3D Printed Turbo Impeller VS Dyson Motor

this video is sponsored by PCB way look at that his Dyson handheld vacuum Motors are pretty cool they make a ton of suction for how small they are and when built into a vacuum they are remarkably quiet a few videos ago I built a snail blower housing around the Dyson impeller assembly to redirect the exhaust flow to a nozzle turns out two of these motors produced enough thrust to make this airplane fly although it did work this is a super impractical way to power an airplane these motors have a terrible thrust to weight ratio in this video I'm going to try on design and 3D print my own impeller that outperforms the Dyson in thrust and suction Force this will be no easy task because Dyson is known for using super advanced technology in their vacuums Step One is choosing an electric motor that will rival the dysons Dyson uses what they call a digital motor this is just marketing lingo for a brushless motor with an electronic speed controller similar to the ones we use on our RC planes and drones the thing that makes the Dyson Motors different is that they put a ton of effort into optimizing for efficiency and manufacturability this led them to a design that is quite a bit different from any hobby motor I've ever seen mainly due to the stator layout it uses 2 c-shaped stator cores with coils on each core I think this means it's a two-phase brushless motor which is unheard of in the Hobby World I saw online that the older versions only have two coils and two poles on the rotor low pole counts lead to higher RPMs so this makes sense but with the later version that I have they switched over to a 4-pole motor I'm not exactly sure why maybe to reduce torque Ripple or something like that for smoother starts and RPM sensing they are using a hall effect sensor that is positioned next to the rotor if I'm going to build a vacuum impeller with similar performance to the Dyson I'm going to need a motor that spins really fast the dysons Apparently spins at over a hundred thousand RPM I found this 4900 KV motor that's designed for an RC EDF jet for some reason KV stands for RPM per volt so if I run this motor at 24 volts it should theoretically spin at 117 000 RPM however the KV rating only gives us a rough approximation as there are many other factors that determine the actual RPM such as motor efficiency and load so now that I've got my motor it's time to design the impeller centrifugal impellers work by flinging air molecules outwards in every direction so in order to make this thing blow air in a single direction to generate thrust we'll need to add this snail-shaped housing around it I'm using on shape for cat design all these models are available for your own use at the link in the description you'll just need to sign up for a free onshape account and then you'll have access to everything including the original sketches so it's easy to go in and modify Dimensions to suit your own needs I tried to copy the dimensions of the Dyson impeller as closely as possible both have a 20.5 millimeter intake diameter and a 36 millimeter outer diameter after the design was done it was time to print the parts now in my previous impeller video I just fdm printed all the parts but this was far from ideal fdm prints have lower tolerances so it was difficult to keep the air gaps as small as possible between the impellers and the duct walls also the little layer ridges that fdm printers make lead to poor aerodynamics to compete with the Dyson impeller I needed something much better so I'm SLA printing all the parts on the form Labs form 3 plus these SLA printers can make parts with extremely fine detail but what really sets the formlab stuff apart from your typical resin printer is the material selection their engineering resins are so much stronger than your typical photopolymer hobby grade resin for this project I'm going to use their rigid 10K resin it's supposed to be similar in strength to Glass filled thermoplastics which is what the Dyson impeller is made from more specifically a carbon fiber reinforced polymer the first part I printed was the impeller housing and I printed it all as one part which in hindsight was probably not the best idea but it did surprisingly turn out pretty much perfect after that I printed some impellers and the motor mount to remove all the excess resin from the parts formlabs has this washer machine that circulates isopropyl alcohol over the parts and lifts them up out of the alcohol when their timer is done after that the parts go into the form cure which uses heat and UV light to ensure all the resin is fully cured after that I removed all the support material and cleaned up the parts with some sandpaper this rigid 10K resin almost feels like ceramic it's really cool stuff here's the motor mount getting screwed onto the motor and here's a close-up of the the impeller I made the impeller blades super thin they are only like 0.7 millimeters thick the idea here is to keep the impeller as light as possible to reduce centrifugal force as much as possible the impeller sits on a nut that threads onto the motor's threaded shaft and then another nut holds it on from the top the motor mount then fits in the back of the snail housing and that completes the assembly the impeller is a very precise fit in the housing the air gap between the blades and the ducts are very small similar to the Dyson here's a comparison of the two this is without the snail housing on the Dyson and this is with the snail housing on the Dyson the new one is quite a bit more compact the first time I spun it up there was a bit of rubbing but that problem quickly solved itself thank you [Music] but then with a little more power it blew up all the blades broke off the impeller this probably happened because the motor mount worked itself loose I didn't have it properly attached to the snail housing then the impeller hit the sides and shattered here's some mesmerizing footage from the freefly systems wave camera at 420 frames per second with 4K resolution the motor was running at a fairly low throttle here probably around 40 watts of power which was only like 10 throttle I didn't want to blow it up quite yet but after I had already got some awesome high-speed smoke shots I decided it was time to give it some real power [Music] foreign [Music] [Music] looking at the high speed footage it's obvious that this thing didn't fail because the impeller broke apart due to centrifugal force it failed because this motor mount part here started falling out and then of course once the impeller contacted the side of the housing it just exploded so I need to attach this in there better it was just friction fit then I reprinted everything this time splitting the snail housing into two halves so that they could be printed more easily this would also allow me to better check the alignment before locking the motor in place this time around I also printed the impellers with a higher resolution setting on the form 3. the print quality really did appear quite a bit better I used the built-in lathe to shave down a little bit of elephant's foot that was present and then I tried it dynamically balancing the impellers this basically just means balancing by trial and error I used a little bit of Scotch tape to add weight at various positions around the impeller and a guess and check which position led to the smoothest operation then I would grind off a little material at the opposite side and repeat the process eventually it started sounding and feeling pretty smooth next I powered up the stratus LEDs 150 watt module and cranked to the frame rate all the way up on the wave camera my goal here was to measure the RPM although in hindsight this was kind of pointless since the impeller was not in the housing I think the housing blocks a little bit of airflow which would reduce the load and lead to higher RPM but anyhow I calculated that it was spinning at a little over 80 000 RPM this is not much slower than the Dyson which I measured at 82 000 RPM like I said earlier the Dyson is supposed to spin out over a hundred thousand RPM but I think having the motor out of the rest of the vacuum leads to a higher aerodynamic load and therefore lower RPM now to finish assembling the next blower this time around I was sure to use Loctite on the motor screws I think all these little fins around the inside of the snail are great for maximizing pressure but I'm not sure that they're effective for increasing flow it would be an interesting thing to test in the future the two snail halves got glued together with CA and then I set it up for a test with the power meter this run is going to Full Throttle with a 4S lipo hi it's me Daniel I just want to show you guys something we're going to take a little break from the impeller real quick if we come over here and lift up the tarp on my boat you'll notice these nice big aluminum brackets here that hold the solar panels ooh solar panels anyways stay tuned for the video about putting these aluminum brackets and the solar panels on the boat but I wanted to show them to you because this video is sponsored by PCB way and those big aluminum brackets were made by PCB way and their CNC Machining Services PCB way offers custom full feature prototype PCB fabrication services at a low cost they also offer pcba component sourcing and assembly services but that's not all for those of us who are less electrically inclined they also offer 3D printing injection molding CNC Machining and Sheet Metal Fabrication to get these giant CNC brackets made for my solar panels all I had to do was draw them up in onshape and send the files off to PCB way you can easily get a quote by just uploading your CAD files then a couple weeks later they showed up at my door and the quality was outstanding be sure to check out PCB way at the link in the description thanks to PCB for sponsoring this video now back to the impellers 333 Watts on a 4 cell battery you dare try a six cell I think I will try a six cell I just won't go to Full Throttle because this motor gets really hot when I do go to Full Throttle 106 so I guess it's already ruined so maybe I should just go for it listen to this yeah these bearings are bad they must have gotten damaged when the motor mount fell out and it spun away oh well I'll test it with a six cell and see what happens [Music] holy that was insane 445 Watts wow that might be more than a Dyson and I didn't even go to Full Throttle on that run next it was time for a Thrust test here we go [Music] 581 wait what that was so many watts 518 Watts holy crap that's nuts okay now I think I'm just gonna try and go Full Throttle maybe smoke this motor and blow everything up [Music] there's little chunks of stuff everywhere that's so brutal oh that thing just turned into powder wow oh there's shards everywhere oh god let's see if the motor even still spins yeah it still works so Peak power on that one was momentarily upwards of 700 Watts unfortunately the scale had Auto powered off and I didn't realize it so we didn't get any thrust readings but I don't know if we actually would have gotten a good reading because it was only at Full Throttle for a fraction of a second from the high speed you can see that the impeller was way off center and everything was vibrating a lot enough to knock the light over this is probably because the motor shaft was slightly bent from the previous explosion so even if we assume that 290 grams from the previous run as this thing's Peak thrust then we still have the Dyson beat by 70 grams power wise the Dyson motor pulls 350 watts and mine momentarily did 530 Watts without exploding so I also went in that category if we're looking at the power to weight ratio the Dyson blower weighs 285 grams and makes 190 grams of thrust and mine weighs 110 grams and made 290 grams of thrust beating the Dyson yet again so that's pretty impressive for our little 3D printed impeller here to me it certainly seems vibration was the cause of all the failures I had if this holds true it it goes to show that the speed of Dyson's digital Motors might not actually be all that special but instead what makes the Dyson so impressive is their ability to balance the impeller and the motor super precisely I didn't want to ruin any of these looking for balance marks on the impeller but this high RPM hair dryer axial fan has some balancing marks on the front that are easy to see here's a YouTube video I found that shows the automated balancing process it looks like they basically just use an accelerometer or something to sense the mass distribution relative to the rotor position and then an end of Mill goes in and removes material from the heavy side I would think Dyson uses a similar system the Ridgid 10K resin that I was using to print these parts was probably the best choice but I also had some formlabs clear resin so I thought I'd give that a try as well I got a new motor since the old one had a bent shaft [Music] this impeller sounded pretty bad right off the printer so I needed to do some balancing this time I used little dabs of hot glue to add weight in different sides and then feel if it made the vibration better or worse once I found the heavy side I would grind off a bit of material and eventually I got it sounding much better [Music] still nowhere near as smooth as the Dyson though not that it matters but it's kind of cool being able to see the impeller in there through the clear resin with this one I was able to get it up to 300 grams of thrust without exploding that's 10 grams more than last time I don't think that is because the clear resin is any stronger but instead it's probably just because this one was slightly better balanced after that I gave it even more power and it got up to 340 grams of thrust before vibrating apart There She Goes yet again another impeller Bites the Dust so I've won this unconsensual competition with the Dyson Engineers my blower makes more thrust and weighs less but they definitely have me beat when it comes to reliability noise and efficiency efficiency is tough to know for sure though because I would need a way to measure the volumetric airflow which isn't easy I tried to measure the static vacuum force that these motors can pull but ran into some issues first of all the vacuum meter that I was using maxed out at negative 206 millibar that happened right before a good portion of the impeller turned itself into dust the Dyson motor seems to have some sort of over RPM protection or it starts to pulsate when the flow is cut off so I was never able to get a good reading but for reference my 5 horsepower shop vac only hit negative 139 millibar but that's kind of an unfair comparison because it could have been leaking air through the entire tub scene whereas I measured my vacuum directly from the intake so that's it for this video in the next video I'm going to be taking my Dyson Motors and doing something more useful with them my impeller on the other hand will not be getting used for anything practical because it's loud inefficient and is likely to explode thanks for watching bye

Channel: rctestflight

Views: 555,862

Keywords: Dyson, motor, impeller, vacuum, turbo, 3D printed, 3D printing, DIY

Id: 8RSoA-aB7ZU

Length: 14min 8sec (848 seconds)

Published: Tue Jun 13 2023