Liquid Rocket Engines 6: Nozzle

Today we talk about how we design the nozzle!
Links:
Arocket Mailing List: www.arocketry.net/forum.html
Translation of Ievlev Paper: apps.dtic.mil/dtic/tr/fulltex...
Nozzle Design by Method of Characteristics: www.researchgate.net/publicat...
Patreon: / charliegarcia

[@CopenhagenSuborbitals]

Will be nice to compare TEOS results! :)

[@whcolours9995]

Charlie Garcia: That one guy that decides to make a rocket engine and shares with anyone who wants to make one themselves.

[@Squeezmo]

probably looking for advanced engineering work in something really cool... 3D printed nozzles.... wow. saw this coming years ago. add forging/cold rolling to the printing process...

[@howardbartlett3419]

As someone who has worked on (albeit significantly smaller) liquid fueled engines, you have done a fabulous job of explaining all of the physics behind your decisions and the mathematical design constraints. Thanks for making these videos!

[@YukonK9]

Walk and talk in Boston with a MIT Rocket Team shirt talking about rockets, Awesome!

[@Squeezmo]

an MIT physics prof's estate has his lectures free online. He's fun to watch... sounz like a stereotypical mad scientist...

[@YourLocalFBIAgent-gl7pu]

I searched for long time for this information.

[@theelectricwalrus]

Commenting for the algorithm!

[@antoniomontana9262]

Just back from a vacation can’t wait to catch up on your channel.. thanks for the content!

[@CanineDefenseTechnologies]

Charlie we envy you mission patch wall very much, but perhaps use some aerospace grade adhesive?

[@vincentbelpois2600]

Such a nice project, I love the approach you take and how you explain the tools you used and the ressources ! Keep up the good work !!

[@mmb3006]

i would just like to thank you for making these videos and because im currently in my first year of high school i have bi way to acess this sort of data and it really helps me to learn im also teaching myself some cad programs currently using fushion 360 and i hope i get enough knowledge to be able to make something similar o a rocket engine in cad untill the end of the year and thank you again for making these videos they must take a lot of time to make and i know how busy you must be but youre still here to teach us nothing but respect for you!

[@RyanDobler]

I think you may have accidentally put 2 takes of the same part in there talking about how your cooling channels were designed too large and you had to reduce the cross section.

[@AstroCharlie]

Oh well. Sorry folks!

[@RyanDobler]

@@AstroCharlie shit happens, as they say. Keep up the excellent content, I look forward to the knowledge you share!, Also, as a machinist myself, I'd love to see the machining work!

[@AstroCharlie]

I'd love to show it. I'm about two hands short though. If you have any pro tips on how to film machining/welding/angle grinding everything let me know!

[@RyanDobler]

@@AstroCharlie i use a GoPro or similar with a clip mount to attach it to the machine watching the action.

[@MichaelLloyd]

@@AstroCharlie Check out Clickspring, Joe Pieczynski, OxToolCo, This Old Tony, Abom79, etc for examples of filming machining. In your spare time... and Mike Patey (his carbon fiber work is amazing)

[@vineethba1503]

Thanks for the video Charlie.... lots of knowledge u have shared which is very valuable ... 👌

[@Spirit532]

This project is getting quite interesting! Hopefully we'll see a rocket engine firing by the end of it! :)

[@anglerfish61]

Im invested in this. cant wait for the next vid

[@Turcian]

Nobody knew h̵e̵a̵l̵t̵h̵c̵a̵r̵e̵ rocket science could be so complicated!

[@aristeidislykas7163]

Can you find a way to cover the inside wall of the throat and nozzle with a thin film of Beryllium Oxide or Silicon Carbide?

[@TheNadOby]

Regarding electroplating/electroetching there is a possibilty to use 3dprinted FDM/FFF party's as a guides for etching electrodes. That's make the process accessible and sufficiently precise.

[@miguelribeiro5165]

Watched your video to learn for a university work -> watched Joey B cooking show -> remembered I saw you somewhere -> Came back to this video to tell this :-)

[@spoot]

Non-scripted content is better, feels more like a conversation versus a lecture, and feels much more natural.

[@theelectricwalrus]

How are you going to add the fuel film cooling into the nozzle? Drill holes in the nozzle?

[@AstroCharlie]

Probably just on the injector face

[@YukonK9]

0:34 Apollo...A Easter Egg or Coincidence? We shall see

[@nirodper]

what are the advantages of copper compared to brass or bronze, considering that the latter ones are easier to machine?

[@AstroCharlie]

Copper is more than twice as thermally conductive as bronze, helping to carry the heat into the coolant. Copper also has a higher working temperature than bronze or brass by over 200 degrees.

[@Squeezmo]

make a mold and cold roll copper tube onto a slug for the inner and outer sleeves *use etched-out slug for in between sleeves (like lost wax).

[@raphaelsantanasouza]

Hi Charlie, very good initiative! You stopped the liquid rocket engine project? Regards

[@fgarza]

Charlie, please finish your thought on the Rocket Propulsion Elements Book, I recently bought it and I'm actually studying it so I can build my own Liquid rocket engine and start testing it. Also a note on the nozzle throat, have you given any consideration of carbon foam?, there is a company that builds and machines it per your specifications (can't remember the name right now) and for what I see, it has a much higher melting point than stainless steel. Maybe you could design it as an insert while you are testing your engine.

[@saimafa5579]

Hey, how is your rocket engine building journey so far. How was the book? Is it useful?

[@mmb3006]

Hey I don't really understand how you prevent leakage of fuel into the combustion chamber without welding the parts together I was looking at your 3d printed steel and what I think I saw was jou just put the parts together and screwed them from the outside and from the inside you just made them fit really nicely but wouldn't that still have a chance of leakage of gasses from the combustion chamber into the fuel in the nozzle or the other way around?

[@vaibhavrajpaneru9302]

pls make a video on Impinging Injectors and how to design a combustion chamber

[@johnortiz9789]

Hey Charley, great video! Just wondering about the dynamics of flow... if you over expand a flow through a too big of a nozzle, does that mean that the pressure at the exit is higher or lower than atmospheric?

[@nanocompa2411]

Too much expantion would mean the pressure is too low, atleast with supersonic flow. Other way around with subsonic.

[@jshatfield115]

Expansion is expansion, sub or supersonic. Overexpanded means that the flow pressure has been reduced to below the ambient pressure. This isn't possible to do at subsonic velocities bc the backpressure CAN be communicated to the expanding flow. So it will not be allowed to overexpand.

[@johnortiz9789]

Justin Hatfield that’s was a great explanation! What type of means or tools can you use to measure the flow velocity at the nozzles exit, at an affordable cost. I believe a Mach meter can be use, but I recently learn that a pitot static tube diameter should be 1/30th the diameter of the channel flow. Something virtually impossible in my project, since I work with a 1 in diameter nozzle exit for my thruster.

[@jshatfield115]

@@johnortiz9789 that is a can of worms. I would first try a pressure and temperature transducers but you would have to assume flow to be perpendicular to the pressure transducer or your results would be basically meaningless. You could put it on a test stand and measure the thrust, and then find the mdot and find it that way!

[@jshatfield115]

@@johnortiz9789 let Charlie know you are confused. He made a tiny mistake at 8:43 in this video. He meant to say "lower than ambient pressure", but said "higher than ambient". Even the sharpest engineers mix up this terminology often.

[@electrolysisresearch8013]

I have a Large Electrolysis cell I Built for welding that makes 65CFH of a 2 to 1 mix of Hydrogen and oxygen at up to10psi. Is there a way to make a Type of rocket with that amount of pressure? I was thinking to make a combustion chamber with a flash port, And hit it with a spark plug at around 60Hz to make a bunch of consecutive explosions. Do you think any of that will work or do you have any suggestions?

[@gregwarner3753]

Buy a couple of compressors, one for Oxygen service, and store the test fuel gasses in SCUBA tanks. Place them on recording scales so you can measure mass flow rates.

[@skyzip4k171]

Nice

[@MrSirPain]

Hello... I am working on an aerospike engine. I was wondering if you could take a look at and let me know your thoughts about it. Thanks

[@AstroCharlie]

Hello, you may find my contact information via my website: www.garciasystems.space/

[@anshuman_eek]

Why doesn't the entire propellant within the tank start combusting at once? I mean when rockets start ignition sequence where they pump propellant, why doesn't the spark run through the entire tank leading in an explosion??

[@AstroCharlie]

Importantly, the propellants are not mixed before you inject them into the combustion chamber, but especially on the oxidizer side this is a very real problem that requires careful design of the pressure drop leading into the chamber (the fuel and oxidizer are at a higher pressure than the combustion) as well as safe guards (one way check valves in the plumbing preventing an entire tank from exploding if this does happen)

[@mhan25]

if someone with high school knowledge wanted to design a rocket, do you have any recommendations on where to learn how to do that?

[@lukasdimmler2622]

Could one increase the energy efficiency of a rocket engine in the atmosphere by increasing the mass flow with air from the outside? I'm thinking something like a ditchable ventury nozzle with the engine exhaust acting as the fast fluid and the air is drawn in from the side and speed up.

[@AstroCharlie]

Absolutely! This suggestion can be implemented many ways, such as rocket heated scramjets, fuel rich rocket with ramjet afterburner, and very complex solutions such as the Reaction Engines designs.

[@lukasdimmler2622]

@@AstroCharlie Awesome! I' ll have to try that immediatly! Theoretically you just have to ditch it before the additional drag is greater than the additional thrust.

[@MOQAZI]

Kindly, make more...

[@cylosgarage]

Is this the real Scott Manley?

[@AstroCharlie]

I'm not an astrophysicist nor am I Scottish, but I do play Kerbal Space Program.

[@minties01]

As somebody who has recently renewed their interest in rockets and space (30 years since engineering studies - not aerospace), I think you do an excellent job of explaining what you're doing and what the limitations of your project are. After this video, I'm really appreciating the challenges involved in trying to get a usable nozzle design that can be manufactured. I think that most of us lay people don't appreciate how limited rocket engine designers are by the material properties available. This has helped me understand why ablative nozzles are frequently used, due to the cooling challenges. I'm looking forward to your next video.

[@michaeljaneschitz-kriegl9598]

13:30 This may not be correct. Please check: The conductivity in copper is SO high, that no heat transfer coefficient can copmpete. It may from a cooling point of view be advantageous to increase the wall thickness in the throat, thus giving a greater surface on the fuel coolant side, which may even get extended surface to improve the cooling further.

[@dillonbledsoe7680]

Who is this genius?!?! Where did u learn man I wanna know what u know!!

[@rumixhd7213]

Are the rocket engine blueprints going to be open source ?

[@daniele.2944]

Are your python scripts available anywhere?

[@thomasrichardson8327]

I was hoping you would continue giving the depth of detail through explaining how the nozzle converts the thermal energy and pressure of the particles into velocity through the nozzle but nope.

[@nanocompa2411]

The exhaust velocity of a well designed turbofan is waaaaaaay above 4 m/s. Most subsonic engines have an exhaust velocity right below the speed of sound. However i dont know if you mean the exhaust velocity after exiting the combustion chamber or the actuall outlet velocity after exiting the nozzle. Intrestingly with jet engines the mass flow coming into the engine is allready travelling at whatever airspeed you are flying at, so you will have to add even more speed to this air. The Delta here is what actually creates thrust. so M*(Vout-Vin) whereas rockets accelerate the gasses from a standstill. Maybe with props the acceleration is about 4m/s since the massflow is so much higher but i dont actually know.

[@AstroCharlie]

As always, it depends on the reference frame you're in, and in the vehicle reference frame, many turbofan engines absolutely do have exhaust velocities that low near their top speed. As you correctly identified, thrust is M_dot * (V_out - V_in), however jet engines still accelerate the gasses from a standstill, since they are decelerated as they enter station 1.Typically this is modelled as an adiabatic deceleration, so that enthalpy is maintained, however you then have problems in your compressors and turbines as you accelerate, due to overheating, limiting how much heat you can add. Eventually V_out ~ V_in and the engine does not produce more thrust than it generates in drag, since m_dot is about the same plus a very small m_dot_fuel

[@nanocompa2411]

@@AstroCharlie Hmm. But if you have an aircraft flying at near its max speed you also have to overcome some pretty significant resistance, if the massflow is high enough i guess 4 m/s might do the trick but youre limited by how much air your compressor can suck into the engine. And as is the case with turbofans there is bypass air that is usually routed through two stages of compression and then through a converging nozzle accelerating it quite alot and avoiding the heat problems caused by a narrow primary nozzle. of course if the flow becomes supersonic that goes down the drain. 4 m/s just doesnt seem high enough to maintain any sort of high top speed. With fighter jets flying at supersonic speeds the resistance is also much higher and the engines are usually designed for reduced massflow and higher exhaust velocities to reduce frontal area. The saab J39 gripen actually has a turbofan allthough with a very low bypass ratio. Really enjoyed the vid btw might be some language issues since all ive learned has been in swedish.

[@somber_soul]

4:23. You said isentropic and then said constant enthalpy. Isentropic and isenthalpic are two very different assumptions.

[@diruvodestp6341]

What is Battleship cooling?

[@rorypenstock1763]

at 8:38 you mean: overexpand it to *less* than the ambient pressure, right? Or is my understanding of nozzles grievously wrong?

[@AstroCharlie]

Err, what I mean is the expansion is more than an expansion which would yield ambient pressure, not that the pressure is more than ambient pressure. I think we both say the same thing.

[@rorypenstock1763]

@@AstroCharlie Oh, I see.

[@Ameliaross106]

So i am going t build a liquid rocket of my own. I am going to use gasoline as my fuel and liquid oxygen as my oxidizer. The problem is i live above sea level is there a formula or some math to figure out what my diameter of then end of the cone. I haven’t even done any math to figure out how much gas i could squeeze through my throat. I am going use soda bottles for my propellent storage. And I don’t know what PSI i could achieve. I have never made a rocket ever before so it probably wouldn’t work. I am making this for a school science fair. I am going to ask the question what is the ideal ratio of gasoline to liquid oxygen. I am going to test a few different things such as ISP, thrust, height, and max temperature in the ploom. Also I don’t know what material to use it would take much more work to actively cool the nozzle. So if i only use copper to make it would it work? Like make the whole engine our old copper.

[@Flyguy779]

so i gather you don't want to use turbopumps since they are sooper complicated, right? what about an electric pump like rocketlab? i mean it doesn't need to lift anything right so batteries shouldn't be an issue?

[@AstroCharlie]

It turns out that pumps in general are hard. I might look into them after I get the engine working, but since I've already spent the money for a pressure feed system, I'll use that.

[@Flyguy779]

@@AstroCharlie well maybe later then ^^ i suppose pressure feed is the simplest and probably most inexpensive thing in any case but if you ever decide to try turbo pumps i think using 2 iniviual electric pumps would be a nice especially for ajusting flowrates and throttling the engine

[@Guywiththetypewriter]

A factor of 2? Seems really cavalier for your first liquid fuel rocket my guy... I wouldn't want to risk flow separation or worse... "rapid un scheduled disassembly" because my exhaust pressure was way too low.... :S Seems like a huuuge risk... But good luck 😅

[@AstroCharlie]

It's not my first liquid ;) The theoretical limit is down at ~42% depending on some combustion product parameters.

[@BillySugger1965]

Be careful using too thin an inner wall. The regenerative fuel pressure is higher than the chamber pressure by two factors: the injector pressure drop at max flow, and the pressure drop within the cooling jacket which increases around the throat. But as CopSub found in a spectacular failure, there is scope for a way higher pressure differential during startup. As you transition from pre-stage to main-stage, you have main-stage furl pressure in the jacket, but only pre-stage pressure in the chamber. This can severely overpressure your jacket if you assume steady state conditions, leading to a costly failure.

[@AstroCharlie]

You're correct, but my pressure is relatively low pressure, even during the startup transient, compared to say commercial engines. The structural analysis will have to come later once I've closed the cooling story.

[@gregwarner3753]

That was not a failure. That engine test yielded significant information. That is success.

[@RyanDobler]

CRS-12 just refuses to stay up. 😆

[@jordanlowry590]

It would help if there were more pictures and examples pit on screen when you are talking would help for visual learners good videos though thank you!

[@Ameliaross106]

Manufacturing and plumbing is probably the biggest head aches in aerospace engineering.

[@vedantmomaya1533]

Hey can u make a video of all the sources that one can use to become someone like u from being a rookie like me,plssss

[@gregwarner3753]

Read, research, think, try, read more, think more, try again. Wor,ed for Dr. Goddard, shpuld work for you. I suggest starting with cold gas like compressed air. Set up a test stand, suggest suspended pendulum, take measurements and have fun.

[@vedantmomaya1533]

@@gregwarner3753 tyssssm for replying but should I start building physical projects or should I read or listen to lectures before going practical ? And can u recommend some material for a rookie

[@maxk4324]

Great video, except your explanation of isentropic flow was a little off. You said it was a flow wit constant enthalpy but thats called isenthalpic flow. Isentropic means constant entropy. I know you know this, but for anyone interested, in the text book "Design of Liquid Propulsion Rocket Engines" by Huzel and Huang, theres a really good and short description of what a nozzle basically does. Effectively it converts enthalpy into kinetic energy.

[@AstroCharlie]

It is still constant enthalpy though. Adiabatic and reversible are the necessary conditions for isentropic flow, which make it constant enthalpy, and enthalpy is the useful measure here since I'm not discussing the second law.

[@maxk4324]

​@@AstroCharlie I don't think that is correct, at least not when comparing the equations for nozzle flow and design that I am familiar with. As a disclaimer, I do not mean to question your experience, nor claim my experience is in anyway greater (I'm not sure where you are in your education but I am going into my third year of engineering), however the relevant equations as I have been taught them conflict with your description. So I guess I will try to simply state my understanding and let you judge for yourself or point out my errors. (Note: I recommend reading this on a computer rather than mobile as I have written out equations) First, I would like to point out that the wikipedia page for "de Laval nozzle", under the "Operation" tab does state that they are assumed to operate nearly isentropically. I know that wikipedia isn't a source that would hold up in, say, a masters thesis, but I think for casual debate its a fairly accurate source most of the time. However that is not the basis for my argument, just some back up. My first argument would be to look at the equation for energy rate balance for a one-inlet one-outlet control volume. For rates I will follow the variable with "_dot". Looking at the steady state energy rate balance for control volumes with one-inlet one-outlet the equation comes to: 0 = Q_dot - W_dot + m_dot * { (hi - he) + [(Vi^2 - Ve^2)/2] + g * (zi - ze)} We can get rid of Q_dot and W_dot since we are looking at the nozzle on its own, and we can also assume the height, z, change is negligible which brings us to: 0 = m_dot * { (hi - he) + [(Vi^2 - Ve^2)/2] } Which we can ditch m_dot and gives us: 0 = hi - he + [ (Vi^2 - Ve^2) / 2 ] Most commonly, as far as I have seen, it is assumed that Ve is so much larger than Vi that Vi could just as well be zero, but even if you incorporated Vi (assuming some magical way to measure velocity at the start of the converging region without melting through your instrumentation) the equation clearly shows that, assuming adiabatic (Q_dot = 0) and that the system is bounded to the nozzle (W_dot = 0) then a change in velocity has to be accompanied by a change in enthalpy. My second argument is similar but comes from a more intuitive approach to the issue. I think we can both agree that, as a gas moves through a properly designed converging-diverging nozzle, the pressure and temperature are dramatically decreased from the nozzle start to the nozzle exit. So one must ask oneself, will a gas that is very hot and high pressure have the same enthalpy as that same gas at a much lower temperature and pressure? It is for these two reasons (which are sort of the same reason approached from twp different perspectives), that I would assert that CD nozzles do not operate at constant enthalpy from inlet to exit. P.S. Regardless of the outcome of this debate, your statement at 4:21 about isentropic flow implying constant enthalpy is unarguably false. Did you mean to say isenthalpic? You can have isentropic processes that still incorporate enthalpy changes.

[@AstroCharlie]

@@maxk4324 Sorry, KZbin replies keep disappearing, and I get less inspired to re-type this response each time. Long story short: I use total enthalpy, which includes kinetic energy in its definition so constant enthalpy is true. Knowing that and knowing that the nozzle is isentropic (deltaS=0 so Q=0) is enough to claim constant enthalpy in this case, but not generally, as you point out.

[@maxk4324]

@@AstroCharlie oooh, ok, I see now, thank you for clarifying.

[@profishinjr4880]

Is 25 psi safe

[@jshatfield115]

Small mistake at 8:44 "Overexpand the flow to a pressure more than the ambient air"... I noticed that you like to correct your mistakes. ("Propellant" instead of "Fuel" in an earlier vid)

[@AstroCharlie]

While it's certainly not my best English, overexpand more = lower pressure so I believe it is correct.

[@jshatfield115]

@@AstroCharlie listen to your statement at 8:43. You say "if you arent careful, you can overexpand the flow to a pressure higher than the ambient air". Simple mistake but it may confuse a small number of viewers.

[@jshatfield115]

Also did you ever finish the engine?

[@jshatfield115]

@@AstroCharlie no offense but I am positive that you made a tiny mistake man. It wouldn't matter but I saw some confusion in the comments.

[@danielemarchelli]

HI Charlie, this video, like others, is very very good. exhaustive and in-depth explanations which as there are no in others tubes; but.... you talk too fast for my poor english! :(

[@rahadyanmuhammad271]

Hi everyone, I am sorry, I still don't get the point what really happens if incompressible fluid flow through the CD nozzle? why the density change can affect the acceleration of the compressible fluid thanks

[@AbgezocktXD]

Hopefully you do a better job welding up your rocket engine than @thunderfoot does when he builds a cesium still ;)

[@HuskyMachining]

I would love to machine some rocket parts for you! Ill only charge you the material cost

[@mrlazda]

I am not rocket engineer or aerospace engineer, but I would like someone to give me example of "well designed turbofan" with exhaust velocity of ~4m/s and it application, that must be really slow aircraft (that aircraft must fly below average running speed of human), because as I remember physics from middle school (and that was long time ago) (simple) moment conservation equation say that air breathing engines can't produce trust if they move faster then it exhausts velocity (to be totally exact they can for amount that is equal to ratio of mass of fuel to mass of air, which is typically maybe around 0.03 so they can move like 3% faster then exhaust velocity). Or to put it differently net thrust of air breathing engines is gross thrust (mass flow rate of exhaust gases times exhaust velocity) minus ram drag (mass flow rate of free stream times free stream velocity) so if velocity of free stream is higher then exhaust velocity then net thrust would be negative

[@AstroCharlie]

The exhaust velocity would be measured in the frame of reference of the aircraft, otherwise you would need to know what the speed of the aircraft is to calculate thrust. Since air enters the front of the jet engine with some velocity, and you decelerate it, it mostly keeps that energy in the form of a pressure rise, so even at cruising speed of ~250 m/s the exhaust velocity of a large turbofan is ~4m/s relative to the aircraft.

[@mrlazda]

@@AstroCharlie then you could say that exhaust velocity of rocket engine can be 0 even negative if we use rocket as frame of reference. By the way I never heard someone use aircraft as frame reference for exaust velocity, I just heard for actual exaust velocity and effective exaust velocity but both do not use as I know airplane as reference, but you are rocket engineer and I am just electronic engineer

[@AstroCharlie]

Since rockets add significant m_dot to their flow, unlike jet engines, the current velocity of a rocket does not affect thrust generation. Jet engines experience drag from their engine based on their velocity, so their ability to continue accelerating is a function of their relative exhaust velocity.

[@user-we9xi2ff5p]

привет друг , не смогу правельно выразится на англиском, можешь сделать обзор на компанию Roket Lab они делают не большие ракеты и хорошо помогут тебе в развитии ракетно житкосных двигателей (my friend vomits, I can’t correctly express myself in English, you can do a review on the Roket Lab company, they make not big rockets and will help you well in the development of rocket engines)

[@AstroCharlie]

Thank you for your comment! I know about RocketLabs and am very excited to follow the progress of their Rutherford engines!

[@hamadaag5659]

He kinda looks like black widow, right?

[@trcxsa]

What about asking other KZbinrs for help with machining? Say ones that do machining videos?

[@wetbredloaf]

did bro just say gooey inseatd of GUI

[@azerwhite8870]

“My Nozzle is going to melt, so the problem here is...” no you’re actual problem is the fact that your nozzle is going to melt. 😂

[@proto_hexagon5649]

some ilustration was better. Bell and conical nozzel. To me look the same. At least what my english undesrtud. Cherrs from Spain

[@gregwarner3753]

More graphs and pictures and less of you. You can talk over the information instead of distracting

[@Saleemsan]

Why in the name of Allah are you using imperial units? CGS/metric, paleeze