I worked building RL-10's back in '89-90 when they were developing the improved thrust version that has the moveable skirt and they really ran into problems. When you increase the fuel flow needed to increase the thrust the skirt cools down which then reduces the expansion which slows the pumps. It's a really delicate balancing act on the early engines and the first attempts to increase the thrust did achieve more but they lost impulse which is no good for a second stage engine. The nozzles supplying fuel and oxygen in the 'shower head' were flowed then carefully adjusted by hand to get exactly the flow needed which was a bit tedious and often took many trips to the flow test to get it right. Happily things were worked out and the engineers at P&W were and still are the finest in the world.
DC-X was a 'first stage' that was lifted against Earth Gravity, using 4 RL-10's! Another cool idea is to use an expander cycle with an Aerospike nozzle. This would supposedly give some more nozzle area so the total expander cycle engine could have more thrust than with a de laval nozzle (it would 'scale' better).
AGREED to the different mozzie geometry video idea! De Laval was an awesome hydrodynamicist, but his design is one of many! Heck, a video on just the differences between a plug and aerospike nozzle would doll your time.
I like the spike. One of the early complaints was that the external nozzle materials couldn't hold up against the heat well. That was before the new advanced materials research that gave us the heat shield tiles of the shuttle. There are a lot of great ideas that got put on a shelf that would do better a second time around.
@@Bartekkru100 in the Destin video, you could see him and his wife standing next to the Delta Heavy and talking deep rocket science. Can't wait for Vulcan to become a 21st century rocket.
I wish I understood 3% of what you're saying, but it's absolutely fascinating. Okay maybe I understand 3% but I love how much detail you get into. Thanks again, Scott.
Not going to lie, I had to back up and replay parts of this video a bunch of times - but it was really fascinating! Thanks for the introduction to Expander Cycle rocket engines!
Thank you Scott for talking about a great Engine like the RL-10, but Sorry Scott RL-10A5 used on the DCX and DCXA launched many times. I worked on the engine test program from 1991 to 1995 for P&W. You also didn't note that the RL-10A5 was very throttleable from down to 5% to 100% in less than 3 seconds, This is why it was selected for the CC program to land on the Moon. Most likely will be again just a modified Aces engine RL-10C-5. Best regards Ed.
Thank you for the engine, it made my late game Moon and Mars landings an efficient breeze in KSP with Realism Overhaul. :) The only other option was the Lunar Descent Engine and its questionable efficiency.
Did you work at the so Florida R&D center? I had family that worked there. On the drawing pressure was in psia and temp was in R, can you please explain these?
@@marksmovies6191 psia is pounds per square inch absolute. 0 psia would be a vacuum as apposed to psig where 0 psi would be atmospheric pressure. I don't know about R. :)
Cryogenic H2 & He ... In another life I use to service vacuum pumps and systems. The most unusual pump I had to service (for the first time) was a cyro-pump using liquid helium in a "cold head" that's purpose was the condense and freeze all the residual gas left in the vacuum chamber. The liquid helium was produced by the system and there was a "temperature gauge" to indicate when the helium was cold enough to liquefy. How this was done was with SS sealed system pipe system with pressure gauge and a small bulb in the cold head and was filled H2 at 7 bar. The pressure gauge scale wasn't in pressure, rather it read the lower end of the Kelvin scale. The idea was that as the temperature dropped to cryogenic levels so did the pressure. The gauge needle would sit there doing nothing and then would slowly drop to 14K at which point the H2 would freeze in the bulb. The pressure would still drop past this point as more H2 condensed and froze. From memory the low end of the scale was 4K which is the temp of LH2. The reason for the service was they opened the connection to the gauge and let all the H2 out so it didn't work as it should. All I had to do was evacuate the system, fill with H2, evacuate again to and then fill to 7bar. The client was very surprised that it was so simple to fix and they just paid a lot of money for me to do something they could have done. Its the old story - "you're not paying me for turning the screw; rather you're paying me for knowing which screw to turn." haha
I work with a machine that uses a cryo pump, it's a wonderful combination of a stupidly simple concept (let's make a vacuum by freezing all the air solid) that requires crazy hardware. It still blows my mind a little seeing that temperature gauge casually ticking over at 10-15 Kelvin.
I haven’t really payed attention to your subscriber count for a while and I still thought you had like 25000. Now you have 860,000!! That’s insane. Good job
I don't use it much I find, usually because I tend to get cost-obsessive in RO games, and RL-10s are great but cost an assload. They're damn fine for probe transfer stages though, as long as you insulate it well.
Thank you Scott, 99% of that just washed over me and did not stick, we really enjoyed it, and that 1% is a lot of new knowledge, you could easily win the Galactic Institute's Prize for Extreme Cleverness. It is amazing that the most tech heavy industry relies on kit and ideas that are decades old.
Hey Scott Manley. If it were not for you I would never know about staged combustion closed cycle rockets! Love your channel Scott. You are the best spokesman for space exploration I have know in my 69 years. Keep it up. I have a PhD in Chemistry and learn new stuff from you daily.
I could not teach you a thing about rockets but in our labs at UT Austin we did not work on anything that wasn't pyrophoric. Imagine metals at 3000 deg C with liquid nitrogen cooled flasks.
@@Azivegu I got you beat. I been here since he uploaded a legendary video of his daughter playing Eve Online (A game known for it's merciless learning curve) as a pirate and wasting people. Before KSP he played EVE.
Idunno what Bob Ross was doing in that web page but now all I can think about is adding a 'happy little rocket' to the background of a painting. ..Maybe some ejecta from stage-sep .. Little sun glistening off the main body .. and a slightly broken con-trail
That's the first time since Uni that I've come across Rankine being used, and then it was because I was too cheap to buy aerodynamics textbook and checked out the old imperial edition for non-stop 3 years (nobody else ever requested it).
4:10 that cut out gives you a view of how intricate and technologically advanced these engines are. And apart from spacex all these are thrown away after one use. 🚀🗑😱
@@motoboggin2619 The little impeller on the left is the LOX pump. You can see it's driven slower than the fuel impeller from the gearing. The two-stage LH2 pump (centre) spins at turbine RPM. The turbine is at the bottom. The turbine on RL10 spins at about 30k RPM.
Yeah i did understand (watching the video) that it was the pump section but still it is impressive and expensive and an important part of the engine that is discarded. 🧐🧐
Like Isaac Arthur says: The reason space is so expensive is we're throwing away our car after one trip and buying a new one, rather than just refilling the gas tank.
Love your videos as always! I would love to see a video about the combustion tap-off cycle. As a thought, I always wondered about a combustion tap-off cycle that uses an external starter such as an electric motor to get the turbo pump spinning before the rocket launches. I figured that starting such an engine would be extremely simple. Get the turbo pump spinning, ignite the pyros, then open the propellant valves and the engine will start and be able to run on its own. An overrunning clutch will automatically decouple it from the electric motor once the engine fully ignites, and off it goes!
@@General12th > It's possible we're stuck on Earth with chemical rockets forevermore. It may even be possible that we're eventually stuck on Earth but without rockets, for all resources required to make the fuel and special materials are exhausted.
@@vladimirdyuzhev That's unlikely. Rocket fuel can be made from water, or carbon dioxide, and the only time those are gone is if the Sun has consumed the Earth and we're all dead anyway.
Curious Marc sent me here but I'm already an avid subscriber. Somehow I missed this video in 2019. I now understand rocket science. There should be a pin or something.
I'm amazed that after so many years that rocket design hasn't settled on a few basic designs that everyone uses. It shows that rockets are still in their infancy and have a lot of improvements yet to go.
Technology advances, new manufacturing methods and materials make old, previously discarded ideas relevant again, changes in prices make other ideas irrelevant.
It is true that rockets haven't settled on a few basic designs, and there's probably a lot of improvement to go, but a major part of this is just that different rockets and designs are optimal for different things. Do you want a cheap engine? Do you want high ISP? Do you want high thrust? Do you want easy reusability? Do you want high reliability? Is this for a large rocket or a small rocket? These are many of the different things and depending on the desired goals, different cycles will be better.
The RL-10 is about as standard as it gets for rocket upper stages, and at this point it basically writes its own resume. What's there not to like? It's efficient, cheap, and simple.
Suppose we have an expander cycle engine and scale it up by a factor of two keeping the chamber pressure the same. Then the throath area becomes 4 times larger and so does chamber wall area. With the same chamber pressure the LOX/LH flow (kg/s) becomes also 4 times greater. But then we will have 4 times bigger area to warm the 4 times bigger amount of LH to turn the turbine and there does'nt seem to be any problem. So why don't we get bigger thrust by scaling? My own solution (not quite sure if its correct) is that we need to take into account the best optimazation of the engine: Burning 4 times more LOX/LH does not need 8 times bigger combustion chamber, which would be the volume after scaling everything by factor of 2. It only needs to be 4 times bigger, and in that case the chamber wall would not have 4 times bigger area, but only 4^(2/3) bigger. So not enough are to warm the LH for the turbine. The Vinci engine clearly has a longer chamber than the RL-10 to counter this effect, but increasing the camber size too much might ultimately mean a too big weight penalty. Comments please.
Hi Juha--What you are saying is incorrect. If you double the engine's linear dimensions (which is what I think you're saying), then the throat area becomes 4 times larger (as you said) and pumping the propellants requires 4 times the power. But your statement about the chamber wall area going up by four times is not correct. It only goes up by 2 times. The throat area goes up by the square of the linear dimensions, but the cooled chamber area only goes up linearly with the chamber dimensions. So as we make the engine larger at constant chamber pressure, the required pump power goes up twice as fast as the thermal input power. This is why expander cycle engines have an intrinsic thrust limit.
And the Japanese are going with the (Open) Expander Bleed Cycle for their rocket engine LE-9, which solves the limiting thrust problem of Closed Expander Cycle, so it can give high thrust while also reducing complexity and cost albeit at a less efficiency. But the superiority(by power) yet low cost n simplicity of it is unmatched by other expander engines I believe. (I had made this comment without watching the whole video completely so I thought he missed LE9 of Japan when mentioning others earlier in the video.)
Would it be possible to have an open/closed hybrid expander cycle engine? That way you could run the engine in the open cycle configuration when more thrust is needed, and in closed cycle config when the thrust is not required, so that you can take advantage of the higher efficiency.
Rik Schaaf was wondering the same thing. Would be beneficial for ULA because if you look at their launch profile, they do thrust off-optimal to compensate for it’s low thrust while they try to achieve orbit. Imagine your solution where they can transfer over once orbit is achieved and gravity losses are no longer a problem!
I like how the Rutherford engineer upper stage has two batteries one ejects when drained reducing the total weight to SECO down just a wee bit. Your going to need all the efficiency you can get with a 17m tall rocket.
Well these all seem a bit better then some of my concepts but I remain confident. Although my current prototype was quite promising. Due to refinements in the design and upgrading the rotational power distribution packages I have dramatically improved efficiency. This has allowed me to eliminate 3 of the 7 guinea pi........ engines. This in turn gives me better balance, allows a reduction in lettuce...... fuel and reduces the overall waste produced by the 4 remaining engines. I will continue to be inspired by my hero Ralph Kramden. I too just want to send my loved ones to the Moon.
Can you do a vijeo on modern solid rockets? All this edumaction on liquid rockets makes me realize that they must me more complex than I've thought about...
I thought I was pretty sharp with propulsion with a career in business and some military aviation. Space technology...how hard can it be.? Well you just crushed me with this one. Thank You. I needed a come-uppance. 😉
Your voice sounded great on the intro! New mic, or configuration? And do you think Kerbal 2 will be that much better? I mean the laws of physics remain the same, what could be taking them so long to establish?
In the final design you mention, why not dump the gas generator exhaust into the combustion chamber? Is that exhaust at a lower pressure? Or just fully combusted and not providing any benefit at that point?
I used to learn a pretty interesting expander cycle engine design. In this engine, the pump is driven by helium in a closed system. The helium is heated by combustion chamber and then dirven the pumps of liquid hydrogen and liquid oxigen; then the helium is cooled by the tank of liquid oxigen. It was proposed by PWR and called EX-Hex cycle engine. This design could also be applied on methane-oxigen engine.
I have an idea! We should pre-mix the oxidizer and fuel in one bigger fuel tank to reduce the complexity of the rocket designs! Less tanks, fewer moving parts, more fuel! What could possibly go wrong?
One question: why once heated up and expanded the hydrogen drives the turbine instead of just pushing back into the tank? And if it pushes against the tank back pressure wouldn’t that be the same as a compressed gas turbine?
You can use a gas generator turbine to provide power for the gas generator which can provide power for both turbines. This means you have to spin up your gas generator turbine using compressed gas (or electric engine or anything else) before turning on your gas generator. In any case, the gas generator turbine can be a lot smaller because it needs a lot less pressure than the one for the combustion chamber.
Man, I feel really pumped up right now, I'm going to buy some Estes rocket motors and build a multi-motor crazy albatross rocket and see what happens! 🚀💥🔥
Same as the RS 25s could, since they both use regenerative cooling around the engine bell. I believe the difference is just in that the Merlin dosn't use that heating to help pump the fuel since RP-1 doesn't boil quite as easily
@12:02 we see a Closed Dual Expander Cycle (w/ split oxidizer side), and later is shown a Closed Dual Split Expander Cycle. Is there a simple explanation for why, if only one side were to be split, it would be the oxidizer side?
What I would love to see would be a paraffin hybrid rocket driven by an expander cycle, since you only have to pump the oxidizer, you could get higher mass flow rates, and if you used the turbine exhaust for film cooling of the nozzle extension you could pull more energy out of it (i.e. have a lower turbine exhaust pressure). Also, since the combustion chamber wall is insulated by the fuel, you would only have to actively cool the nozzle which could allow for larger engines that are still on a closed, or partially closed cycle.
Scott-- Very good presentation. Just 1 minor editorial comment. The LH2 fuel is never 'boiled'. The LH2 is pumped to a pressure above what is called its 'critical pressure'. Above the 'critical pressure', boiling is not possible. There is no 'discontinuous phase change', only a 'smooth' drop in density at temperatures above the 'Critical Temperature'. Rocket engine regenerative combustion NEVER operate under conditions under which the fuel can boil in the regen system. Reason--the heat transfer characteristics under boiling conditions are not stable, and quickly decay to levels that cannot support operation of the hardware. For this reason, regenerative coolant systems (at least the ones that I have experience with) all use coolant that is at supercritical pressure. Thank you very much.
Good point, should have been more clear with wording. After the regen system and before the turbines it does make the phase change to gas though right?
@@scottmanley No. The critical pressure of Hydrogen is 186 psia. The original chamber pressure of the RL-10 was about 350 psia (its higher than that now), so the 'back pressure' that the H2 system is 'working against' is essentially twice critical pressure. As long as the H2 stays above the Critical Pressure, the words 'phase change' have no meaning. It is certainly warm enough to be called a 'gas', and nobody would argue with you, but it didn't get there as a result of 'boiling', or as a result of any 'phase change'. I hope I don't sound like I'm being 'too critical'--I only commented here because the notion of 'boiling' in the regen system is not what happens at all. Thank you. I enjoy watching your stuff. We should talk sometime.
@@scottmanley Not to 'dwell' on this too much, but if you want to present that kind of thing, a thermodynamic chart of H2 properties makes it easier to visualize. Do you have access to such? If not, I could probably help.
@@dandeprop Oh I learned all this in thermodynamics back in university (taught in the same lecture theater used by Kelvin), I'm already with you on this, it's the difference between knowing the theory and recognizing where it applies in practice.
why not feed some of the pressurized fuel coming from the cooling vanes directly into the combustion chamber, use the rest to power the turbopump and then feed the "exhaust" of the turbine into the fuel fuel tank? You could do that for the oxidizer tank too to pressurize both propellant tanks. Couldnt that improve the chamber pressure by a lot since its not limited to the turbopump pressure delta anymore?
My question is how much power does a turbo pump need to pump the fuel and oxidizer in the RL 10 engine. The V2 turbo pump generated something like 600 Hp if I am not mistaken.
Another way is to take some of the combustion chamber gas and send it off to a hydrogen heat exchanger instead of using a separate pre-burner to augment extra heat. The now cooler flue gas can be sent to the interior of the rocket bell to recover a little more "push" that would normally be lost if it was just bled out the side.
The Rutherford engine has half of a really good idea. It uses battery powered electric motors to drive the pumps. The better idea is a hybrid expander rocket engine. Take an RL-10, add an electric motor/generator on the hydrogen turbopump shaft. Remove the gearbox and drive the LOX pump with an electric motor. You get the advantages of the electrically driven rocket: no seal between LOX and fuel requiring helium, vastly simpler startup, simpler and much faster throttle control (good for landing!), less careful balancing of the turbine power to pump demand. But you can use a much smaller battery, since the majority of the pump power comes from the turbine. Finally, you can scale to a significantly larger thrust than an expander cycle, since when you increase engine thrust, expander turbine power may not keep pace but it does increase significantly. So as the engine gets bigger the battery scales faster than the rest of the engine. Eventually that's a problem but you can get quite a bit bigger and enjoy the cost and thrust:weight advantages of a larger engine before the battery weight is a loser.
The turbopump of the RL10 is in the order of a megawatt... So going electric definitely won't improve thrust:weight ratio and you can't get more thrust because you can't build such a powerfull electric motor/generator or it will be insanely heavy. Or you would need to lower chamber pressure (less pumping power required) but that would either hurt specific impulse or you would need a much larger nozzle (making the engine heavier again hurting t/w).
Can the propellent in a closed split expander cycle that powers the turbine then be sent to the propellant tank for autogenous pressurization, instead of to the combustion chamber? That would probably reduce the back-pressure on the turbine....
Hey, a couple of us were wondering about a hybrid between the expander cycle and the expander bleed cycle where you can switch back and forth to suit your needs. Would that work? Idk with the turbo pumps probably being so different for each. Also, could you drive only one pump with the expansion cycle and the other with another cycle to maximize efficiency and thrust?
I don't see why not. Probably the best combination (IMHO) would be to use the energy collected by heat transfer from the walls of the combustion chamber and nozzle to drive only the liquid hydrogen pump, with all of the hydrogen flowing through this expander cycle section. Then have a secondary cycle generating just enough power to drive the liquid oxygen pump, since with any hydrolox engine the volume of liquid oxygen needed is less than a quarter of the volume of liquid hydrogen used. This is partially because liquid hydrogen has such a low density - pieces of styrofoam actually sink in liquid hydrogen. But also because hydrolox engines are always run very fuel-rich, primarily to get as much unburned hydrogen into the exhaust as possible to minimise the average molecular weight of the exhaust, which maximises the average exhaust velocity and hence increases specific impulse. Another reason for running fuel-rich is to keep the temperature down - if a hydrolox engine was run at stoichiometric ratio to burn 100% of the hydrogen used, the exhaust temperature would be several thousand degrees higher. Consider than these engines already run at more than 3000 degrees C, so even running all the propellants through the walls of the combustion chamber and nozzle would not be enough cooling to stop everything melting unless you made the hot bits out of tungsten, but then it would be too heavy. Anyway, the secondary cycle could either be a conventional pre-burner or a combustion tap-off, again running fuel-rich, though it probably wouldn't make much difference which of those two approaches you used. You could even run this through another liquid-hydrogen cooled heat-exchanger first to bring the temperature down to a level the turbine could more easily cope with. This would also give the liquid hydrogen pump more energy, which might allow you to scale the whole engine up a bit, though I don't know by how much.
Scott, you're a computer programmer. I know that takes smarts, but how in the world did you teach yourself the complexities of all this rocket science? I'm amazed at your level of knowledge! Very impressive!
Apparently the BE7 engine is a dual expander cycle. I've no idea if it's bleed or closed (guessing bleed), but that cycle seems to have left the drawing board now.
Hey Scott, in all this time almost no-one talks about catalyst converters in mono-propellant engines, such as the catalyst pellets depicted in the diagram of Redstone's steam turbine. Think you could an episode on that? Like, exactly is it? What is it made of? What is the chemical reaction taking place? Is it the same thing that makes the bubbles in my contact lens solution, etc. That would be cool.
So the dual expander cycle is kinda like the full-flow version of expander cycle engines. I do wonder, though. Do rocket engines overall gain any significant performance when it preheats its fuel and/or oxidizer as a coolant before burning it?
A thing that would solve many of these problems is some sort of bearingless pump. Now there is no leak, so it doesn't matter what drives what. It'd be really cool if something like that existed for the pressures required...
I have a suggestion.. If you make the video as a documentary like A Discovery channel show instead of just narration, it will be very much good and interesting..
I feel like you made this video to answer my questions. Thank you very much, it's great.
Hey Dustin! Nice to see you watching Scot's videos!
hello destin!
I worked building RL-10's back in '89-90 when they were developing the improved thrust version that has the moveable skirt and they really ran into problems. When you increase the fuel flow needed to increase the thrust the skirt cools down which then reduces the expansion which slows the pumps. It's a really delicate balancing act on the early engines and the first attempts to increase the thrust did achieve more but they lost impulse which is no good for a second stage engine. The nozzles supplying fuel and oxygen in the 'shower head' were flowed then carefully adjusted by hand to get exactly the flow needed which was a bit tedious and often took many trips to the flow test to get it right. Happily things were worked out and the engineers at P&W were and still are the finest in the world.
Thanks for the bit of history! It's always interesting to get a measure of historical insight from the people who were there.
Sounds like a great example of why heat exchanger design is the bane of most mechanical engineering students!
Would this be such simpler nowadays with simulation? I've heard it's still pretty tough to get usable results sometimes
Love having professionals chime in!
Was the moveable skirt design actually patented by PWR ?
Curious as to whether the Vinci engine has to licence the technology...
DC-X was a 'first stage' that was lifted against Earth Gravity, using 4 RL-10's!
Another cool idea is to use an expander cycle with an Aerospike nozzle. This would supposedly give some more nozzle area so the total expander cycle engine could have more thrust than with a de laval nozzle (it would 'scale' better).
I forgot about that one.
@@scottmanley Can you please do some more on the different engines ? Including this aerospike?
This topic looks like an great pool of content!
AGREED to the different mozzie geometry video idea! De Laval was an awesome hydrodynamicist, but his design is one of many!
Heck, a video on just the differences between a plug and aerospike nozzle would doll your time.
I like the spike. One of the early complaints was that the external nozzle materials couldn't hold up against the heat well. That was before the new advanced materials research that gave us the heat shield tiles of the shuttle. There are a lot of great ideas that got put on a shelf that would do better a second time around.
I love how Elon, Tim, Tory, Destin and Scott just have a casual discussion in rocket engines on in some tweet reply chain of something unrelated.
Tory is a really underrated CEO. He replied to me three times on Reddit. He seems to be genuinely nice person.
@@Bartekkru100 in the Destin video, you could see him and his wife standing next to the Delta Heavy and talking deep rocket science. Can't wait for Vulcan to become a 21st century rocket.
I wish I understood 3% of what you're saying, but it's absolutely fascinating. Okay maybe I understand 3% but I love how much detail you get into. Thanks again, Scott.
This was great, please do more of these more in-depth videos on other cycles, other engines, and other space tech stuff.
Hell yeah! More technical as well!
Not going to lie, I had to back up and replay parts of this video a bunch of times - but it was really fascinating! Thanks for the introduction to Expander Cycle rocket engines!
Thank you Scott for talking about a great Engine like the RL-10, but Sorry Scott RL-10A5 used on the DCX and DCXA launched many times. I worked on the engine test program from 1991 to 1995 for P&W. You also didn't note that the RL-10A5 was very throttleable from down to 5% to 100% in less than 3 seconds, This is why it was selected for the CC program to land on the Moon. Most likely will be again just a modified Aces engine RL-10C-5. Best regards Ed.
Thank you for the engine, it made my late game Moon and Mars landings an efficient breeze in KSP with Realism Overhaul. :) The only other option was the Lunar Descent Engine and its questionable efficiency.
Did you work at the so Florida R&D center? I had family that worked there.
On the drawing pressure was in psia and temp was in R, can you please explain these?
@@marksmovies6191 psia is pounds per square inch absolute. 0 psia would be a vacuum as apposed to psig where 0 psi would be atmospheric pressure. I don't know about R. :)
Ed, did you know Bill Adair, who also worked on rockets and jet engines for P&W?
I love expander-cycle engines. It's just such an elegant concept.
Cryogenic H2 & He ... In another life I use to service vacuum pumps and systems. The most unusual pump I had to service (for the first time) was a cyro-pump using liquid helium in a "cold head" that's purpose was the condense and freeze all the residual gas left in the vacuum chamber. The liquid helium was produced by the system and there was a "temperature gauge" to indicate when the helium was cold enough to liquefy. How this was done was with SS sealed system pipe system with pressure gauge and a small bulb in the cold head and was filled H2 at 7 bar.
The pressure gauge scale wasn't in pressure, rather it read the lower end of the Kelvin scale. The idea was that as the temperature dropped to cryogenic levels so did the pressure. The gauge needle would sit there doing nothing and then would slowly drop to 14K at which point the H2 would freeze in the bulb. The pressure would still drop past this point as more H2 condensed and froze. From memory the low end of the scale was 4K which is the temp of LH2.
The reason for the service was they opened the connection to the gauge and let all the H2 out so it didn't work as it should. All I had to do was evacuate the system, fill with H2, evacuate again to and then fill to 7bar. The client was very surprised that it was so simple to fix and they just paid a lot of money for me to do something they could have done.
Its the old story - "you're not paying me for turning the screw; rather you're paying me for knowing which screw to turn." haha
I work with a machine that uses a cryo pump, it's a wonderful combination of a stupidly simple concept (let's make a vacuum by freezing all the air solid) that requires crazy hardware. It still blows my mind a little seeing that temperature gauge casually ticking over at 10-15 Kelvin.
Nice one Scott !!!! super informative !! extremely high KTT ratio (Knowledge To Time). loved it ...
I haven’t really payed attention to your subscriber count for a while and I still thought you had like 25000. Now you have 860,000!! That’s insane. Good job
5:38 when you see this after Simple RL-10 schemat
Only 90s kids will understand
If you look closely, you can faintly see the outline of what looks like a nozzle.
Isnt that the Mueller Report?
I wasn't ready for this yet
Yeah, uh, how is anyone supposed to _read_ this?
The RL-10 is my Realism Overhaul vacuum engine of choice!
I always find myself using the “Russian version” due to it being lower in the tech tree
Yah I use it and it's Russian equivalent a lot in RSS.
I don't use it much I find, usually because I tend to get cost-obsessive in RO games, and RL-10s are great but cost an assload. They're damn fine for probe transfer stages though, as long as you insulate it well.
@@Patchuchan RL-10 for the win
Thank you Scott, 99% of that just washed over me and did not stick, we really enjoyed it, and that 1% is a lot of new knowledge, you could easily win the Galactic Institute's Prize for Extreme Cleverness. It is amazing that the most tech heavy industry relies on kit and ideas that are decades old.
Those were some cool (or maybe hot...) new expander cycles I had not seen before. Thanks for some rocket design with my morning coffee!
Hey Scott Manley. If it were not for you I would never know about staged combustion closed cycle rockets! Love your channel Scott. You are the best spokesman for space exploration I have know in my 69 years. Keep it up. I have a PhD in Chemistry and learn new stuff from you daily.
I’m sure you could teach me a lot of chemistry.
I could not teach you a thing about rockets but in our labs at UT Austin we did not work on anything that wasn't pyrophoric. Imagine metals at 3000 deg C with liquid nitrogen cooled flasks.
One of the best channels on UA-cam. Keep up the good work.
Just w8 for the day this chanel hits long deserved 1 milion subs
jeez I have been here for a long time. Still remember the heated debates of how to pronounce Mün.
It's close...
It also says something about people’s interest in space!
@@Azivegu :o)
@@Azivegu I got you beat. I been here since he uploaded a legendary video of his daughter playing Eve Online (A game known for it's merciless learning curve) as a pirate and wasting people. Before KSP he played EVE.
It is truly amazing how something so impossibly complicated is made to seem trivial.
That blog post, and the follow up NTR post, are PURE GOLD Internet manna!
Idunno what Bob Ross was doing in that web page but now all I can think about is adding a 'happy little rocket' to the background of a painting.
..Maybe some ejecta from stage-sep
.. Little sun glistening off the main body
.. and a slightly broken con-trail
best I can do is that he was Air Force....didnt work on rockets though
there are no explosions, just happy little RUDs
Flat
It's a play on the title "art of the expander cycle engines", there is a link in the description :)
Tap-tap-tap.
6:00 Scott is surprisingly confident in our knowledge of the rankine scale.
That's the first time since Uni that I've come across Rankine being used, and then it was because I was too cheap to buy aerodynamics textbook and checked out the old imperial edition for non-stop 3 years (nobody else ever requested it).
6:18 You forgot about the RL10A-5 wich flew on the DC-X wich was used in atmospheric conditions
Thank you for explaining that. I did know that limitations in the Closed system were due to cube-square law, but it is nice to refresh memory.
Thanks for expanding my knowledge on this subject!
Really looking forward to Japan’s LE-9.
4:10 that cut out gives you a view of how intricate and technologically advanced these engines are. And apart from spacex all these are thrown away after one use. 🚀🗑😱
thats an oxidizer turbo pump
@@motoboggin2619
The little impeller on the left is the LOX pump. You can see it's driven slower than the fuel impeller from the gearing. The two-stage LH2 pump (centre) spins at turbine RPM.
The turbine is at the bottom. The turbine on RL10 spins at about 30k RPM.
Yeah i did understand (watching the video) that it was the pump section but still it is impressive and expensive and an important part of the engine that is discarded. 🧐🧐
Like Isaac Arthur says: The reason space is so expensive is we're throwing away our car after one trip and buying a new one, rather than just refilling the gas tank.
That's actually really simple, compared to just an engine, let alone a rocket.
Love your videos as always! I would love to see a video about the combustion tap-off cycle. As a thought, I always wondered about a combustion tap-off cycle that uses an external starter such as an electric motor to get the turbo pump spinning before the rocket launches. I figured that starting such an engine would be extremely simple. Get the turbo pump spinning, ignite the pyros, then open the propellant valves and the engine will start and be able to run on its own. An overrunning clutch will automatically decouple it from the electric motor once the engine fully ignites, and off it goes!
I hope the day will come when we look upon this technology as being antiquated and quaint kind of like steam locomotives.
Norman Mattson
Allow me to add on to that
I hope most of us live to see that day
That day may never come. It's possible we're stuck on Earth with chemical rockets forevermore.
That day came and went when they got NERVA to work and then canceled it.
@@General12th > It's possible we're stuck on Earth with chemical rockets forevermore.
It may even be possible that we're eventually stuck on Earth but without rockets, for all resources required to make the fuel and special materials are exhausted.
@@vladimirdyuzhev That's unlikely. Rocket fuel can be made from water, or carbon dioxide, and the only time those are gone is if the Sun has consumed the Earth and we're all dead anyway.
Curious Marc sent me here but I'm already an avid subscriber. Somehow I missed this video in 2019. I now understand rocket science. There should be a pin or something.
The best space channel on YT
I am more than amazed when I saw the Chinese subtitles. AWESOME
I'm amazed that after so many years that rocket design hasn't settled on a few basic designs that everyone uses. It shows that rockets are still in their infancy and have a lot of improvements yet to go.
It can also mean, that people want more tools for the growing number of jobs.
Technology advances, new manufacturing methods and materials make old, previously discarded ideas relevant again, changes in prices make other ideas irrelevant.
It is true that rockets haven't settled on a few basic designs, and there's probably a lot of improvement to go, but a major part of this is just that different rockets and designs are optimal for different things. Do you want a cheap engine? Do you want high ISP? Do you want high thrust? Do you want easy reusability? Do you want high reliability? Is this for a large rocket or a small rocket? These are many of the different things and depending on the desired goals, different cycles will be better.
The RL-10 is about as standard as it gets for rocket upper stages, and at this point it basically writes its own resume. What's there not to like? It's efficient, cheap, and simple.
By that reasoning the ICE engine is in it's infancy instead of coming to the end of it's life.
I love how you set up your Delta 4 and Atlas 5, while talking about their glorious upper stage and the engines that it uses.🚀👌🚀👌🚀
tl;dr: Expander Cycles are the rocket engine equivalent of Pulling Yourself Up By Your Bootstraps
:D
Suppose we have an expander cycle engine and scale it up by a factor of two keeping the chamber pressure the same. Then the throath area becomes 4 times larger and so does chamber wall area.
With the same chamber pressure the LOX/LH flow (kg/s) becomes also 4 times greater. But then we will have 4 times bigger area to warm the 4 times bigger amount of LH to turn the turbine and there does'nt seem to be any problem.
So why don't we get bigger thrust by scaling?
My own solution (not quite sure if its correct) is that we need to take into account the best optimazation of the engine:
Burning 4 times more LOX/LH does not need 8 times bigger combustion chamber, which would be the volume after scaling everything by factor of 2. It only needs to be 4 times bigger, and in that case the chamber wall would not have 4 times bigger area, but only 4^(2/3) bigger. So not enough are to warm the LH for the turbine.
The Vinci engine clearly has a longer chamber than the RL-10 to counter this effect, but increasing the camber size too much might ultimately mean a too big weight penalty.
Comments please.
Hi Juha--What you are saying is incorrect. If you double the engine's linear dimensions (which is what I think you're saying), then the throat area becomes 4 times larger (as you said) and pumping the propellants requires 4 times the power. But your statement about the chamber wall area going up by four times is not correct. It only goes up by 2 times. The throat area goes up by the square of the linear dimensions, but the cooled chamber area only goes up linearly with the chamber dimensions. So as we make the engine larger at constant chamber pressure, the required pump power goes up twice as fast as the thermal input power. This is why expander cycle engines have an intrinsic thrust limit.
And the Japanese are going with the (Open) Expander Bleed Cycle for their rocket engine LE-9, which solves the limiting thrust problem of Closed Expander Cycle, so it can give high thrust while also reducing complexity and cost albeit at a less efficiency.
But the superiority(by power) yet low cost n simplicity of it is unmatched by other expander engines I believe.
(I had made this comment without watching the whole video completely so I thought he missed LE9 of Japan when mentioning others earlier in the video.)
I love learning from you, Scott. Your voice is like my light in the dark.
I appreciate the new Starship models behind you )
Yeah I really liked those as well.
I now have a 6" Newtonian on an Equatorial with a clock drive. I'm super exited! :D
Would it be possible to have an open/closed hybrid expander cycle engine? That way you could run the engine in the open cycle configuration when more thrust is needed, and in closed cycle config when the thrust is not required, so that you can take advantage of the higher efficiency.
Rik Schaaf was wondering the same thing. Would be beneficial for ULA because if you look at their launch profile, they do thrust off-optimal to compensate for it’s low thrust while they try to achieve orbit. Imagine your solution where they can transfer over once orbit is achieved and gravity losses are no longer a problem!
in my office at work i have that F1 @0:52 engine turbopump assembly cutaway hanging on my wall 36x24.
great conversation starter
I like how the Rutherford engineer upper stage has two batteries one ejects when drained reducing the total weight to SECO down just a wee bit.
Your going to need all the efficiency you can get with a 17m tall rocket.
Here from Curious Marc channel. Great explanation!
The rl 10 is a workhorse of an engine logging so many flight hours awesome look into how it works
IIRC the RL-10 was used on DC-X and was also planned for DC-Y. So it WAS used as a liftoff engine.
Scott: ...neutral hydrogen gas...
Hydrogen embrittlement: what am I, a joke to you?
You are nothing compared to oxygen free radicals.
Hey Scotty! thanks for beaming me up again!!
I LOVE THAT SHIRT! EXCELLENT CHOICE YOU HAVE GREAT TASTE!
Well these all seem a bit better then some of my concepts but I remain confident.
Although my current prototype was quite promising. Due to refinements in the design and upgrading the rotational power distribution packages I have dramatically improved efficiency. This has allowed me to eliminate 3 of the 7 guinea pi........ engines. This in turn gives me better balance, allows a reduction in lettuce...... fuel and reduces the overall waste produced by the 4 remaining engines. I will continue to be inspired by my hero Ralph Kramden. I too just want to send my loved ones to the Moon.
Can you do a vijeo on modern solid rockets? All this edumaction on liquid rockets makes me realize that they must me more complex than I've thought about...
Is the RL-10 the engine that whistles at 80,000hz? And damaged a payload back around 2009?
What perfect timing. I was just thinking about that...
Also FIRST!!!
I thought I was pretty sharp with propulsion with a career in business and some military aviation. Space technology...how hard can it be.? Well you just crushed me with this one. Thank You. I needed a come-uppance. 😉
There's beauty in every cycle type..
Your voice sounded great on the intro! New mic, or configuration? And do you think Kerbal 2 will be that much better? I mean the laws of physics remain the same, what could be taking them so long to establish?
In the final design you mention, why not dump the gas generator exhaust into the combustion chamber? Is that exhaust at a lower pressure? Or just fully combusted and not providing any benefit at that point?
How do they keep the boiling liquid from creating massive back pressure, or simply locking a check valve closed?
I used to learn a pretty interesting expander cycle engine design. In this engine, the pump is driven by helium in a closed system. The helium is heated by combustion chamber and then dirven the pumps of liquid hydrogen and liquid oxigen; then the helium is cooled by the tank of liquid oxigen. It was proposed by PWR and called EX-Hex cycle engine. This design could also be applied on methane-oxigen engine.
Now that answered my question of of tubes wrapped around the engine bell of many rocket engines. Thanks scott manley
Ah, the smile in my face when I saw 14m43s of video!
I have an idea! We should pre-mix the oxidizer and fuel in one bigger fuel tank to reduce the complexity of the rocket designs! Less tanks, fewer moving parts, more fuel!
What could possibly go wrong?
Hey we should do that with hypergolics
Ka...BOOM!
@@awuma Wait, is producing a constant Ka...BOOOOOOM at the bottom not the goal?
I know your joking, but they also have different storage temps to lower bulk
What a concentration of knowledge here, both by our safe flyer and commenters.
One question: why once heated up and expanded the hydrogen drives the turbine instead of just pushing back into the tank? And if it pushes against the tank back pressure wouldn’t that be the same as a compressed gas turbine?
You can use a gas generator turbine to provide power for the gas generator which can provide power for both turbines. This means you have to spin up your gas generator turbine using compressed gas (or electric engine or anything else) before turning on your gas generator. In any case, the gas generator turbine can be a lot smaller because it needs a lot less pressure than the one for the combustion chamber.
Man, I feel really pumped up right now, I'm going to buy some Estes rocket motors and build a multi-motor crazy albatross rocket and see what happens! 🚀💥🔥
your "fly safe" at the end gave me threat vibes 0-0
This made me wonder how tf merlins survive all that and actually can be reused...
Or rather - how confident can we be in the reuse. How much the probability of failure increases with each firing.
By not being damaged by normal operation beyond the same kind of wear as you expect from operating your car's engine.
Same as the RS 25s could, since they both use regenerative cooling around the engine bell. I believe the difference is just in that the Merlin dosn't use that heating to help pump the fuel since RP-1 doesn't boil quite as easily
I always feel smarter after I watch one of your videos. :)
Actually the longer I watched this video the dumber I felt. This is so far over my head its not even funny.
Just watch it again....
@@scottmanley Your too kind.
@12:02 we see a Closed Dual Expander Cycle (w/ split oxidizer side), and later is shown a Closed Dual Split Expander Cycle. Is there a simple explanation for why, if only one side were to be split, it would be the oxidizer side?
Don't forget about the Relativity Space Aeon, it's going to be a fully 3D printed open expander cycle engine.
What I would love to see would be a paraffin hybrid rocket driven by an expander cycle, since you only have to pump the oxidizer, you could get higher mass flow rates, and if you used the turbine exhaust for film cooling of the nozzle extension you could pull more energy out of it (i.e. have a lower turbine exhaust pressure). Also, since the combustion chamber wall is insulated by the fuel, you would only have to actively cool the nozzle which could allow for larger engines that are still on a closed, or partially closed cycle.
Scott-- Very good presentation. Just 1 minor editorial comment. The LH2 fuel is never 'boiled'. The LH2 is pumped to a pressure above what is called its 'critical pressure'. Above the 'critical pressure', boiling is not possible. There is no 'discontinuous phase change', only a 'smooth' drop in density at temperatures above the 'Critical Temperature'. Rocket engine regenerative combustion NEVER operate under conditions under which the fuel can boil in the regen system. Reason--the heat transfer characteristics under boiling conditions are not stable, and quickly decay to levels that cannot support operation of the hardware. For this reason, regenerative coolant systems (at least the ones that I have experience with) all use coolant that is at supercritical pressure. Thank you very much.
Good point, should have been more clear with wording. After the regen system and before the turbines it does make the phase change to gas though right?
@@scottmanley No. The critical pressure of Hydrogen is 186 psia. The original chamber pressure of the RL-10 was about 350 psia (its higher than that now), so the 'back pressure' that the H2 system is 'working against' is essentially twice critical pressure. As long as the H2 stays above the Critical Pressure, the words 'phase change' have no meaning. It is certainly warm enough to be called a 'gas', and nobody would argue with you, but it didn't get there as a result of 'boiling', or as a result of any 'phase change'. I hope I don't sound like I'm being 'too critical'--I only commented here because the notion of 'boiling' in the regen system is not what happens at all. Thank you. I enjoy watching your stuff. We should talk sometime.
Thanks, I wish I’d thought it through a bit more, this would have been a really interesting distinction to spend a bit of time on.
@@scottmanley Not to 'dwell' on this too much, but if you want to present that kind of thing, a thermodynamic chart of H2 properties makes it easier to visualize. Do you have access to such? If not, I could probably help.
@@dandeprop Oh I learned all this in thermodynamics back in university (taught in the same lecture theater used by Kelvin), I'm already with you on this, it's the difference between knowing the theory and recognizing where it applies in practice.
why not feed some of the pressurized fuel coming from the cooling vanes directly into the combustion chamber, use the rest to power the turbopump and then feed the "exhaust" of the turbine into the fuel fuel tank? You could do that for the oxidizer tank too to pressurize both propellant tanks.
Couldnt that improve the chamber pressure by a lot since its not limited to the turbopump pressure delta anymore?
6:00 It's good to see the Rankine Scale used every once in a while...
My question is how much power does a turbo pump need to pump the fuel and oxidizer in the RL 10 engine.
The V2 turbo pump generated something like 600 Hp if I am not mistaken.
The more you learn about rocket science, the more complicated it gets!
How much of a challenge is it to keep the boiling hydrogen from rupturing the pipes in the nozzle?
consider using the expander cycle for a fuel-rich gas generator, which then drives the turbopumps to increase chamber pressure
I love this kind of video’s, educational and entertaining
I love you Scott, your the best space youtuber, I wish I could travel to the US to meet you
trm me too
The expander cycle was used as a first stage on the SSTO Mcdonnell Douglas DC-X
Another way is to take some of the combustion chamber gas and send it off to a hydrogen heat exchanger instead of using a separate pre-burner to augment extra heat. The now cooler flue gas can be sent to the interior of the rocket bell to recover a little more "push" that would normally be lost if it was just bled out the side.
Could you make an RD-170 style expander cycle, to recover proportionately more heat energy from the exhaust of a larger engine?
The Rutherford engine has half of a really good idea. It uses battery powered electric motors to drive the pumps.
The better idea is a hybrid expander rocket engine. Take an RL-10, add an electric motor/generator on the hydrogen turbopump shaft. Remove the gearbox and drive the LOX pump with an electric motor. You get the advantages of the electrically driven rocket: no seal between LOX and fuel requiring helium, vastly simpler startup, simpler and much faster throttle control (good for landing!), less careful balancing of the turbine power to pump demand. But you can use a much smaller battery, since the majority of the pump power comes from the turbine. Finally, you can scale to a significantly larger thrust than an expander cycle, since when you increase engine thrust, expander turbine power may not keep pace but it does increase significantly. So as the engine gets bigger the battery scales faster than the rest of the engine. Eventually that's a problem but you can get quite a bit bigger and enjoy the cost and thrust:weight advantages of a larger engine before the battery weight is a loser.
The turbopump of the RL10 is in the order of a megawatt... So going electric definitely won't improve thrust:weight ratio and you can't get more thrust because you can't build such a powerfull electric motor/generator or it will be insanely heavy. Or you would need to lower chamber pressure (less pumping power required) but that would either hurt specific impulse or you would need a much larger nozzle (making the engine heavier again hurting t/w).
Scott, if you ever make it out to Utah (my home state) you need to go out to ATK and experience the booster rocket tests!
Hast thou forgotten the DC/X (and DC/XA)??
Can the propellent in a closed split expander cycle that powers the turbine then be sent to the propellant tank for autogenous pressurization, instead of to the combustion chamber? That would probably reduce the back-pressure on the turbine....
Hey, a couple of us were wondering about a hybrid between the expander cycle and the expander bleed cycle where you can switch back and forth to suit your needs. Would that work? Idk with the turbo pumps probably being so different for each.
Also, could you drive only one pump with the expansion cycle and the other with another cycle to maximize efficiency and thrust?
I don't see why not. Probably the best combination (IMHO) would be to use the energy collected by heat transfer from the walls of the combustion chamber and nozzle to drive only the liquid hydrogen pump, with all of the hydrogen flowing through this expander cycle section. Then have a secondary cycle generating just enough power to drive the liquid oxygen pump, since with any hydrolox engine the volume of liquid oxygen needed is less than a quarter of the volume of liquid hydrogen used.
This is partially because liquid hydrogen has such a low density - pieces of styrofoam actually sink in liquid hydrogen. But also because hydrolox engines are always run very fuel-rich, primarily to get as much unburned hydrogen into the exhaust as possible to minimise the average molecular weight of the exhaust, which maximises the average exhaust velocity and hence increases specific impulse.
Another reason for running fuel-rich is to keep the temperature down - if a hydrolox engine was run at stoichiometric ratio to burn 100% of the hydrogen used, the exhaust temperature would be several thousand degrees higher. Consider than these engines already run at more than 3000 degrees C, so even running all the propellants through the walls of the combustion chamber and nozzle would not be enough cooling to stop everything melting unless you made the hot bits out of tungsten, but then it would be too heavy.
Anyway, the secondary cycle could either be a conventional pre-burner or a combustion tap-off, again running fuel-rich, though it probably wouldn't make much difference which of those two approaches you used. You could even run this through another liquid-hydrogen cooled heat-exchanger first to bring the temperature down to a level the turbine could more easily cope with. This would also give the liquid hydrogen pump more energy, which might allow you to scale the whole engine up a bit, though I don't know by how much.
Scott, you're a computer programmer. I know that takes smarts, but how in the world did you teach yourself the complexities of all this rocket science? I'm amazed at your level of knowledge! Very impressive!
That T-shirt is so cool. I wish I could buy one.
6:05 Pretty sure this is the first time I've seen Rankine used literally anywhere outside the Wikipedia page.
😂😂😂
5:32 - _Powered_ the space shuttle, unless they pulled them back out of the museums while I wasn't looking.
Apparently the BE7 engine is a dual expander cycle. I've no idea if it's bleed or closed (guessing bleed), but that cycle seems to have left the drawing board now.
Hey Scott, in all this time almost no-one talks about catalyst converters in mono-propellant engines, such as the catalyst pellets depicted in the diagram of Redstone's steam turbine. Think you could an episode on that? Like, exactly is it? What is it made of? What is the chemical reaction taking place? Is it the same thing that makes the bubbles in my contact lens solution, etc. That would be cool.
So the dual expander cycle is kinda like the full-flow version of expander cycle engines.
I do wonder, though. Do rocket engines overall gain any significant performance when it preheats its fuel and/or oxidizer as a coolant before burning it?
A thing that would solve many of these problems is some sort of bearingless pump. Now there is no leak, so it doesn't matter what drives what. It'd be really cool if something like that existed for the pressures required...
Hi Scott Manley, I wonder if/when we will see the Aerospike engine in use? Thoughts?
Expander Bleed Cycle is just another word for, "turbo waste gate"... CHANGE MY MIND!
SHHHHHHHH TU tut tu tu tu to space!
@@onogrirwin ?
@@charadremur333 look up videos of the soud of a overturbo charged engine with a waste gate. shhh is the turbo noise. "TU" is the waste gate
My head hurts. All I remember from engineering class is, the wheels on the bus, go round and round!
Think of the turbine as a fancy wheel, then work forward from there.
@@scottmanley Hey, come on now Scott, it's 30 years since I was at university, bit late now!
I have a suggestion.. If you make the video as a documentary like A Discovery channel show instead of just narration, it will be very much good and interesting..
If you want more surface area in your combustion chamber, can't you just put vertical "ribs" that protrude from the internal walls towards the center?