Flying Wing Project Uses 3D Printing To Reach New Heights

3d printed glider gets engines

A team of engineers from the Advanced Manufacturing Research Centre at the University of Sheffield have just put the finishing touches on their 3D printed Flying Wing with electric ducted fan engines — a mini electric jet so to speak.

Earlier this year they had created a completely 3D printed fixed wing UAV, which the new Flying Wing is based off of. Designed specifically for the FDM process, they were able to optimize the design so that all parts could be printed out in 24 hours flat using ABS plastic.

The new design also almost exclusively uses FDM technology — however the wings are molded carbon fibre… using a 3D printed mold of course!  The original glider weighed 2kg, and with the upgrades to the design, the Flying Wing weighs 3.5kg, with speed capabilities of around 45mph.

To save weight, neither plane has landing or take off gear, so the team had to create a slingshot catapult in order to launch the UAV’s. Also created using FDM components, it’s capable of launching the planes at 12m/s, or around 30mph.

Down the road they hope to double the size of the plane to have a wingspan of around 3 meters, and use miniature gas turbines to take it to new heights, literally!

[Thanks Joseph!]

58 thoughts on “Flying Wing Project Uses 3D Printing To Reach New Heights

  1. Looks sexy, but if kinda flys like a brick. They need some vertical stabs and turbines or simply props instead of edfs. I do think scaling it up will help, however. Also, since there are two edfs, they cound use the differential throttle and a gyro to increase yaw stability, cuz i noticed some wobble in that axis

    1. I’m not sure how “turbines” would help? Yaw stability by differential throttle with relatively closely spaced engines would be difficult to achieve due to throttle lag and inertia effects. BTW it doesn’t fly like a brick – I know, I flew it. Considering that was the first ever flight, it performed extremely well.

    1. Well the bottom is more likely to come into contact with something that’s underneath all the lovely free air… the ground. And putting a camera at the front would dramatically affect the center of gravity.

      1. Depends on the mass of the camera. Flying wings and conventional aircraft tend to become more longitudinally statically stable as the cog moves ahead of the wing’s aerodynamic center. Too far ahead though, and you lose controllabiity/maneuverability.

  2. They could make the launching platform more compact by using those Ladder locking mechanisme that only click in the right angle.
    And since the wings are already detachable ,why not make it that the whole wing moves instead of the little tiny flaps. less turn radius and what not, don’t know what that does to the dynamics tho

    1. Those weights are comparing an unpowered previous version to the current ducted fan version. Their case study for the current version notes a self-imposed upper bound on mass at 5 kg, so they’re doing pretty well on that.

  3. They should have attached the cam at the bottom. ’tis more useful than looking at the clouds… I’m sure, if you could put pilots in planes upside down, all planes would have cockpit windows at the bottom as well…

    1. We can put pilots in upside down. We still don’t because it messes up their hair (coincidentally why fighter pilots, those who have to be upside down from time to time, hide theirs with helmets), and is silly.

    1. So you figured out that an equivalent foam airframe would be lighter? Very well done.

      Obviously we could have CNC milled the entire airframe from foam, or used a host of other commonly (or not so commonly) used techniques to build the aircraft, quite possibly lighter than the model you mentioned, but we would have learned absolutely nothing about design for FDM, which was one of the main objectives for the project.

  4. Very nice build, but for me, this ain’t no hack.
    The files are not released nor open source, and they don’t say a word about releasing them somehow (mainly because that’s not the purpose of this group).
    It’s interesting, but without more details, kinda useless (sounds more like an ad for a company than a real hack, and given their means, of course they succeeded!).

  5. Absolutely cool and awesome outcome, however, I fail to see the point. Usually engineering is to make something better. Building a model aircraft is certainly not engineering but hobby, without wanting to insult the latter. It’s heavy, it’s expensive and the properties in the air seem to be only ok’ish. That’s why we fund engineering departments with several hundred thousand bucks a year?
    There is so much room for improvement here! Print molds for foam and fill the foam with aerogel to make a super light plane, do something with the printing that can’t be done with injection molding, directly print a foam aircraft whatever. But making a carbon fiber flying wing? Not really convincing use of the money and resources.

    1. If you read the case study, you’ll find the main objectives of the project were to give the team experience of designing complex structures for rapid manufacture. Completely eliminating the need for support structure on the glider version thereby reducing manufacturing time and cost, and using printed moulds for carbon fibre on the powered version were both steps which have applications in a wide range of other projects totally unrelated to aircraft (in other words “making things better”). In addition, validation of CFD results through physical testing was achieved. In other words, the project achieved all its objectives and more.

      1. Not sure how 3D printing and carbon fiber molding helps with the rapid manufacture??? If you are looking for rapid manufacture consider foam molds and CF reinforcement. In this case you can manufacture a single peice in about 10 minutes. Or CNC foam routing / foam cutting. Let’s say 1h.

        3D printing as you mentioned is 24h. Manual carbon fiber molding is at least 3 or 5 hours.

        Plus your design is just to heavy for no reason …

        1. Obviously we could have CNC milled the entire airframe from foam, or used a host of other commonly used techniques to build the aircraft, but those methods were not what we were particularly interested in.

          1. I think it would be useful for you to do a Google search for “Rapid Manufacturing”. In the context of the vast majority of definitions, it doesn’t mean simply “making this quickly”, which is what you appear to be assuming?

      2. Yeah, I thought about it and it’s really an educational project, and in that sense it fulfills its goals scientifically, not so much, but if that wasn’t the goal, one shouldn’t complain about its lack. But mold printing is not new. Just applying it to aircraft models is also a trivial step. And eliminating support structure is also not new, foam gliders don’t have it either. There isn’t anything new in this one but as a team experience, yeah, why not.

  6. Wow GN, you’re mean.

    Anyway, project is pretty cool. I’ve been thinking about 3d printing molds for awhile. I think that 3d printing is a great addition to the toolbox. But it doesn’t quite get you there by itself since you still need to take it to mass manufacturing. CNC is still my faves.

    1. Again, if you’d read the case study (or even other comments on here) you’d see we were working to a span limit and a max. weight limit. This pretty much defined the maximum wing loading, given a reasonable planform. Since we were well under our max. weight limit, and at our span limit, obviously we were under our wing loading limit. So no, we were not particulary concerned by it.

        1. Firstly, your wing loading/stall speed assumptions are wrong, and you could easily have figured this out for yourself if you’d studied the test flight video more carefully (or indeed, at all). You’d have then seen that the catapult launched the aircraft to a peak velocity of 9.57m/sec (around 34 km/hr.) at the end of the rail, and at that point it was obviously not in a stalled condition. So please tell me how the minimum stall speed for the aircraft can be what you seem to assume i.e. 42.8 km/h? If we assume that our launch speed had a safety factor built in, then it’s clear that your assumptions are even further out.

          The fact that you would be concerned about the stall speed of the aircraft is irrelevant: The fact that the aircraft was designed and built in the way it was, and performed as predicted during testing more than justified the effort that went into the project. Remember, this is not intended to be a mass-produced item, rather a prototype designed and built to further our knowledge of certain manufacturing and design techniques.

  7. Wow, that whole 3D printing stuff is amazing. You’ve duplicated a flying wing that’s been made better, faster, cheaper since the 70’s and it almost kindof sortof flies like those earlier models. Not exactly the best example if you’re trying to show your rapid manufacturing chops.

      1. E.g. HK Sonic 64 EDF Wing EPO 1230mm
        – flying weight 700g
        – costs $100 including EDF, ESC and servos
        – it is rapidly manufactured
        – has good flight characteristics with low wing loading

        On the other side your design:
        – looks sexy
        – uses agile design techniques
        – does not use rapid manufacturing techniques
        – it is too heavy
        – top speed does not justify the weight
        – top speed does not justify 2 EDFs

        1. So you think a mass-produced, moulded EPO aircraft is Rapid Manufactured? It certainly isn’t within what our definition of RM is. Here is a common definition:

          “An additive fabrication technique for manufacturing solid objects by the sequential delivery of energy and/or material to specified points in space to produce that part”.


          …and you also think that an airframe that uses FDM does not use rapid manufacturing techniques?

          Nor does the Sonic 64 have twin ducted fans.

    1. Pretty much any blended-wing-body with integral engines will look something like a Horten 229, although I don’t think the Ho.229 had canted, swept wingtips, nor did it have a duck tail for pitch trimming, nor did it use only elevons for control.

      The project wasn’t about coming up with a particularly unique aircraft planform, it was about learing about applying design specifically for rapid manufacturing techniques such as 3D printing…In fact I believe that’s mentioned in the case study.

        1. Sure, we could have designed a coffee mug, but then again it’s unlikely we would be discussing it as we are here. One of the many benefits of building an aircraft is that people love to see things fly, and as such this project has generated an incredible amount of interest in our group. Even inane, cynical and ill-informed comments made on forums such as this ensure our work is brought to an ever wider audience. So thanks for your input.

  8. @GN I don’t think most of the comments on here came from people that went to engineering school otherwise they would understand the case study and reasons behind the design. I’m sure if your goal was to purely build a fixed-wing UAV then the design constraints would have been much different, so there’s really no point in arguing the aeronautical design.

    1. “@GN I don’t think most of the comments on here came from people that went to engineering school..”

      You’re not kidding! Thanks for a considered, intelligent reply.

      From our YouTube viewing statistics, this thread has had a significant positive effect, so it was well worth the pain. Job done, over and out :-)

      1. i know i’m late to the party, but first, you just showed us an excellent example of feeding the trolls, even if they were trolling unintentionally, and seconly, i think half the arguments on heere werre basicallyy about word choice. HAD usually describes 3D printing as “Rapid Prototyping” which it very clearly is, vs rapid manufacturing, which is a less easily defined and kind of alien sounding to me. especially considering there have been a few arguments on here about people using printers to mass produce objects, when they should use the printer to make a mold, as printers are prototyping platforms. it’s weird, but those are my two cents.

Leave a Reply

Please be kind and respectful to help make the comments section excellent. (Comment Policy)

This site uses Akismet to reduce spam. Learn how your comment data is processed.