3D Printed Quadcopter Props

Here’s something that isn’t quite a hack; he’s just using a 3D printer as a 3D printer. It is extremely interesting, though. Over on Hackaday.io [Anton] is creating 3D printable propellers for quadcopters and RC planes. Conventional wisdom says that propellers require exceedingly exacting tolerances, but [Anton] is making it work with the right 3D file and some creative post-processing treatment of his prints.

These 3D printed props are a remix of an earlier project on Thingiverse. In [Anton]’s testing, he didn’t get the expected lift from these original props, so a few small modifications were required. The props fit on his 3D printer bed along their long edge allowing for ease of slicing and removal of support material. For post-processing, [Anton] is using acetone vapor smoothing on his ABS printed design. They come out with a nice glossy sheen, and should be reasonably more aerodynamic than a prop with visible layer lines.

Although [Anton]’s prop is basically a replica of a normal, off-the-shelf quadcopter prop, 3D printing unique, custom props does open up a lot of room for innovation. The most efficient propeller you’ll ever find is actually a single-bladed propeller, and with a lot of experimentation, it’s possible anyone with a well-designed 3D printer could make turn out their own single-blade prop.

52 thoughts on “3D Printed Quadcopter Props

      1. But then you can’t speed up/slow down the individual props, which is the easiest way to achieve control of a quadcopter. Not the only way… others have done it through variable pitch. But that makes yaw a bigger challenge that requires a custom (and so-far closed-source) controller.

        1. Variable pitch single bladed props–that would be really unique. Also, I’m surprised no one has come up with an open source controller for the Stingray (or others), especially given the proliferation of open-source multirotor options..

    1. That depends on if you are counting the ~$700+ price of having a 3D printer or not.
      Just tossing some figures into a filament calculator ~32c US course it prints 2 blades so ~16c ea and that would sidestep the delivery time by having them ready in about an hour instead of ~3-7 days shipped.

    1. It’s more efficient aerodynamically (drag and turbulence-wise), since you’re not running a blade through air that is turbulent from the prior blade moving through it. This also reduces vibration (and noise). The thrust profile (for a prop it’s technically not a direct thrust output like a jet) can be altered to match that of a multi-blade prop by changing the geometry.

      1. isn’t the air turbulent from the previous turn of the single blade? In this case that means the effect is “only” halved (ad don’t know how much that means in the grand scheme of things).

        1. It can be turbulent, but far less (generalization) than if there was another blade involved. It tends to not be a linear effect either, but again, it really depends on the geometry of the blade, airspeed, etc.

      2. 3D printing would be a great way to do a science fair project on this topic. Easy test would be to measure the wattage required to make a quadcopter hover with a particular propeller, then compare a number of prop geometries head to head.

        This holds thrust force constant but not RPM, so (as some have mentioned) you would really just be testing the combined efficiency of the motor and propeller and would not know which variable is contributing more to the measured results.

        A better project might be to attach the quad to a pedestal attached to a sensitive scale and measure thrust produced at set RPM values.

      1. Or you could increase the angle of the prop to provide more force at the same RPM. So no, you don’t need to spin faster for a single blade prop you just increase it’s angle so it pushes twice as hard and use the same RPM.

  1. “Here’s something that isn’t quite a hack”

    How surprising, coming from Mr. Benchoff

    [Seeing more and more posts deleted each day. Censorship is what the hacker spirit is about right fam?]

    1. Traffic optimization. Goal is to increase advertising revenue. So every day with not much to repeat, is used to fuel traffic with controversial, sensationalist articles. More page views, more money. HaD is a commercial magazine, now!

        1. It’s not nearly as bad as some commenters seem to suggest. Certainly not worthy of constantly attacking authors. I certainly don’t read every article, but I get a lot of value out of the site.

          Again–if you don’t like it, GTFO. Start your own site.

          1. While the “find another site” is a valid statement, it is equally valid to complain to prevent a beloved site from straying into mediocrity. I know that personally, I don’t prefer to see the non-hacks on here which have become so prolific as of late.

    2. I bet you that kwtoska is is actually Brian in disguise and he’s starting turmoil to make his posts more popular. Or Maybe kwtoska simply doesn’t understand how these things work.

      1. If you interpret “kwtoska” as a series of initials for states of the USA, you’ll see it spells Kentucky, Washington, Texas, Ohio, South Dakota, Kansas, and Alabama. Further investigation will uncover that these are all States that Brian has visited! B^)

  2. The most efficient prop is that one that matches that produces the trust you need at the sweet spot of high efficiency on the torque/speed curve of your particular motor. At the system level if you have to spin faster to get the same amount of air flow as the conventional prop, then that efficiency might not be there.

    1. it should be a minimal (but still non-ignorable) due to the layer direction, at least within the blade section. I would like to see more structure around the hub, that bothers me slightly

  3. “Conventional wisdom says that propellers require exceedingly exacting tolerances…”

    If you can do it by hand with a popsicle stick held in a jet of steam, I don’t think it can be that high-tolerance.

    1. Really? Maybe wiping with acetone on a rag would be better then, smooth off the edges without affecting the middle.

      Then again maybe a little moulding setup would be good, use the printed blade to make a mould from, say, plaster. Then something like a hotend to inject melted plastic. Maybe heat the mould up in an oven first. Actually that might be a good way of replicating prints cheaper, perhaps something 3D printerers should consider owning.

  4. 1 bladed prop might be more efficient that 2 bladed (because last turbulence is 1 rotation in front, instead of 1/2 rotation for the 2 bladed prop), but the thrust is lower (half of surface area) for the same size (and RPM), so you either need to increase the size (which might not be possible for some quads) or increase the RPM. Increasing the RPM seems to me to work against the supposed efficiency, so there might not be so much gained.
    I am actually thinking about multi(5) bladed props being better. Either the same size as the 2 bladed props (slower rotating, with change of KV of motors) or smaller in size for the same KV.

  5. As someone who is in the 3D printer business (Some might know me, created Cura, working for Ultimaker for 3 year now), and some hobby experience in model planes.

    Please, please, please! Don’t use this in an actual setup. Props are not that expensive, and real props don’t explode in shrapnel.
    The slicing&printing process will also guarantee that it’s no balanced at all.

    I love 3D printing, but without proper processing these props are a danger to yourself and your surroundings.

    If you do want to print props. Print them, and after that give them a fiberglass or carbon fiber + epoxy coating for strength. And balance them.

    1. I agree with you on the dangers involved. It can, with added security measures, be a nice way to try out different sizes and geometries to optimize control or efficiency. But for normal copter pilots i would always recommend to work with store bought injection moulded props. They are cheap and a LOT more safe and reliable.
      Of course, as others have mentioned, you could try and make your own injection moulds based on the results of the printed design, but again, proper mould design and moulding processes are not that easy to make for a safety-critical part. In the end, i would not want my copter to crash and have expensive equipment destroyed or even people injured, just because i had to save a few cents with 3d-printed props.

  6. I’d love to see someone do a finite element analysis on an RC prop. I know it’s been done on real props, but these plastic ones work a bit differently due to the materials and thickness.
    I’ve had props sling blades before due to damage, and it’s not pretty. A 10×6 slo-fly prop sounds like a shotgun when it explodes at full rpm.

  7. Nice work. I recently presented similar work at an aerospace conference, here is the link if anyone wants to readhttps://www.researchgate.net/publication/281030455_Evaluation_of_Additive_Manufacturing_Techniques_for_Fabrication_of_Propellers_for_SUAVs

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Google+ photo

You are commenting using your Google+ account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s

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