Launch of the 3D Printed Rocket at NASA

17 responses to “Launch of the 3D Printed Rocket at NASA”

1. 

   jurvetson

   Mar 17, 2013 at 5:38 pm

   Aerotech E15-7W… with a photogenic plume-to-rocket ratio… =)
   

   Flew straight as an arrow with two body sections and no nose weight…
   
   But, going up is the easy part…

   For reasons unrelated to the rocket design, our parachute failed to inflate and stayed wadded up like a streamer ball. So the rocket came in fast, and hit the tarmac hard. This shattered two fins, popped off the internal motor retention, snapped the body tube and cracked the nosecone:
   
   I detached the shock cord from the fin can for the photo, but it remained attached throughout the flight and through the crash landing. It is a leftover elastic cord and chute from an Aerotech kit. You can also see the white coupler, a simple cylinder we printed from Tinkercad.

   Close-up of nosecone:
   
   You can see a cool build tip for the Makerbot Modular Model Rocket here. If you take the two metal clips off a medium-sized binder clip, they snap perfectly into the shock cord loops on the nose cone and above the motor tube. This makes it much easier to attach a shock cord (especially inside the fin can), and it is centered and easily exchangeable in case you need to print some replacement parts… =)

   Also, the toughest part to print was the fin can; the fins kept lifting off the build plate. Any suggestions there? I tried different resolutions, rafts, textured base plate.
   
   The problem seems to be vertical structures with low surface-area for contact adhesion below. I was wondering if it would help to design an angled rear to the fins (is 45* the limit for printing outward into free space without pillars?). This way, the lift-up is part of the design, and the fin is less likely to strike the ground first on landing.

   I had a similar problem with the body tube, and finally had to literally duct tape the cylinder to the plate mid-build to keep it from sliding freely near the end of the build, which results in something like this:
   
   This is why I had extra partial builds of body tubes to extend the rocket length, and shift the CG forward for stable flight.

   The standard build took a bit of sanding to accommodate the Aerotech motor. I had to use a Dremel to widen the motor tube from within. The motor slid out easily afterward, so the waste heat was not enough to affect the PLA.

   Reply
2. 

   rocketmavericks

   Mar 17, 2013 at 9:46 pm

   This is one reason I recommended the other printers. I am using Ultem, with a tensile strength over 10,000 PSI. Poly Lactic Acid has a Tg of about 60 C not to mention a much lower tensile and compressive strength. I have been woriing with Navy and Air Force Research Labs here, and they are fabircating parts of the F18 Hornet as well as Predators, and other aerospace applications. The Makerbot is great for hacking the Makebot and making your own 3D printers, which I think I pointed out in my recommendations to you. Reprap is particularly usefull here, but as I recommended before, if you want to build useful things and leverage digital materials and digital fabrication capabilities, you do not want to go with the Makerbot.

   Reply
3. 

   Spaceaholic.com

   Mar 17, 2013 at 11:35 pm

   Or PEEK (as an alternative to Ultem)

   Reply
4. 

   PhOtOnQuAnTiQuE

   Mar 18, 2013 at 12:53 am

   Sweet !!!!!! Merci for all the feed backs, will see what we can do here =)

   Reply
5. 

   js.brain

   Mar 18, 2013 at 1:47 am

   Had many ideas for 3D printing, but printing a Rocket what a great idea!!!
   Very cool going up!!!

   Reply
6. 

   paranoidroid

   Mar 18, 2013 at 2:29 am

   This is so awesome.

   Reply
7. 

   rocketmavericks

   Mar 18, 2013 at 3:00 am

   Mavericks has a new program for elementary schools and middle schools for printing your own rocket. Stay tuned!!!

   http://www.rocketmavericks.com

   Reply
8. 

   whitefire990

   Mar 18, 2013 at 4:43 am

   You can still reinforce the fins with fiberglass even if the thing is 3D printed. Even with the fiberglassing process it will still save time compared to otherwise. Besides, as soon as you enter the transonic area and cross the sound barrier, without the fiberglass the thing will fall apart. A single Aerotech G80 easily breaks the sound barrier as long as the fins are fiberglassed and the model is not much bigger than the engine itself.

   Reply
9. 

   jhagerty52

   Mar 18, 2013 at 10:38 pm

   "… our parachute failed to inflate and stayed wadded up like a streamer ball."

   We call this a "para-wad."

   Nice flight, Steve! This is a brave new world for the hobby. It will be interesting to see where it goes. The possibilities are especially good for scale models.

   – Jack

   Reply
10. 

    websumm.it

    Mar 19, 2013 at 7:24 am

    it’s still in the range of hobby rockets

    Reply
11. 

    johnbeans@yahoo.com

    Mar 19, 2013 at 4:56 pm

    One idea for reducing the tendency to twist or sag is to add a broad base into the design with many thin connection points (or just very thin, easily trimmed sections) that can be removed after printing. Similarly, for the rear sweep, just add an angled thin perforation line to your current fin shape, and snap off the rear corners when you are done. Think of them as "printed fixtures."

    Reply
12. 

    jurvetson

    Mar 20, 2013 at 2:57 am

    John – great suggestions. It was good to see you at NARCON, and it would be fun to go to a launch together. We are planning on DairyAire…

    websumm.it – hobby is good. SpaceX has a 3D printer for titanium. I am starting small. =)

    whitefire990 – I want to minimize labor if I go the 3D printed route, so I am wondering if an epoxy layer alone adds composite strength when lathered over the PLA.

    You don’t need fiberglass for the fins to survive supersonic flight. I have broken Mach 1 routinely with plastic fin cans from GLR (two different airframes) and wood fins lathered in epoxy (Binder Thug and LOC Expediter), but I do rely on carbon-fiber tip-to-tip reinforcement for the Mach 2+ flights (four times with Mongoose98), mainly because you have to do that with a minimum diameter airframe. Supersonic Flight Examples

    Reply
13. 

    Spaceaholic.com

    Mar 20, 2013 at 10:55 am

    Is the SPACE-X 3D printer utilizing sintered titanium powder (cant imagine it would be extruded)…

    Reply
14. 

    jurvetson

    Mar 21, 2013 at 4:55 am

    yes, with lasers

    Reply
15. 

    rocketmavericks

    Mar 23, 2013 at 5:34 am

    Your problem is the base plate of the makerbot. It is smooth. You don’t want that. You need to reflect the shear load back in to the part as it is fabricated to prevent surface shear and separation. Biggest problem is the thermal instability ambient to the Makerbot causing thermal gradients and non-uniform cooling. That is why they are so inexpensive, as they do not regulate the cooling process of the melt. This effects tolerance, dimensions as well as melt adhesion. Putting down a soluble support layer first will help increase the surface area adhesion and you can embed the part in it for added support. I think there is a PVA plastisol that is pretty cheap that 3M makes you can use. Have fun!!!

    Reply
16. 

    xwinder99

    Apr 18, 2014 at 12:45 am

    On your question "Does anyone have experience lathering the exterior of Replicator 2 prints with epoxy to strengthen?" – Sure – you need an X-Winder. 2-axis version is on Kickstarter and with the 3-axis version we’ve got a lot planned for the HPR field …

    Reply
17. 

    jurvetson

    Apr 22, 2014 at 8:53 pm

    reposting sub photos to unhide: Aerotech E15-7W… with a photogenic plume-to-rocket ratio… =)
    

    Flew straight as an arrow with two body sections and no nose weight…
    
    But, going up is the easy part…

    For reasons unrelated to the rocket design, our parachute failed to inflate and stayed wadded up like a streamer ball. So the rocket came in fast, and hit the tarmac hard. This shattered two fins, popped off the internal motor retention, snapped the body tube and cracked the nosecone:
    
    I detached the shock cord from the fin can for the photo, but it remained attached throughout the flight and through the crash landing. It is a leftover elastic cord and chute from an Aerotech kit. You can also see the white coupler, a simple cylinder we printed from Tinkercad.

    Close-up of nosecone:
    
    You can see a cool build tip for the Makerbot Modular Model Rocket here. If you take the two metal clips off a medium-sized binder clip, they snap perfectly into the shock cord loops on the nose cone and above the motor tube. This makes it much easier to attach a shock cord (especially inside the fin can), and it is centered and easily exchangeable in case you need to print some replacement parts… =)

    Also, the toughest part to print was the fin can; the fins kept lifting off the build plate. Any suggestions there? I tried different resolutions, rafts, textured base plate.
    
    The problem seems to be vertical structures with low surface-area for contact adhesion below. I was wondering if it would help to design an angled rear to the fins (is 45* the limit for printing outward into free space without pillars?). This way, the lift-up is part of the design, and the fin is less likely to strike the ground first on landing.

    I had a similar problem with the body tube, and finally had to literally duct tape the cylinder to the plate mid-build to keep it from sliding freely near the end of the build, which results in something like this:
    
    This is why I had extra partial builds of body tubes to extend the rocket length, and shift the CG forward for stable flight.

    The standard build took a bit of sanding to accommodate the Aerotech motor. I had to use a Dremel to widen the motor tube from within. The motor slid out easily afterward, so the waste heat was not enough to affect the PLA.

    Reply