3D printed "shell" molds and process for hobbyist - experiments - RC Groups
Shop our Airplanes Products Drone Products Sales
Thread Tools
Oct 14, 2017, 03:15 AM
Registered User
Discussion

3D printed "shell" molds and process for hobbyist - experiments


Hello,

I would like to investigate the potential of 3D printed "shell" molds and process for hobbyist. At least do few experiments.

The intention is to find an alternative to the proven vacuum bagging techniques, with more accuracy and still keeping costs and labor reasonable. The technique should be suitable for one off builds, or few builds, which is quite different from (even small) manufacturers.

I won't develop here, but CNC finally does not appear to me an easy path in that scope.

I discovered the concept of shell molds in Stratasys white papers dans design guides.
http://blog.stratasys.com/2017/05/25...osite-tooling/
http://www.stratasys.com/landing/composite-tooling

The shell is placed inside the vacuum bag, which makes that the stress on the shell focussed on the thickness of the shell. Not bad. Neither need for vacuum tight mold, as all the mold is in the bag.

I designed a piece for a first segment of a SynerJ ALES wing, and I'm currently printing it.

Crunching some numbers: one side mold for first 20 cm of the wing would represent 150g of PLA and 12 hours printing, for a 3mm thick shell and 50% infill.

Building the full wing would be equivalent to 24 segments like those ones (6 segments for Left wing bottom x4 for top and right)
=> 3,6 kg of PLA approximatly 72€
=> 12 days printing.

Cost OK, but very long printing. However, it is not me working, my printer is one of the cheapest and this may be optimized.

JMF
Sign up now
to remove ads between posts
Oct 14, 2017, 03:25 AM
Registered User
As of the building process, as CNC is not in the game here, the easiest path could be to go to hollow wing:
- no need of foam cores,
- I imagine that you should not need too much vacuum to make the half wing sandwich with balsa or thin foam
- glueing the two halves together doest not need again too much pressure, which would be compatible with PLA molds.
- no expensive or difficult to source stuff needed (laminate could be carbon / balsa / glass)

Key point to address: surface finish of the mold. Yes, I know. Different options:
- epoxy filler + sanding (labor)
- adhesive film as proposed in the Stratasys design book, like Tooltec CS5 (this one seems difficult to find here)?
- vacuum bag a glass layer with mylar on top ?
- other options ?

Notes: PLA mold won't like temperature. Keep curing below 50-60°C

Comments and ideas welcomed :-)

JMF
Oct 14, 2017, 07:17 AM
myday.. myday
LA8PV's Avatar
I think this is interesting, what surprised me when it comes to 3D printing and CNC carving is the time that it takes. I have a nice CNC router myself and when I got it I had big plans of carving molds and such, but I realized that if I wanted to carve molds or positives with the advocated accuracy that these tools can achieve that would take enormous amounts of hours to do. Considered in the manual of my machine it is advised to do a partial service every 10 hours, that would be quite some work involved too with lots of interrupts.

Personally I like to create and build myself and I like everything that I can actually achieve with my own hands, so I mostly use my CNC machine for routing out parts and such in 2D. Also I like to avoid using molds if I can because I only use them once and then they just takes up a lot of valuable space in my workshop or in the trash bin.

Anyway I think things like this 3D printer is the future, because kids nowadays don't start out with piece of wood and a knife, they start out on a computer. It will be interesting to see your findings on how you can use this as an aid for making composite models or parts.
Oct 15, 2017, 04:55 AM
Registered User
Quote:
Originally Posted by LA8PV
I think this is interesting, what surprised me when it comes to 3D printing and CNC carving is the time that it takes. I have a nice CNC router myself and when I got it I had big plans of carving molds and such, but I realized that if I wanted to carve molds or positives with the advocated accuracy that these tools can achieve that would take enormous amounts of hours to do. Considered in the manual of my machine it is advised to do a partial service every 10 hours, that would be quite some work involved too with lots of interrupts.
Same for me. That was a real surprise and a quasi show stopper. I believe that it is more easy to achieve with 3D printer where less eneregy is at stake that with CNC, but with other drawbacks.

The interesting thing is that it is reliable and working in the background, then you can do something in parallel.

One magic thing is the laser cutter: so fast, easy and clean.I can't compete with that thing...

I realized how efficient the classic vacuum bagging process is: quite fast and low cost with not bad results at all. Not esay to compete with :-)

At the moment I try to print without warping, which I didn't succeeded to do yet on large samples.

JMF
Oct 15, 2017, 01:38 PM
myday.. myday
LA8PV's Avatar
Vacuum bagging foam cores with Mylar have it's limitations when it comes to modern designs with a true elliptical plan form and such, but I am happy with that and the results regarding accuracy and the actual joy of the finished model - I don't need it to look that fancy in the mirror to have all the fun and satisfaction. I think the joy of this hobby as in designing, constructing, building and flying very quickly turns in to a nightmare if you let yourself get trapped by the psychotic focus on that everything needs to be so perfect and accurate to be fun or satisfactory like so many people tend to be - it's actually quite the opposite. If you get too hung up on those things you'll soon find that buying a ready made 3 or 4 meter world class competition model is actually amazingly cheap compared to trying and making it yourself, you could buy many models for the same cost - and that is without taking in to account the labor involved, include that with a decent hourly salary and you could buy a brand new car for that money.


I first learned about composite processing from my older bro', but he's all about 3D printing now, he started out with a kit and then built a bigger 3D printer from scratch with he's own selected components and parts - some of which are 3D printed of course Seems to me it's a lot of fun, he uses it to make all sorts of things. I guess it's a little like everything else; there is a lot of parameters that you need to learn gradually how to use to get the results you want.
Oct 19, 2017, 01:14 AM
myday.. myday
LA8PV's Avatar
So how are we doing? Have you got some new results or have you given up already? I know my brother made a v-tail joiner mold for one of he's competition planes with he's 3D printer and it turned out perfect.
Oct 19, 2017, 02:36 PM
Registered User
Hi,

I experimented a bit, but did not went so far. You can see the results in pictures below. Far up to now from being OK. The main issue I have at the moment is "warping". When the PLA cools down, on big pieces like those ones, the corners tend to curl/warp. This destroys the geometry of the piece.

Up to now, I was printing on blue painter tape. This is fine for small pieces or when you don't heat the bed. But when the bed is heated, the glue gets softer and the printed piece tends to lift the tape.

Option now is to print directly on the bed, with hairspray. But first tests showed improvements but not sufficient.

The shell mold shape, with 2mm thickness and 50% infill proves too bendy. I'm afraid it would not be sufficient even putting the mold in the vacuum bag.

The filled one looks more promising.

However, I have at the moment too many projects on the workbench. Many friends have helped me for my Journey F5J, hotwiring foam cores and so on. Their help deserves my dedication to finish those wings. I have to focus on that and come back to my 3D printed molds after.

Your brother's experiments could be interesting to share. Especially if you have a link to some photos or explanations.

JMF
Oct 20, 2017, 11:18 AM
The Junk Man
A glass bed was the best thing I ever added to my printers (I have two). White Rain hairspray and the parts literally fall off after cooling.

No need for expensive glass either. My beds are window glass cut at my local Ace Hardware store.

Tom
Oct 20, 2017, 01:11 PM
Registered User
Thanks Tom,

I will give it a try. I have read several reports that it had helped solving this type of issue.

Did you tried to print molds with your printers ?

JMF
Oct 20, 2017, 02:01 PM
The Junk Man
Quote:
Originally Posted by JMF11
Thanks Tom,

I will give it a try. I have read several reports that it had helped solving this type of issue.

Did you tried to print molds with your printers ?

JMF
IMHO 3D prints do not lend themselves to making molds. The surface, even from a pretty good printer, is still rough enough to give problems. I usually make molds from foam plugs with UltraCal 30 as the mold material.

I have a MPCNC machine that is to the stage where I can drive it around with the electronics but it is not yet complete. Maybe use it, but not a 3D print.

Tom
Oct 23, 2017, 05:10 PM
Registered User
New attempt,

I almost succeeded to print a full wing subsection. See pictures.

Checking with a laser cut wooden template shows that the profile is accurate.

I printed directly on the Aluminium bed + hair spray. It lifted just a bit at the extreme tips. I shall try glass.

This is 20x20 cm section. Surface finish is not bad at all. I will try dry sanding, wet sanding. If not sufficient, I may try some spray filler.

Let's see what can be done :-)

JMF
Oct 23, 2017, 06:48 PM
The Junk Man
Make sure your bed temp is about 65 degree C and there are no drafts blowing near printer. Set your cooling fan to stay off for the first 5 or so layers.

That should help with the warping at the corners.

Tom
Oct 24, 2017, 08:05 PM
myday.. myday
LA8PV's Avatar
Very nice!
Oct 25, 2017, 03:13 PM
Spy's sappin' my thermal!
Chris Sutton's Avatar
I'm watching this with great interest, as I want to try 3D printing a mold for a DLG fuselage. For small, low-quantity parts, I think there's a lot of potential in 3D printing.

One thing that might help with with your corner adhesion issues is to radius the edges that run vertically in print orientation (i.e. spanwise). This reduces the forces trying to pull the corners up as the material cools. It won't solve serious issues, but it looks like you're almost there, so it might do the trick.
Oct 25, 2017, 03:51 PM
Registered User

Just started with this process, also!


This thread encouraged me to write it up. The past month or so I've been working to make the tails for a discus launch glider (the rest to follow).

https://sites.google.com/view/dlg-co...r-construction

It's my first time working with composites, and I'm exploring new techniques, so there has been a few false-starts. I did, however, get some tails!

I think this method holds promise. I will be using it on the main wing, eventually.


Thread Tools

Similar Threads
Category Thread Thread Starter Forum Replies Last Post
Gallery 3D printed hexapod robot (from Ghost In The Shell) PauliusLiekis 3D Printing 0 Jun 08, 2016 02:21 PM
Discussion Anyone planning to do 3D printed fiberglass mold plugs? jzanutto Classic Pattern Flying 10 Jan 28, 2016 04:22 PM
Build Log 3D Printed Molds for Urethane Casting meerkat13 Composites Fabrication 4 Apr 02, 2015 12:15 PM
Discussion 3D-Printed Mold for casting Urethane Foam. methuso Scratchbuilt Indoor and Micro Models 5 Nov 12, 2014 01:11 AM