because it needs some material when printing your arms. The printer cannot all of the sudden put liquid filament in the free space!

Similar proceeding would be applied when he is printining you canopy, the base is easy, but as the ribs go more vertical in the arch it will require some support structure under this arch.

Spapeways printer would od this in one piece, as the powder bed provides the necessary support, which is later during cleaning blown away.

bis denne

Christian
gaborka
_VISITCOMMUNITY
Borsod-Abauj-Zemblen, Hungary
Joined: October 09, 2005
KitMaker: 626 posts
AeroScale: 264 posts
Posted: Wednesday, December 07, 2016 - 02:36 PM UTC
I think the quality minimum that is acceptable for our scale modelling purposes is the Formlabs 3d printer which still costs around 3500,- USD plus the material you use.

If your daughter is ready to present you with such a gear... have a go:) I would not hesitate:)



KitKan
_VISITCOMMUNITY
Bangkok, Thailand / ไทย
Joined: November 20, 2016
KitMaker: 18 posts
AeroScale: 15 posts
Posted: Wednesday, December 07, 2016 - 03:21 PM UTC
Greetings Christian,

While that is theoretically correct, that is not strictly true in practice. Most slicer softwares (these are software which convert 3D files like .stl format into Gcode) contains packages which causes the 3D printer to create "scaffolding" which are thin plastic structures which support parts projecting horizontally from the main body. Once print is completed these scaffolding could then be excised from the main body.

This is definitely an option for computer based slicer system, whether the "smart" proprietory software allows this or not is another matter completely.
drabslab
_VISITCOMMUNITY
European Union
Joined: September 28, 2004
KitMaker: 2,186 posts
AeroScale: 1,587 posts
Posted: Wednesday, December 07, 2016 - 05:55 PM UTC
I have no doubt that 3d scanning / printing will fundamentally change or hobby, and not only that.

For my work I recently met with a company producing high quality industrial powders for 3D printing. The statistics of the amount of powders they sell is exponentially growing year by year, and the same for the mechanical properties that they achieve. I am not only talking plastic here, but also titanium and aluminium based powders. In certain configurations the 3D printed parts are already achieving the same mechanical properties as a part casted a more traditional way in the same material.

Quite promising but also potentially quite schocking for our society. Labour used to shift from high income to low income countries. Now it seems that work is simply dissapearing completely.

Back to modelling.

If today I would have the opportunity to do something related to 3D printing, I would certainly concentrate first on making a design in pofessional software. Buying a printer seems not effective to me for three reasons:

> because of the expense (if quality is a requirement, and in our hobby it is)
> because of the very fast evolution of the printers, whatever you buy will be outdated in a few months time
> because there are 3D printing companies offering high quality prints at a price you will not be able to compete with.

On the other hand, any design you make now remains usable for any future generation of printers.

In the above, I don't know about 3D scanning as I have no view on the quality that is achieved here. It may be a good option to scan, improve/adapt the model using good software and then have it printed elsewhere.
MikeyBugs95
_VISITCOMMUNITY
New York, United States
Joined: May 27, 2013
KitMaker: 2,210 posts
AeroScale: 63 posts
Posted: Tuesday, December 13, 2016 - 07:31 AM UTC
I just want to chip in a bit here. There are 'affordable' desktop printers out there. Some from the likes of XYZPrinting with their DaVinci line. The problem with these printers for us isn't necessarily the limited capabilities (but that's one big reason) but the method of printing.

Most of these affordable desktop printers use Fused Deposition Modeling or FDM. It's also known as Fused Filament Fabrication (FFF) or Plastic Jet Printing (PJP). All of these methods are the same and are only named differently to get around trademarks and copyrights. The FDM process heats a plastic filament up to its melting point and extrudes it through a small nozzle onto a build platform. The technology is limited by the simple fact that it's extruding a line of plastic - the nozzle needs to be large enough to allow plastic to flow. The smallest print resolution (how small the bead of plastic is, not the detail resolution) I've seen is around 0.1mm. The detail resolution we seek just isn't there. The most egregious limiting factor for us, though, is that this technology will also leave visible stepping unless processed after printing. You can bathe it in acetone or some other solvent to smooth out the roughness but detail can be lost.

Sintering printers bring you closer to modeling quality but not necessarily all the way there. Although they are good for large prints (at least from Shapeways).

The best printing technologies for modeling are DLP/SLA and MJM/MJP. DLP, Digital Light Processing, and SLA, Stereolithography, are very similar technologies differing really only in the projection method. DLP uses a projector to project successive layer images onto a vat of a liquid photopolymer. This cures the polymer and the build plate moves up slightly while a second thin layer of the polymer is laid down. SLA differs only in that it uses a laser to cure the polymer. As you might have figured, SLA takes more time than DLP but the results can be pretty similar. DLP printers can have higher detail resolution and better XY accuracy because it projects an entire layer slice instead of a single laser dot. A very good quality about DLP/SLA printers is that the printed surfaces come out very smooth. Nearly as smooth as injection molded plastic. You can still feel the layers if you run your fingernail over the surface but it's not very noticeable. On the downside, the models require printed supports. These need to be removed following printing and thus can mar the surface.

MJM/MJP (multijet modeling/printing) are the same technologies but, again, with different names. They're also called Polyjet and Material Jetting. They work similar to desktop inkjet printer but with a liquid photopolymer instead of ink. The printer lays down a very, very thin (around 16 microns on up; 1 micron = 0.001mm) layer of photopolymer and support material and cures the polymer with a flash of UV light. The accuracy and detail resolution of MJM/MJP printing is very high. This process can render details smaller than 0.1mm. This process will result in stepping on angled or curved surfaces but, depending on the layer height used, might not be noticeable. Printed supports aren't needed because it uses a solvable or meltable support material.

Unfortunately, MJM/MJP is out of reach for all but the wealthiest of modelers. The machines cost a fortune, they require constant upkeep and the materials are also very expensive. DLP/SLA are actually well within reach of modelers, though. Some average quality DLP/SLA printers cost from $800 to a few thousand. Many of these have average resolutions with minimum feature sizes being over 0.2mm and layer height over 30 microns but there are few that cost between $2k-$4k with pretty nice specs. Unfortunately you won't really find many with a 16 micron or less layer height and minimum feature size of less 0.1mm within that price range but there are one or two. The DLP/SLA desktop printers tend to be a bit more user friendly with less maintenance being needed thanks to the lack of a dedicated heater/extruder head. But the materials used tend to also be rather expensive. Many printers require the use of proprietary materials in company or printer specific cartridges. There are some printers that can use open-source or non-proprietary materials. Two of the more common consumer SLA printers are Formlabs Form 1 and Form 2. These are solid printers but print with a 25 micron or higher layer height. For all intents and purposes though, 25 microns is good enough. I think they also require proprietary cartridges.

A general rule of thumb to remember is the higher the print quality, the higher the price.
BigfootV
_VISITCOMMUNITY
Colorado, United States
Joined: December 24, 2005
KitMaker: 1,624 posts
AeroScale: 385 posts
Posted: Tuesday, December 13, 2016 - 09:26 AM UTC
Hello Michael,

I've been popping in on your forum over at Armorama concerning this topic. Your input is very important a long with all the others.
I'm still researching these systems. For now, we'll just wait.

If my daughter and I had the budget today, the AIO Zeus system would be a start. But as you stated, the high quality print, the higher the price.
That system seems to be a good starting point for what I would like to start doing in my opinion.

Thanks for the reply.

See ya in the funnies..........

 _GOTOTOP