I Built The First LAMINAR FLOW ROCKET ENGINE

3D Models

Materials

Machines

#Rocket #engine #3dprinting

[jeppelarsen]

Hi Integza, I'm a research assistant in a research group researching in combustion engines. Personally I specialize in injection of liquid ammonia fuel into into IC engines, and I'm an old fan of rocket engine design! For me, rocketry is the ultimate engineering problem, combining so many engineering fields together!

Let me just say i think your concept is absolutely awesome in the department of OF mixing.

But in regards to the feasibility in real applications, I can foresee a cooling problem of the internal mesh if you are thinking of pushing the engine to its max capacity reliably every time to achieve the ultimate goal of maximizing thrust. Ultimately, the higher temperature you can create in the chamber, the better efficiency and the more thrust available, thereby creating the beautiful trade-off balance between efficiency and structural integrity. This is also one of the reasons to use pure oxygen, because then you don't waste a lot of energy heating up the 79% nitrogen in the air. Make it with pure oxygen! 😂😅🤪

(I really hope we get to see the divergent nozzle as well in the future!)

I'm quite surprised that you did not use your software to create a regenerative cooling jacket around your nozzle chamber, maybe I have missed some details in this regard? If you decide to run with pure oxygen some time in the future, maybe you could consider using the liquid Butane in the flask to run through a cooling jacket in the outer wall of the rocket engine. Instant cooling AND potential higher fuel pressure!

But absolutely fantastisk video! Super fan of your work!

Br
Jeppe

P.S. Tomatoes are disgusting!

[inoob]

Somewhere out there in Alabama, Destin is fist pumping the air

[Skatuser]

Hello, Integza!
Rocket engine designer/test engineer here.
Very good to see you using RPA, if you can please share your config, very interested to see it (your L* is about 1.5?).
Main problem with your injector type is heat managment. At low pressures, low flows and fuel-rich mixture ratios it will work, but at some point your combustion chamber inner wall will overheat and melt.(with beatifull but dangerous explosion).
To answer your question "Is this a viable way of injecting propellant into the chamber?". It is, for your specific case. Ideas of using porous walls for fuel injectors was proposed and tested since invention of powder metallurgy. The problem is - wall overheats and then combusts.

Problem with flame detachment comes from very low flowrates and as a result very small combustion chamber. You flame detachment is like 3/4 of your chamer length. I think you got numbers for injections speeds somewhere from literature, but this numbers are for normal-sized rocket engines. If you want to go with such low flowrates, next time search for "rocket engine igniters", some of them are very low power and low pressure rocket engines, exact thing you are searching for. Also you can cheat by using tangential flows.

If you want to go in this field, you need invest some money and time in your instrumentation. Your way of raising butane pressure was very dangerous (I think you know it yourself, and if someone saw that in my test facility, it will be very bad). If you need more pressure, you can switch no natural gas in cylinders, they are about 13 bars, with pressure reducer you can get it to about 5 bars, and have cylinder installed outside your garage, with flame arrestors and purge lines, as it should be. You can use something like nodered for control interface and arduino hardware and modules for test control (Won't recomend NI hardware:) for it's huge price, therefore arduino).

For your next designs invest in modular water-cooled jacket. It will make your experiments more safe, because your walls will always be as cool as water.

If you have questions, feel free to contact me. I would like to discuss your safety measures, your instrumentation and data collection.

[JoeTheScientist]

Professional carpenter here. I can definitely confirm that is metal and not wood.

[Jesse-sfhh]

Coffee machine technician here, that is probably too hot to make espresso. Carry on

[nikaross]

Professional rocket engine designer here (and fan of your videos). Interesting idea using gyroidal geometry for injection. Very creative solution to a common problem with small liquid injectors. Without access to super expensive equipment to hold tight machining tolerances small injectors are very difficult to get good mixing out of and it's easy to run into a lot of stability problems like a lot of your attempts at more traditional injector types (impinging doublet, coax swirl, premixed showerhead) did. The gyroidal geometry will create a lot of turbulence and disruption in the flow down the combustion chamber which significantly improves residency time and mixing (both allowing for more steady combustion).

Before I get into why the gyroidal injector geometry likely isn't a great I thought I'd point out that there is quite a bit of research going into 3D printed gyroidal geometry as a base for catalyst beds for monopropellant thrusters. 3D printing gyroidal geometry and plating or coating the high surface area bed with a catalytic element can result in a high specific surface area and a low restriction which results in better monopropellant performance for catalytic decomposition like the operating principle of a lot of the industry standard monopropellant hydrazine thrusters.

Anyway, at the performance levels required for getting things to orbit, I foresee a couple of issues with the gyroidal injector element design. Namely thermals, pressure drop, and performance issues.

Thermals: Generally, the performance requirements for liquid rocket engines require the combustion temperatures to be well above the melting point of any metal. This means you need to have mitigating circumstances to contain the combustion and any metal the is in the flow will likely erode away very quickly (like the gyroidal injector). The propellant flowing through the injector would likely provide some cooling, but it is almost certainly not enough to prevent any material from eroding very quickly.

Pressure Drop: In a rocket engine, pressure drop from the highest pressure point to the combustion chamber is ideally minimized (excluding stability and throttling concerns). This is done to prevent losses which benefit nothing to the engine efficiency. If your supply pressure is limited like in this case, this results in lower chamber pressure and thus lower performance, and likely a less stable nozzle flow. In a context with pumps, this would mean that more input energy is required for the pumps to achieve the same chamber pressure as a lower pressure drop design (meaning more wasted energy).

Performance: This one is kinda complicated, so I'll keep it a touch higher level. The high-level explanation of a rocket engine is that mass is accelerated as quickly as possible out the nozzle which provides a reactive force. Efficiency or performance comes from what percentage of the heat energy is converted into kinetic energy in the correct direction. Many things can effect this like exhaust products, combustion temperature, nozzle geometry etc. Something else that can effect this is to create more turbulence than required in the combustion chamber. Any swirling or non-axial flow effects result in energy losses and a lower exhaust exit velocity. The gyroidal geometry creates a lot of these non-axial flow effects.

Unsolicited advice on if you attempt to approach this again with a more traditional injector design:

1.) Start with your constraints to work out your engine parameters. In this case it would likely have been the propellants you're using and the supply pressures from your tanks.
2.) As a general rule of thumb, injector stability should be fine at dP - Pc ratios of ~10% (dP - Pc ratio is the ratio between the pressure drop across the injector and the chamber pressure). In this case, that would mean a 10% total drop from your supply pressure in the chamber. I'd probably guess the chamber pressure should be a touch under 1.8bar.
3.) From the above, you should be able to use the combustion byproducts to figure out a good throat geometry (use the combustion byproducts to find the gas constant and the speed of sound, from there work out the throat diameter and decrease it a touch to make sure the flow gets to M=1 with the efficiency losses from imperfect combustion, chamber wall losses etc.
4.) Double check your numbers with RPA at a few different OF ratios and mass flow rates to make sure your rocket will stay choked over a range of your possible operational conditions. (I would go for a smaller throat than RPA needs due to inefficiencies stated above, and a throat that isn't sonic results in a 'fast candle' with pretty minimal thrust.
5.) Once you have the above, move onto an injector element design. The next bit is a bit dependent on your element choice, but there is literature out there for a lot of them (old NASA papers go hard). Essentially, you're aiming to get your pressure drop right, while making sure your flow rate of each propellant is correct for the OF and mass flow rate you're trying to target. With a propellant combination of air and butane it's about 15:1 Air:Butane by mass ratio for stoichiometric combustion. Good practice is to try and run rich of stoichiometric for better performance and even richer if you're aiming for low temperatures. If you're not looking for too much performance I would look to run at about 11:1 or so, will help keep the heat down. Each element type has other considerations to try and optimize mixing (momentum ratios, cone angle, jet angles, etc.). Another tip could be to try and use some film cooling to keep thermal issues away (just very tiny holes around the outside of the injector head that just get fed fuel)
6.) Next thing to consider would likely be the length of your chamber. Ideally you want to minimize this to minimize heat loss and potential longitudinal instability issues, but it needs to be long enough to allow for proper mixing. Again, I'd recommend checking literature for your element type (gas-gas flow is usually pretty low chamber lengths required)
7.) Send it (safely, please use appropriate pressure vessels for your propellants)

Side note on point 6, the raptor engine uses really high pressure, hot-hot, gas-gas mixing in it's injector elements (the injector uses the exhaust from the pre-burners/turbines). This means the mixing and combustion happen super quickly and they can get away with using a really short chamber. This is likely partially why you were having issues with the exact replica as the propellants didn't have time to mix properly before entering the contraction at the throat.

p.s. or maybe I'm not a rocket engineer, don't believe everything you read on the internet

[flyinn]

hey integza! NASA engineer here. i can confirm that the rocket engine in this video is metal

[maxk]

Hi integza! Huge fan here!

So to cut straight to it, I'm a propulsion engineer, specifically I work on liquid rocket engines, and I have a very very important warning about your design!

The porous injector idea is fascinating, and tbh I didn't expect it to work as well as it did, but it also poses a safety risk to you. In the region upstream of the porous injector you effectively have an enclosed volume of premixed fuel and oxidizer that, if the engine gets hot enough, will spontaneously combust before ever reaching the combustion chamber itself.

However while the combusion chamber has the nozzle to allow the gasses to flow out, this volume does not. It would most likely result in an explosion within the injector inlet volume. Best case scenario is it blows the hoses off. Worst case scenario is it shatters the thin and brittle (strong, but still brittle because it's 3D printed) walls of the injection volume and sends metal shrapnel flying everywhere.

Please stay safe, and looking forward to your next video!

[TheWhiteDragon]

It really has been absolutely wild that this last decade has produced such incredible advancements in metal 3d printing.

[RStagarts]

Delivery driver here and this was pretty awesom.

[aBradApple]

I'm just a former aviation maintenance technician but you answered a question that my crew mates and I had for a long time. The Harrier's Pegasus engine had a showerhead flex hose (about 3 in diameter) which fed fuel into a check valve from the fuselage (where fuel and outside air pressure meet). I never understood why the inlet required such an interface but I have a much better idea now. Thanks for the insights!

[connor_mckinley]

Hi Integza, Professional Youtube Commenter here,

I see some problems with your rocket in that it is attached to a table and not something that has wings, perhaps this is a design flaw that you could not have foreseen, however it is vital to the design of a rocket motor that it be attached to something which can bring people to space, or even perhaps monkeys. Come up with some news designs and make a new video by next Thursday, thank you.

[HexerPsy]

I like how this attracted some real professionals in the comments :D It was worth the read and the watch!

[harleybmx]

I'm not rocket scientist and do not know very much towards this subject, however I do play with jet engines, and the biggest no no towards small turbines is the use of pure butane, it doesn't provide the heat required nor the combustion, using an ISO/butane (60 butane/ 40 propane) you should definately see different results, unless that is what you use. The properties of the fuel that is used is key to the magic of internal combustion in both jet and rocket applications.

[connarcomstock]

"If there's any kind of specialist"

A man is looking for a literal, actual, rocket scientist.

[Talik]

It's not every day that someone just stumbles across a super cool engineering feat that nobody tried before. This was so dope

[ErinMoth]

This is actually fucking crazy, you have an engine that could easily be self cooled, cryogenically too, that could possibly use HHO det gas, and have a constant detonation, not a rotdet, not a pulsedet, just the det. This is insane just from the fact that a civilian could make something with an ISP of roughly 1.3 km/s, be self cooled, and be roughly the size of a water bottle. I love this.

[kennyhagan]

Strangely, the more organic looking the components are, the better the performance is. I am impressed.

[CastleRene]

7:12 Thank you, sir. We appreciate your transparency.

[skyeline]

@integza
Safety Engineer with a degree in Aerospace Engineering here:

NEVER stand next to an experimental rocket motor during testing.

If there is an error in the design, or something unexpected happens, that rocket motor will explode.

Especially with complicated internals like this, if an inside part melts and clogs the nozzle, that combustion chamber will become a bomb.

Please put more safety disclaimers in your videos. Most people don't have the tools to make these, but you don't want someone blowing themselves up because they tried to copy you.