ฝัง
- เผยแพร่เมื่อ 18 มี.ค. 2022
- Designing, 3D printing, and testing a turbo pump for injecting fuel and air into a rocket. All turbo parts are made with resin 3D printing, to eventually inject fuel and air to a rocket.
3D files:
www.thingiverse.com/thing:532...
Turbo uses 3mm bolts and nuts.
Small bearings:
www.amazon.com/dp/B082PS3QDJ/...
Large bearings:
www.amazon.com/dp/B07R11PBZ8/... - วิทยาศาสตร์และเทคโนโลยี
This is a really cool project. I love the fact that there’s a whole group of TH-camrs collectively attempting to build DIY rocket engines and stuff! Soon we’ll have the first completely TH-camr run space agency lol!
Look up "Copenhagen Suborbitals" if you want to see something similar to that idea, it's really neat!
Look up bps space
@@Infinitescale-oni That's who I am talking about in part when I said "a whole group"
Me and afew friends are nearing the end of our SSTO research and we hope to have our test plane done in 2 years
I've seen a lot of comments on the turbine/impeller geometry already, but I think another equally important thing to improve is the shaft constraint. Using a single large bearing like that creates a lot of resistance, and a single deep groove ball bearing has angular play (by design) which makes keeping tip clearances tigh basically impossible. I design high speed shaft systems for my job so while I can't comment much on the blade geometry, I can say that using two much smaller bearings (I'd guess 4-6mm ID) spaced a short distance apart will provide much better angular constraint with mush less resistance. Deep groove ball bearings like that are pretty much always used in pairs to constrain an axis as each bearing constrains a point. You will need a steel shaft though, as a plastic one spinning at these speeds is unlikely to be stiff enough to avoid resonances.
With that said though, it's still an amazing achievement to even make a functional turbopump and injector, nevermind to make it with 3d printing
You could pour a slurry of heat resistant ceramic into the bottle and then spin it at a high speed to so that centripetal force will coat the walls evenly. Or better yet just paint the outside of the bottle and let the Plastic burn away with use. Integza did this a few times with 3d Printed parts.
This is incredible! Way to go!
Turbomachinery is hard when we do it for a full time job.
Me everytime I introduce my "revolutionary" college project to the judge: 0:07
“The difference between mine and the real ones is that the real ones work”
The most important thing to know when designing a pump is how”
Immediately subscribed.
How do you only have 28k subscribers?! This is some of the coolest and best presented engineering on TH-cam
WOW! I didn't even notice! Geeeeze You better be on the Up in up because this is awesome.
TH-cam likes to promote junk that make people dumb so this channel has a long climb ahead.
he only has 28k because it's a niche of building amd 3d printing with a niche within a niche of making rockets, which is not for everyone
@@VerbenaIDK lol explain smarter every day and integza
@@willhaney96 integza is same thing and pretty sure smarter everyday has a bunch of science in it, whoch not all people will br interested, idk about mehdi tho
maybe because he almost kills himself every video, that man is lucky he's alive, he bamboozled death at the very least once
One thing that might be worth looking into, that would definitely make this way easier from the sealing and fluid dynamics POVs, is going with an electrically driven pump. Small BLDC motors are available with frankly terrifying power densities, and your pump isn't putting out that much hydraulic power (volumetric flow rate times pressure) anyways.
Exactly what I wanted to recommend after watching this video.
While improving the efficiency of the turbopump is certainly possible and should be done either way, using drone motors and maybe some VESC based programmable speed controller would at least be far easier to get working.
However if the goal is to use the developed technique in a model rocket to actually fly it the weight considerations of using a motor with lipo and vesc come to mind.
Hehe, I was just watching your throttled solid rocket motor video and thinking it would be neat if you made a turbo pump system.
I then looked down and saw this.
- Was the hotdog tasty?
- Is it safe testing the fire over a patch of grass like that LOL!
- Amazing to see how a "relatively simple" injector made such a big difference!!
perhaps try using both turbos in series making a multi-stage turbo
The hotdog reminds me of the time that I was smoking a pork shoulder, and I accidentally spilled some of the fat drippings pan on the floor of the smoker. It was maybe 3-4 ounces. It was night time, and it burned so fast that it actually flash blinded me and I couldn't see. Grease and fat have a tremendous amount of embodied energy!
So glad I stumbled on this channel! Please keep the 3D printing and project videos coming!!!
I'm competing at the Texas State Science and Engenering Fair next week. My project was a 3D printed alchohol and oxygen rocket. Iv had the Idea of 3D printing a rocket engine since my school won a 3D printer in a grant my freshman year. While researching I came across your channel. You've been a great inspiration in project and I just want to say thank you.
Well, how'd it go?
@@FreshAsFxck didn't place, only top 3 in their category know what they placed but I was at least 27th in state (in my category) because thats all that was competing at state
5:20 when your pump is so efficient you struggle to measure how efficient it is xD
There is another type of turbopump that works slightly differently; instead of using the flow of oxidizer to move the fuel (or vice versa) directly, it runs the oxidizer (or fuel) around the walls of the rocket bell for cooling and then uses the expansion of the respective gas (due to having picked up heat) to drive the pump, which then extracts more fuel/oxidizer to feed the cycle.
While you are completely correct a few engines do that it will not work here as that system needs very high heat to work.
Your projects continue to entertain me, excited to see all these concepts put together!
The only way that i could imagine to solve both the combustion chamber melting and reduce the backpressure into fuel line is to reduce the ratio between fuel and oxidizer until you get a workable combustion pressure.
Another way to solve, at least, the pressure is to assembly several pumps in series to increase the fuel pressure.
You could combine both methods to create a rich oxidizer mixture by feed parallels pressure air lines to run the pumps, witch would be assembly in series into the fuel line. Maybe you could use this air to cool the chamber by put a bottle inside another and run all of this air between thoes bottles or something like that. You are creative, you figure out how to do it.
that injector is genius!
I don't have much experience with turbo pumps myself. However, what I do know is that there seem to be a few optimization problems worth noting, and possibly writing home about. Fist, dependent on the direction of flow, several things are happening. The fluid must fit in a given volume generating pressure, subsequently transferring the force to the angled fins turning the impeller and thus collecting angular momentum. The reverse happens when external mechanical force is applied to the impeller. Something that stands out - "as I observe the functional impeller device", is that the flow of fluid usually travels around the toroidal casing of most turbo pumps I have witnessed. This process requires that the fluid be traveling at a minimum velocity around the casing to fill the volume, generating pressure. However, the impeller also requires a greater amount of energy to rotate than it would be able to provide as conserved mass. Several improvements I invasion for such a device include the following. { A slightly longer impeller that protrudes through the inlet/outlet. }-{ Slightly longer fins that provide a better energy collection ratio. (The space between the casing and the impeller.) }-{ Revising the angle and geometry of the fins to optimize for
really appreciate the effort yo put into these videos, i must say the way you explain whatever you are explaining is very intuitive and fun. keep up the great work!!
Awesome project and results! You should look up "velocity triangle", it's a key feature for designing decent turbomachinery.
there's tons and tons of literature around turbochargers used in vehicles.
One thing that's important is that the turbine side is smaller than the impeller side in cars.
Biggest thing to mention, is static pressure from a centrifugal pump is very different from pressure at flow.
could you try adding that roach killer that makes green flames when added to alcohol when you do the live fire tests. just think it would look cool. great vid keep em coming
IDK how practical it is to 3D print, but on high pressure steam turbines they use labyrinth seals to reduce friction. This is an alternating series of circular lands and grooves that mesh together with close tolerances, with each groove on the shaft meshing with a land on the stator and vice versa. This forces the fluid to zig-zag along a much longer path, and the path is designed to be long enough and tight enough that fluid friction stops it from leaking all the way down the path.
So impressive! I love how you used graphics and animations, they made it really easy to understand
It will help to start with eulers turbomachinery equation on the turbine and pump to get a very rough idea of what diameters you will need for an operating rpm that won't break the printed plastic but might hit your pressure requirements. Then simple velocity triangles at the inlets will help get the blade angles close-ish. Your first pass worked really well though on the first try though!
I think you are the first TH-camr making DIY 3D printed centrifugal pumps that actually used the term "turbopump" correctly. So many people think that any centrifugal pump is a turbopump and completely neglect the fact that "turbo" means it is powered by a turbine. But you actually made an honest to god turbopump, turbine and all. Way to go man! This is so freaking cool!
Love the design! As for the combustion chamber, you could maybe 3dprint it with thicker ablative walls (abs also makes a good fuel if you're not afraid of the fumes) and/or cooling channels to route air and fuel through? It would free you from the limitations of the 2L plastic bottle while still remaining a plastic rocket engine.
Loved the video mate. Can´t wait for the next one.
I'd think about a staged turbo set up with an identical one downstream and run the Air into it first then up to the 1st turbo from the fuel side and then out to the combustion chamber so the higher pressure fuel has the highest pressure air.
Nice vid. The hotdog probably glows bright yellow because of all of the salt (sodium).
Regarding the hotdog grease :hotdogs contain quite a lot of sodium, which colours a flame bright orange when you light it on fire.
Cool project. Subscribed.
Was searching through the comments to see if someone had already mentioned this.
Gosh I love this channel! Hope you'll get to 100k by the end of this year!
Very cool. Fun project and I love all the helpful comments here. I think youtube recommended this because I'm interested in 3d printers. Awesome to see what you're able to do with one.
Seen a few Tesla valves printed off, along side two different size pipes for fuel and air might solve your back flow issue.
For the impeller you can use a bigger diameter to increase static pressure, since the centripetal force will get higher.
Then you might also need a bigger turbine to get more torque...
You can also vary the amount of flow by making the insides of each lower, depending on what peak flow output you need for a perfect mixture.
Use 4 thinner holes at the injector to get more surface and constrict the airflow with a better nozzle to get faster airflow and both together can improve the mistification process much more effective.
Maybe even give the nozzles a shallower angle to get the mistification startet earlier.
And the bright orange flame is indicative of sodium 👌
This was a pretty great video to watch to be honest
If you're looking for people to talk to about 3d printed turbo compressors, Indeterminate Design designed one from scratch here: th-cam.com/video/1fVT1okcIEc/w-d-xo.html
He can probably point you in the right direction from his research.
Dude, amazing design and video!
9:30 I would assume that the sausage is quite salty. Sodium ions create a really bright yellow/orange glow in flames and the color matches this.
You could try a tesla valve or some other type of one way valve on the fuel side to keep flashbacks from happening.
Your forgetting the combustion ratio or fuel to air ratio for optimal combustion, propane is like 16-1 16 parts air, 1 part fuel, your air fan would have to be 16 times the size of the fuel pump for optimal combustion, and one way valves will solve the problem of air going back into the fuel line
really cool watch! Had no idea the resin prints could do that!
This is awesome! I’ve recently been hitting the books pretty hard on turbo and compressor design. We are also taking the same initial approach: make something that looks right and see where it’s actually wrong. Subbed and stoked to see what kind of
mayhem results from this, hail MOTOR!
Wish you uploaded more often, your videos are dope!
My intuition says the diameter should be bigger on the fuel side. I can't wait to see if that proves correct. Although I suspect this will also have to be multistage.
Amazing work as always!
Make the compressor side bigger then the turbine side. That should extract more energy out of the air at the cost of more back pressure / less air pressure after the turbo
This is some great stuff! Before I say this, do know this is the first video I saw of you so this may already be out but a overpowered water pump that will shoot across a small lake
185 feet!!! That's one incredible small water pump. Dude you should make this into a 12 volt accessory if there is a way to get the air pressure to a reasonable level that a small tire inflation pump could handle. Then it's just an accessory you can buy to pump fuel or water. Super cool man!
Wow, this channel is was under-subbed. Solid quality content. Reminds me of a cross between Integza and Michael Reeves. But he’s much less vulgar, so I feel comfortable sharing his content.
Maybe more fins on the air side of the turbo whilst making sure as little as possible escapes around the fins. Much love
I just found your channel and super impressed! 👍🏻Maybe a small tube coming of the fuel rail after the turbo going back into the fuel tank to keep it positively pressured . This may help stop the pressurised air from feeding back up the fuel line!
Look at the way that actual car turbos run smoothly.
ideas for optimization off the top of my head: gear + ratio
i'm extremely surprised how well both the turbo and the nozzle work
you are interested in applying a check valve in your fuel line between the turbo and combustion chamber.
You need a cricut for cutting those plastics more precisely! Also I'm sure your SO will use it a lot too
Amazing! Cutting the fuel supply while maintaining airpressure for a few seconds at the end of the burn might help to prevent the flame travelling back to the injector.
Use a tesla turbine on the air side.
Compress the fuel in the tank separately.
Or ask JohnnyQ90 for a better turbo design.
Remember that you need high pressure, not flowrate! Youre pushing everything through pinholes. It it will always spray hard, but pinholes are smaller than a hose.
you should get a check valve for the fuel side, but thats basically what an injector is, but could allow for the pressure to first build up, before opening up, and releasing your desired fuel pressure, and then no need to have to mess around optimizing turbo pump, and especially to after you said you have like 150ft of tubing to play with. You can get that punch with out having the length.
Maybe there is a cheap way to cool the bell-housing by circulating the fuel/air around it? Like they do in actual rockets?
Though you'd have to find a way to do this cheap & light-weight.
Maybe using some light-weight 3d printed part?
Another great video! I wish the frequency was a little higher, but it's nice having higher quality! Thanks
I would recommend looking into a means of measuring the dynamic pressure the pump can generate. Centrifugal pumps are not great with static pressure. I'd also be concerned about cavitation on the fuel impeller eating it up in a hurry. You may want to experiment with different ratios between turbine and compressor size and design, as well as their respective housings, but unfortunately my flavor of engineering never got into pump design, just specification. Super excited to see this go forward though!
One idea that works is to hook the air compressor to a soda bottle than can withstand high pressure. Then you have a different hose which pumps the fuel out. Basically you pump 80psi air in which pushes fuel out of the bottle at 80psi. You will need another compressor, though ive tested this with cheap tyre air compressors and it works. Its the same principle as fuel pumps, which cause a high pressure in the fuel tank which forces the liquid out of the tank through a hose
Awesome results!
Maybe make a high-flow one way valve? Maybe use a wuad copter motor to drive instead of air? Maybe get a flow simulation for the pumps?
Without much diving into cad files I would think you need to adjust the AR ratio of the housings or possibly the housing shapes. I expect it's operating as a simple pump and not actually compressing much at all. If all you're getting is 20 psi, the geometry is off somewhere.
that is an awesome idea! I love it!
Every video you make takes me one step closer to buying a resin printer.
You could 3d print a ceramic based combustion chamber and nozzle. Integza does this often with good results.
FYI, the other way you can check pressure is to have some air in the tube on the end of the fuel line and plug up the line. When it pressurizes the fuel will push up the tube and compress the air. Based on the air properties you can calculate how much pressure was required to compress the air that much.
if i understand correctly the way to increase pressure on the fuel side of the turbo would involve either increasing the speed at which the air moves through the turbo or increasing the size of the air side of the turbo allowing for more airflow. in rocketry it's hard to get one turbo that properly feeds both the correct amount of air and the correct amount of fuel and so they have two turbos, one for fuel and one for oxidizer and each turbo pump runs off of a preburner that runs either oxidizer rich or fuel rich depending on what it's pumping. if you're serious about using turbopumps for your rockets you're probably going to want to borrow the idea of a preburner assembly. depending on the pressures of your rocket you could pipe the exhaust of the pumps into the burn chamber or i think what nasa and other rocket groups do is exhaust it to the bell assembly. godspeed fly safe!
one way of improving it would be to use a steel-ceramic bearing and if it is sealed then to remove the seals and grease and lubricate it with some thin oil, to reduce the friction in the turbo pump
You could use something like starlite, nighthawkinlight talked about it, as an abative coating on the inside of the 2L bottle. The only issue would be needing to water it down and spin it until dry and that it'd be opaque, so you wouldn't be able to see inside it.
Might be neat to make one with a pre-burner instead of using air pressure somehow.
The maximum prestige fuel side of the pump can produce is likely higher than the gauge read. If the pump’s flow is completely impinged like it was in your test, the impeller will “stall” for lack of better term. Pumps like that make maximum pressure when there is a balance between the restriction and the flow. The channel Indeterminate Design talked about this a little bit in his video on a two stage turbo.
electric turbo pump, remove the input side dependency, easier to bootstrap, but you could do hybrid-solid NaClO3 - carbon hybrid drill cylinder rocket, two part, separated, gas-solid hybrid
If you're happy using stored pressure in the oxidizer to pump the fuel / the pressure from the air is enough to overcome your chamber pressure, you can just use a venturi injector. it would also mix the fuel.
You will get higher output pressure if the compressor impeller fins are perpendicular to the circumference, sweeping back lowers pressure but improves efficiency a little bit. You need tight tolerance with the housing to get high pressure too.
Stop thinking of it as a turbo pump and start thinking of it as a gear ratio. If a symmetrical drive shaft (same impeller as propeller) shares your incoming pressure 4:1, then make the drive side have 2x the impedance. Then, your combustion air will be at half pressure and your combustion fuel will be at double pressure, and you'll be even.
I don't know if that's actually how the math works, but maybe it's still useful.
Also of note, the flow will necessarily decrease with higher pressures.. so once you max out efficiency, you basically have a sliding scale of flow/pressure for a given input flow/pressure..
maybe you could metalprint a bell? or even ceramic print it
I don't think the pressure in the chamber will send stuff back up the fuel line. Because if the pressure gets too high, your fuel coming out will be reduced and it will be a feedback loop that will never allow it to go backwards. If you're very concerned, try a check valve?
so fun to see this stuff!
Alternate title: DIY 3D printed flamethrower
One thing that would improve the efficiency of the turbo is to make the inlet and outlets tangent to the turbine. The turn it makes from going from tangent to radial causes a loss in energy
What about using two stages of turbo pumps connected in series to increase fuel pressure?
7:42 yeah. Sure mate. I definely believe you. And I love that thing 😂
You could potentially run a mechanical seal on each side, this will keep the "products" away from one another. Another advantage with using the mechanical seals would be non contamination of the bearing by the "products". It looks like the shaft size is about 10mm, mechanical seals this size exist. I think it could be an awesome upgrade for a project like this! Either way I think this is a really cool idea
Awesome! That looks like it works really well! I think the main area of focus (at least from just watching the video), might be a better solution for the seals. 🤔
Use check valve in fuel line to avoid backward flow of fuel
try 3d printing a small regen cooled chamber with thin walls for great heat conductivity
Holy shit that injector is effective 👏
I'd imagine the color of the flames coming off of the hot dog is from all the salt in there.
I haven't tried it yet, but I installed a FreeCAD plugin the other day for turbine blade design. If you also install OpenFoam you can do fluid dynamic simulations. If it is like the finite element analysis side of FreeCAD, it is actually pretty easy to do assuming directions. I was thinking a quick and dirty way to model pressure might be to treat it like hydraulics or pneumatics where the cross sectional area moving the fluid is the area of the piston. If it works at all I suspect it would be fairly inaccurate because of all the leakage, and other reasons. Maybe the best thing would be to keep the air motor constant sized, and print a larger pump. Take many measurements of the existing one and the new one and see if a pattern pops out when you graph stuff. When doing that sort of thing, I like to use prime numbers 3 or larger for sizes so you avoid stuff like 2+2=4 and 2*2=4 and 2^2=4. Anyway, try making your own empirical equation.
And as a silly side note you might not have noticed yet, but water column height is a common unit of measurement in some fields so there are simple conversions for switching to psi, pascals and etcl. Sometimes it is inches of water, sometimes feet of "head" or just "head." While the static pressure is too large for it to be easy to use, it can be handy in other situations. Like you might want to find the pressure drop between two locations in a pipe or duct, kinda like voltage across a resistor.
Awesome content!!! It might be interesting to see if a duel turbo set up would make any improvement. You could experiment with series and parallel connections on the fuel and air sides.
Love this stuff man! Send something to space in the next video
Great video!
I think pressure you fuel reserve with the air, kinda like a blow through carb set up when turboing, you apply. Boost pressure to the float bowl