It Took Me 3 Months To Get This Working - Two-Stage Centrifugal Compressor 100% 3D Printed
ฝัง
- เผยแพร่เมื่อ 27 มี.ค. 2024
- PCBWay Your 3D & CNC One-Stop Solution: www.pcbway.com/
Support the channel and join my Patreon: / letsprintyt
After not finding success in the last two, finally, I finally made a fully 3D printed two-stage centrifugal compressor that works better than I could ask for.
It's a big project and took a lot of time to print and build it, but it all was worth it.
Two previous videos:
• Two POWERFUL 3D Printe...
• Is it Possible to 3D P...
DOWNLOAD THE MODEL: www.thingiverse.com/thing:655...
________________________________________________________________
Follow me also:
MY PATREON PAGE: / letsprintyt
Facebook: www. letsprintyoutube
Instagram: / letsprintyoutube
________________________________________________________________
#3dprinter #3dprinting #diy #homemade #project #viral - วิทยาศาสตร์และเทคโนโลยี
If I’m not mistaken, this is the first time we’ve seen your face, it like to see that build video!
when the video started I was like 'who the hell is that?'
@@TuttleScott i knew it was beethovens estonian cousin right away
Dude he looks so different than i expected😂
@@Javii96right? I totally thought he was gonna be a brown guy! 😂 😂 now I wanna know what his accent is because I've only heard similar from brown guys, like its messing with my head
@@forbiddenera😂 i thought he would look more like the guy that electrocutes himself in all his videos, i know you know who i’m talking about
David Beckham- good looks -squeeky voice. You accomplished 3d printing engineer and good looks. I can only imagine you must be hiding a terrible secret.
lol
💀
the balloon was such a formidable foe, he decided to face reveal after defeating it
Dude, you look absolutely different from what I imagined! 😂
But it's nice to finally meet you! 😊
Fantastic work as usual!!!
Stay safe there with your family! 🖖😊
This is what happened. The ball bearings heat on operation causing the walls to deform on each iteration the ball bearings will be wobbly and that cause your design to fail. You need to add a layer of heat resistant material that hold and dissipate the heat of those areas where you interact with ball bearings. Another topic is to clamp properly the 2 parts together and seal them properly and you will increase your success. Great work so far keep it up.
You can cool plastic near bearings with the air by making air channel insude plastic.
The handsome devil himself!
This series has been a genuine pleasure to watch. Thanks so much for sharing!
Love seeing the things you make with 3D printing. I just love the iterative process, of finding slightly better and better ways to make something more efficient. and 3D printing lets regular people do awesome stuff like this, without 10's of thousands of dollars worth of machinery, and thousands of dollars in material. 3D printing is such a learning and teaching tool, since it lets a regular person design and print something fairly quickly, test it, and make changes, all in their own house.
Can't wait to see what else you come up with!
Woo! So excited to see the balloon pop in the intro. Nicely done. It's amazing to see the progress you've made since I started watching your stuff a couple years ago.
You can use some Orings gaskets rope to get some seal, you can buy it by meter and cut the necessary piece
Congrats! This has been such a fun project to see you adapt and evolve over time!
Regarding the airflow: When I designed my turbopump that was one of my challenges (the pathway of the air) it's preferable if you do it tangentially.
Also, the leakage problem exists when you use PLA!! Everything must be water tight in order to make it work properly, that's why you couldn't pop the ballons up. And to be honest, on your previous designs, I think that was also part of the issue..!
With 600W power I can "propell" water at about 3 metres, the thing was design to operate at 3kW but since it's a 3D printed device, I seriously doubt it won't blow up once I apply that much power, but as proof of concept, yeah, it worked! :)
Try to use CF Turbo (they do have a demo version) to design your props, believe me, once you get the hang on that thing it's just marvelous!
Ah! Another thing, add a pressure gauge to the exit to measure the output pressure!
Excellent work, my guy. Huge respect 🤯
Love it can't count how many times I've tried explaining to people on you tube trying to 3d print turbo's about volute casings. Being as that's where the pressure comes from and how it increases volumetric efficiency through fluidic coupling or entrainment if you prefer.
this is one of the coolest things ive ever seen. keep up all of your good work. youre incredible
Nice work, looks like you pushed resin/plastic to its tolerance levels. Anything better will require metal parts.
so cool to see the progress and the result - keep it up !
I love it, very fun to watch. Keep it up, can’t wait to see what’s next!❤
Good job. For the leakage problem between two printed halfes I use sandpaper on a glasplate to get both halfes as smooth and plane as possible (180 Grid and some water). Then I use some grease ( e.g. white petrolatum). Works well with PLA.
Really enjoy the new video format! This is awesome!
Well done! Love the new set up too - keep going!
I stand impressed. Well done
What most fascinating about stuff like this is seeing up and coming engineers being able to make things they've thought of in the comfort of their own home with these printers. Now imagine if you just take these are then make scalable models with CNC metal parts. You quite literally have just made a jet engine. Saving the R&D costs of multi-billion dollar companies. It's now in the hands of the "average everyday" person. Absolutely brilliant.
Modern tech has brought back the basement tinkerer back into a viable path for inventors. That was the standard for centuries, and the industrial revolution shoved the individual into a corner to do niche things. But what technology undid, it also re-did, and here we are. Ain't it cool?
Speaking of printers, does anyone know what printer he used?
@@viduraherath4008 a mix of SLA and FDM printing
Hey congrats! I knew you'd work it out!
The problem with that compressor being chewed up is either something where the shaft is off alignment by a hair, there's flex somewhere once it's in motion, or in some cases plastic parts like that will expand a bit if they spin fast enough.
Might be neat to see something like this go further into making a rudimentary gas turbine engine, but it'd have to go into green sand or lost PLA casting processes with machining to clean up and balance parts. No way to get around working with metal, once you start dealing with heat engines. Also at that point you'd have to find a way to lubricate the bearings too, because heat and sealed bearings tend to not get along.
Thank you for putting in all this work and not hiding it behind a paywall. You're one of the good ones. If I had it I'd sub to patreon to support you but I'm sure others who have the money to help will! Meanwhile, liked and subbed here to help where I can. ❤😊
Neat. I wouldn't have thought you could push 3D prints this far and I actually work with them.
Thats an impressive and insteresting design
I'm not sure if you know this, but rotors like these, with a flat bottom, and that suck air from the top, have a net force pulling them axially "up" or out of the intake, because of the pressure difference between the plate at the bottom and the impeller at the top of the rotor.
Wow, nice to see you!
Excellent video and fantastic choice of music man! 👏👏👏
It looks beautiful, good sir
Congrats on the successful build. After all of this, i can't recall seeing you hook the balloon directly to the hose, as a control.
A check valve on your outlet will help as the back-pressure is probably causing movement in your shaft and some resistance. Also like turbos etc you need very low tolerance gaps between your impellers and casing (cheap method could be to use something like aluminum tape on the casing or possibly seals on the blades from a soft material)
- Either way super cool, my 3d printer gets hardly used cause i suck at CAD so i just draw things and make them other ways lol
I love the build music!
There are real world applications: two-stage turbochargers. They use exhaust pressure of an internal combustion engine in place of compressed air, but otherwise operate this way. Normally, turbo chargers are just a single stage, but there are some two-stage ones out there. Very cool build!
most of the time you just pipe one turbo outlet to the inlet on the next so they run in sequence, make it a lot cheaper.
@@TommyApel Yeah, compound turbos are crazy, but they aren't really useful for anything besides diesel because air-fuel ratios and pre-detonation issues with normal gasoline. Too bad the only diesels popular in America are the big boat anchor truck engines like Cummins and Duramax. It'd be nice to see more 4-cylinder turbo diesels.
Crikeys mate! Nice Haircut!
What are the velocity comparison between the air coming out of the air line, and the air coming out of the compressor?
Congrats!
:)
This is so GOOD! 👍👍
Yes! Finally, the balloons are exploded!. Good job
Holy incredible!
If I'm not mistaken it looks like there's no point in having two stages. If they both have the same (or similar) sized intakes and operate at the same speed (same shaft) and also have similar outlet sizes then how is there a compounded compression. My understanding is that the compressed air from the first stage will expand to fill the volume of the second stage inlet and be back to almost atmospheric pressure. Great video and I'd love to hear your opinion on this
Looking at the stages, based on a visual assessment of the compressor wheels, it looks like he's got 2 different gas volumes there. I'm not sure how different, but they definitely look to be, also different angle on the blades. Not sure how that affects it either.
If the impedance (output cross-section) is mismatched he could be getting no real increase, sure. However I think he is still getting an increase. The ideal airflow or air velocity can be different at the same rpm because of the impeller geometry. Even two impellers of the same geometry could produce a static pressure increase I believe, but not as efficiently. You could of course put pressure transducers in each stage to verify this.
An interesting test would be to remove one impeller to see if it makes any difference to the time it takes to explode a balloon.
One base line test would be to use your compressor directly attached to a balloon to time it.
Great job!
When is the vane pump coming? Definitely was one of my favorite 3d printed pumps I've created!
Awesome video as always!
It took you 3 months? Damn it took me 3 months to make a nerf gun tht didnt even work😭
Why didn't it work my guy?
@@nachgebaut4176I think because the flywheels were too far away and the motors were slow. Unfortunately I would need to redesign the entire gun and i kinda dont have time for this.
The concept is great I think it can be adapted to be used as a replacement for compound turbos on diesel engines
a little wear is not a bad thing. What that means is that you don't have a lot of leakage in that area. Ideally it wears say 0.01" and then just rotates freely
I love your videos. You will be one great engineer one day. Just be patient and think all over, a few times more.
You should try printing some TPU to use as a gasket, instead of the hot glue. Polymaker TPU HF 95 prints flawless on the Bambu Labs X1C.
Depending on the infill you can alter the rigidity.
Hope this will give you some other perspective.
Keep up with the great work!
we actually do have a use for these when powered by compressed air. pressure boosters.
it's not something i'd recommend and piston boosters are generally better but i have actually made a 1:6 centrifugal booster. it made sense for the project but the deciliters per second it puts out makes me cry, even at 35 bar.
Free video. Thanks for sharing
6:55 Instead of your plastic seals, you should have put seals cut from the inner tube, it's much more efficient and it saves you from using glue. In addition, your joints must be the full height of your air compressor.
Nicely done ! But, such narrow air passages . I would add cross sectional area to the air passages to minimize resistance to flow. also, a thought: if you make a channel in the mating surfaces of the casing from one end to the other you could fit a seal made of polyurethane elastomer into it to seal the leak. Best of luck!
Correct me if I am wrong, but is it the 1st time we are seeing Let's Print?
Also you look damn cool, I love your channel!
You are correct!
Great video, regarding the damage to the turbines (especially in the previous video) i have a hunch could be attributed to the perpendicular force generated by the turbines, in reaction to the compression action (a gyroscopic effect if i'm not mistaken). Placing them vertically was a great idea in-fact seems to do the trick, at least in most part. Hope this could be useful
Glad i said something useful, but actually i mixed 2 forces: the gyroscopic effect (which i think has the higher culprint) and the reaction to the compression by the air.
I think you could analise the vectors in the project fase, and you could avoid the gyroscopic effect using an opposite rotation part that pull in the opposite direction. Although i fear would really complicate A LOT the project
This is really interesting. Thanks for the video. Here are some possible improvements and explanations of the phenomina at play:
First of all, it appears that you dont have a designated fixed bearing. So when standing up, there is some metal to plastic contact from the couplers or even some plastic to plastic contact in the turbine, which increases friction.
Secondly, the reason why the first impeller got shaved and gradually got more play, is either unbalance or an aeroelastic phenomena, where the turbine gets pushed of center by the aerodynamic forces. It could be a feedback loop between the play thats already in the system (due to it beeing mounted on the bottom, rather loosly, at a different height then the center of mass and the center of pressure) and an induced oscillation caused by the tighter side achieving higher pressures, which then pushes the compressor over to the other side - repeating the process. This would correlate with it getting worse faster after achieving higher pressures. To get rid of this phenomena, increase the stiffness of the coupling between the shaft and the compressor wheel. You could also mount it at the top and at the bottom.
The second stage could be improved with a propper stator that realigns the airflow and corrects for induced rotation in the fluid. Since the second stage is rotating at the same rate and direction as the first stage this induced rotation reduces the bite the second stage has on the fluid significantly.
Also as you found out, reducing leakage is key. So implementing a proper pressfit seal (printed from TPU) between the two halfes would help alot.
As for all turbomachinery: tollerances are live. So the tighter the tolerances, the less the airleakage, the better.
Maybe this helps if you decide to continue the project, or anyone interested.
Nice job
Great video. Do you think it's possible that you may be getting some slipping on the set screws on the shaft?
Now we got a face to the voice. OK for some reason I was thinking of an old guy with a beard. I got no idea why. BTW, your videos are awesome!!!!
Thank You!! :)
Excellent work.
It's just a bummer that we still have to be limited on plastic printers...
A lot of jet engines are started with compressed air from another jet engine, the apu, or compressed air tanks. So your design does have a real world analog
Very cool!! Congratss!!
Maybe the first impeler got damaged because the axial force from the system. Try to reduce the tolerances and add an axial bearing. You could add after the gas turbine.
So this is basically like a compound turbocharger setup you'd see on a crazy modified diesel truck, except with a single turbine spinning both compressors. It'd be interesting to see how this sort of design would change the performance and efficiency of diesel engines compared to normal turbochargers.
Even an audible difference in performance before and after sealing it. I was going to suggest sealing the joints.
Es un trabajo formidable,me ha gustado el ensamblaje de lo creado en 3D y la unión con el metal para formar todo el conjunto.
Your hair looks great
Maybe look at using 3d printed TPU seal ring between all 4 outer housing parts. the TPU will compress against the walls and help keep everything sealed in, so less air would leak out.
Add a combustion chamber and turn this into an actual jet engine! I wanted to 3D print one that ran on ISO alcohol, but lost motivation half way through. Even if it runs on its own for ten seconds before it melts, that's a win!!!
I think you missed checking out normal turbochargers. For one, they have a beefy thrust bearing which you seem to have figured out. Second, dual-stage compressors are usually for high pressure, not really for large flow. The 2nd stage wheel is likely too large, or 1st not large enough. Third you need to seal between the inlet AND outlet for each stage, not just one large seal. Fourth, when spool-up isn't a big deal, a larger turbine outlet is better. They work off the expansion of the turbine air, not so much the paddle/pinwheel effect of air slamming into the inducer-side.
The impeller are rubbing the housing because the clearances are considering what is happening at 0rpm, when up to operating rpm the centrifugal forces would cause expansion axially. Automotive centrifugal compressors using aluminum alloys are stiffer(relative to plastic/resin). If you dont want to do the engineering work to figure it out you can just axially scale the model and test the new one.
Really cool build! For the gas turbine, wouldn't it be better to use a pelton type? And for the compressor side you need to consider volume, if the gas can expand again you will loose efficiency.
The funny part is, if you remove the entire 'compressor' in the middle, the greatest efficiency is realized. Going directly from the shop compressor hose to the balloon. Just... LOL
Very nice small project. Imaging putting a rotary gear ratio. These impellers will explode.
Wow, pretty close to some rocket engine turbopump designs, but an impulse turbine to drive the shaft would get you better dP across than the reaction turbine you have now. You should consider adding an inducer to your pump inlet before the impeller for even better efficiency
Not sure what the goal is here. Re-designing the wheel? Very cool print though!
If you just want to blow up balloons. There is a much easier, more efficient way to do so. It's put the balloon directly over the end of your air compressor hose/ blowgun. It would most certainly take lots less air.
I guess it's fun to play though! Or is it?
I will give it a thumbs up for the effort in print skills. 👍
maybe instead of burning stuff to get a volume increase you might inject LN2 into the airflow, which boils becoming N2, and that can then drive the turbine.
to evaporate the LN2 a heat exchanger might be used.
Theres a difference between your 2nd stage and turbine and the 1st stage which is likely causing the rubbing.. the first stage is unsupported while the other 2 have support on both ends. 1st stage is wobbling. Put a support above the intake and lengthen your rod.
Baloon Exploder 1000 is ready (almost)!
Do you think there’s any relationship between the angle of the compressed air on the turbine and a sailboat going faster than wind because it’s traveling diagonally across the airflow?
GOOOOOOD
My Theory: The only turbine that shows wear from eccentric rotation is the only one whose axis is not constrained by a bearing at each end. If you were to extend the shaft and incorporate a bearing at the inlet mouth, you might be able to eliminate that eccentricity. No need to reprint the housings necessarily either: You could adapt a bolt-on or clamp on or press fit bearing holder retrofit. As simple as a y-shaped thing, or even as fancy as the LAX air traffic control tower.
now turn it into a turbo jet engine !!! it already has almost everything it needs to do that actually
Could you test it as a heat energy recovery mechanism?
Use as a way to cool something else, and see if waste heat collected could help power?
It’s what I figured that the other parts were resin prints and the other parts were leaking compressed air . Probably reinforcing the larger prints with resin
The balloon is not an indicator of how well it works. The more you restrict the outlet, the higher the pressure will be at the balloon; if you restrict the outlet fully, it would work at top efficiency going by the balloon measure. It only makes sense to use the balloon if you're driving the turbine with a motor, otherwise the balloon is just an indicator of the air pressure you put in, combined with the amount of resistance the air faces going to the outlet.
Face unveil, eh? Enjoy your videos. Your success is deserved :)
Thank You ;)
Great 👏
What resin did you use for the impeller and with what printer?
Also is the outside fdm printed?
Hello,
For resin prints I used Heygears Ultracraft Reflex and resin called PAU10 (it's ABS like resin but stronger). For FDM I used ABS+ and Qidi Tech Q1 and Max 3.
I think filling seams with silicone would have been much better as you would still leave an option to easily split the parts as long as one side had some light layer of wax or silicone grease to prevent it from sticking to one of the sides
Cheers 👍💪✌
Yay!
Just an observation, I wonder how well it would perform if you made the both impeller and turbine like turbo's impeller and turbine?
Literally sounds like a mini jet engine before you fire off the ingiters
which resin did you use for the impellers of the turbine?
How did you design the impellers?
You should measure the pressure tho
im not sure if the secound stage does anything because it looks like the gas exspands to its initial volume after the first stage, the first stage and the secound stage turbines should have differant sise outlets but they seem to be the same so i think only the last stage actually does anything
what about measuring cfm? incoming vs outgoing to show if it's effective or if it's just a 1:1 transfer
Serious question, could you make one that would fit on a 6.0 powerstroke? Money is not an issue.
Bro's making a rocket engine pump... Now it only lacks some combustion turbine, and boom turbopump
Before watching this video, I watched Made with Layers (Thomas Sanladerer) printer review with the questionable French baguette and now the balloons having inflation issues. 🤣🤣🤣🤣