SpaceX Raptor engine - General Thread 4

[Chris Bergin]

Master Thread 4 (General, updates and discussion) for the Raptor Engine.

Thread 1:
https://forum.nasaspaceflight.com/index.php?topic=41363.0

Thread 2:
https://forum.nasaspaceflight.com/index.php?topic=47506.0

Thread 3:
https://forum.nasaspaceflight.com/index.php?topic=52988.0

--

A 2016 original baseline article:
https://www.nasaspaceflight.com/2016/10/its-propulsion-evolution-raptor-engine/ - by Alejandro G. Belluscio.

Additional Articles relating to Raptor:
https://www.nasaspaceflight.com/?s=Raptor

---
L2 Resources:
https://forum.nasaspaceflight.com/index.php?board=60.0

Includes additional resources ranging back to the start of the McGregor test stand facility for Raptor through to subscale and current Raptors on the stand hires and continually updated overhead photos etc. Now following each new Raptor going to the test stand (SN50/60 range)  Additional L2 areas for performance and vehicle content.

--

All rules apply here. On topic. Civil. Etc.

[Alberto-Girardi]

A few days ago there was a picture showing  six Raptors in a tent.

https://twitter.com/delta_v/status/1379912546430418945

If I saw correctly are new generation raptor (those compatible with sn15+). Are there more old generation raptors? If yes, can they being upgraded or it isn't worth it, even if possible?

[Moritz Kibler]

A few days ago there was a picture showing  six Raptors in a tent.

https://twitter.com/delta_v/status/1379912546430418945

If I saw correctly are new generation raptor (those compatible with sn15+). Are there more old generation raptors? If yes, can they being upgraded or it isn't worth it, even if possible?

From what we know from Mary's pictures there could be one old gen Raptor remaining from that we know of I think (and of course flight proven ones from SN5 and SN6). But also if there where more than that I really do not think that SpaceX would upgrade them. There are just to many changes in plumbing layout, TVC arms, etc.

[Pueo]

It seems to me that the drive for the 30 MPa chamber pressure is almost entirely an issue of thrust, not ISP.  Vacuum raptors, with their space constrained expansion ratio barely gain any Isp (<1 s) in the shift from 25 MPa to 30 MPa.  Sea level raptors gain more on account of their expansion ratio being tied to an exit pressure of ~0.8 atm, but it's only ~3 s.  If the full stack can reach a target orbit with 100 tonnes of payload when running at 30 MPa it can reach the same orbit running at 25 MPa with 96 tonnes of payload, even in the worst case scenario where all three SL raptors on Starship have to be used for the entire burn.

Chamber pressure is even less of a factor for the 6 month transit time Mars burn of ~4.5 km/s.  Again assuming the worst case, that all three SL raptors need to be used to get to the propellant transfer orbit and one SL raptor needs to be used on the Mars burn for gimbal control, it only takes one additional tanker of fuel for the 25 MPa vs the 30 MPa, and that additional tanker would have 73 tonnes of spare payload.
Heck, if you're only concerned with Isp you can go as low as the BE-4 at 13 MPa and still make it to orbit with 78 tonnes of payload, and to Mars with 8 tanker fill ups! (vs 6 for 30 MPa)

The problem is thrust.  With the 28 standard sea-level raptors at 25 MPa the "lift off" TWR is 0.91.  Even at 30 MPa the TWR is only 1.09 which also helps explain why SpaceX is willing to use a worse expansion ratio in order to get more thrust with the non-throttling raptors.  The full stack can achieve a decent TWR of 1.20 using the current 25 MPa sea level raptors, but it would require 37 engines, literally the maximum number of 1.3 m diameter nozzles you can fit in a 9 m diameter circle.  _{That packing would have a rather attractive C6 symmetry group though, and we could talk about the chirality of Super Heavy...}

[baldusi]

Have you stopped to think about T/W? That translates directly to PMF improvements.

[InterestedEngineer]

The problem is thrust.  With the 28 standard sea-level raptors at 25 MPa the "lift off" TWR is 0.91.  Even at 30 MPa the TWR is only 1.09 which also helps explain why SpaceX is willing to use a worse expansion ratio in order to get more thrust with the non-throttling raptors.  The full stack can achieve a decent TWR of 1.20 using the current 25 MPa sea level raptors, but it would require 37 engines, literally the maximum number of 1.3 m diameter nozzles you can fit in a 9 m diameter circle.  _{That packing would have a rather attractive C6 symmetry group though, and we could talk about the chirality of Super Heavy...}

More precisely, chamber pressure is about thrust flux.  There's only so many square meters to put rocket engines on the bottom of a very tall stainless steel tank.

T/W is only slightly related to thrust flux and chamber pressure, because T/W is related to the mass of the engine, not the area it takes up.  Make the engine out of very dense tungsten, T/W changes but area doesn't change.

[hkultala]

It seems to me that the drive for the 30 MPa chamber pressure is almost entirely an issue of thrust, not ISP.  Vacuum raptors, with their space constrained expansion ratio barely gain any Isp (<1 s) in the shift from 25 MPa to 30 MPa.

At vacuum.

But that diffefence in pressure might be critical in allowing the vacuum-optimized engines to be safely operated in dense atmosphere.

Needed for example in case when Starship needs to abort using its own engines when there is trouble with the SuperHeavy.

And the difference in isp will be greater even at some medium-pressure atmosphere.

[schuttle89]

It seems to me that the drive for the 30 MPa chamber pressure is almost entirely an issue of thrust, not ISP.  Vacuum raptors, with their space constrained expansion ratio barely gain any Isp (<1 s) in the shift from 25 MPa to 30 MPa.  Sea level raptors gain more on account of their expansion ratio being tied to an exit pressure of ~0.8 atm, but it's only ~3 s.  If the full stack can reach a target orbit with 100 tonnes of payload when running at 30 MPa it can reach the same orbit running at 25 MPa with 96 tonnes of payload, even in the worst case scenario where all three SL raptors on Starship have to be used for the entire burn.

Chamber pressure is even less of a factor for the 6 month transit time Mars burn of ~4.5 km/s.  Again assuming the worst case, that all three SL raptors need to be used to get to the propellant transfer orbit and one SL raptor needs to be used on the Mars burn for gimbal control, it only takes one additional tanker of fuel for the 25 MPa vs the 30 MPa, and that additional tanker would have 73 tonnes of spare payload.
Heck, if you're only concerned with Isp you can go as low as the BE-4 at 13 MPa and still make it to orbit with 78 tonnes of payload, and to Mars with 8 tanker fill ups! (vs 6 for 30 MPa)

The problem is thrust.  With the 28 standard sea-level raptors at 25 MPa the "lift off" TWR is 0.91.  Even at 30 MPa the TWR is only 1.09 which also helps explain why SpaceX is willing to use a worse expansion ratio in order to get more thrust with the non-throttling raptors.  The full stack can achieve a decent TWR of 1.20 using the current 25 MPa sea level raptors, but it would require 37 engines, literally the maximum number of 1.3 m diameter nozzles you can fit in a 9 m diameter circle.  _{That packing would have a rather attractive C6 symmetry group though, and we could talk about the chirality of Super Heavy...}

Made me think a lot, great post. If that extra thrust allows less engines to be used and slightly less complicated fuel routing then it seems worthwhile.

[ericgu]

The problem is thrust.  With the 28 standard sea-level raptors at 25 MPa the "lift off" TWR is 0.91.  Even at 30 MPa the TWR is only 1.09 which also helps explain why SpaceX is willing to use a worse expansion ratio in order to get more thrust with the non-throttling raptors.  The full stack can achieve a decent TWR of 1.20 using the current 25 MPa sea level raptors, but it would require 37 engines, literally the maximum number of 1.3 m diameter nozzles you can fit in a 9 m diameter circle.  _{That packing would have a rather attractive C6 symmetry group though, and we could talk about the chirality of Super Heavy...}

Exactly this. Having played around with numbers, starship looks pretty good with a thrust/weight of 0.94 with 100 tons of payload, plenty for a second stage (I have Falcon 9 at 0.77 for a starlink launch).

But SH + Starship (100 ton payload) is just too heavy for 28 engines. I was getting a 1.12 with my estimates, but that's just too low - at a low thrust/weight you lose so much due to gravity losses. My Falcon 9 first stage numbers say T/W of 1.34 for a Starlink launch, and that's the range I'd expect them to do.

The "super raptor" variant for the outer 20 engines would give a T/W of 1.48. That would be spicey.

[edzieba]

That matches with the "about 1.5" TWR Elon previously gave:

https://twitter.com/elonmusk/status/1355627125802299393

[schuttle89]

The problem is thrust.  With the 28 standard sea-level raptors at 25 MPa the "lift off" TWR is 0.91.  Even at 30 MPa the TWR is only 1.09 which also helps explain why SpaceX is willing to use a worse expansion ratio in order to get more thrust with the non-throttling raptors.  The full stack can achieve a decent TWR of 1.20 using the current 25 MPa sea level raptors, but it would require 37 engines, literally the maximum number of 1.3 m diameter nozzles you can fit in a 9 m diameter circle.  _{That packing would have a rather attractive C6 symmetry group though, and we could talk about the chirality of Super Heavy...}

Exactly this. Having played around with numbers, starship looks pretty good with a thrust/weight of 0.94 with 100 tons of payload, plenty for a second stage (I have Falcon 9 at 0.77 for a starlink launch).

But SH + Starship (100 ton payload) is just too heavy for 28 engines. I was getting a 1.12 with my estimates, but that's just too low - at a low thrust/weight you lose so much due to gravity losses. My Falcon 9 first stage numbers say T/W of 1.34 for a Starlink launch, and that's the range I'd expect them to do.

The "super raptor" variant for the outer 20 engines would give a T/W of 1.48. That would be spicey.

According to a cursory Google search, the Saturn V had a 1.2 T/W. So my question is how low is too low? I thought anything over 1.0 was acceptable depending on what you are trying to optimize for.

[Herb Schaltegger]

According to a cursory Google search, the Saturn V had a 1.2 T/W. So my question is how low is too low? I thought anything over 1.0 was acceptable depending on what you are trying to optimize for.

Depending on ISP (which translates into how fast you’re moving propellant mass through the engine), “too low” is the point at which your vehicle loses more delta-V to gravity than it can afford to lose. Launch from Earth and you are fighting two things: gravity and atmospheric drag. If your T/W ratio is too low, you’re wasting prop by not accelerating quickly. Too high and you end up with huge drag losses and aerothermal heating in the dense lower atmosphere.

Like all engineering, there’s a lot of balancing and trade-offs involved.

[Keldor]

According to a cursory Google search, the Saturn V had a 1.2 T/W. So my question is how low is too low? I thought anything over 1.0 was acceptable depending on what you are trying to optimize for.

Depending on ISP (which translates into how fast you’re moving propellant mass through the engine), “too low” is the point at which your vehicle loses more delta-V to gravity than it can afford to lose. Launch from Earth and you are fighting two things: gravity and atmospheric drag. If your T/W ratio is too low, you’re wasting prop by not accelerating quickly. Too high and you end up with huge drag losses and aerothermal heating in the dense lower atmosphere.

Like all engineering, there’s a lot of balancing and trade-offs involved.

Propulsive landings change the dynamics too.  For Falcon 9 or Super Heavy, any extra tankage is all the more mass you have to turn around and bring back to the landing site.  This is very different from the expendable model where you can tolerate 80% gravity losses for a portion of the flight and still come out ahead because the dry mass for a booster doesn't matter so much - it will always have a heavy second stage on top of it.

So the dry mass of Super Heavy matters a lot more than an expendable stage, and thus they can't go as far down the road of diminishing gains lengthening the stage to gain a small bit of extra boost at the cost of low TWR wasting most, but not all, of the gains.

[schuttle89]

According to a cursory Google search, the Saturn V had a 1.2 T/W. So my question is how low is too low? I thought anything over 1.0 was acceptable depending on what you are trying to optimize for.

Depending on ISP (which translates into how fast you’re moving propellant mass through the engine), “too low” is the point at which your vehicle loses more delta-V to gravity than it can afford to lose. Launch from Earth and you are fighting two things: gravity and atmospheric drag. If your T/W ratio is too low, you’re wasting prop by not accelerating quickly. Too high and you end up with huge drag losses and aerothermal heating in the dense lower atmosphere.

Like all engineering, there’s a lot of balancing and trade-offs involved.

Propulsive landings change the dynamics too.  For Falcon 9 or Super Heavy, any extra tankage is all the more mass you have to turn around and bring back to the landing site.  This is very different from the expendable model where you can tolerate 80% gravity losses for a portion of the flight and still come out ahead because the dry mass for a booster doesn't matter so much - it will always have a heavy second stage on top of it.

So the dry mass of Super Heavy matters a lot more than an expendable stage, and thus they can't go as far down the road of diminishing gains lengthening the stage to gain a small bit of extra boost at the cost of low TWR wasting most, but not all, of the gains.

Alright that makes a lot more sense thanks guys.

[hkultala]

According to a cursory Google search, the Saturn V had a 1.2 T/W. So my question is how low is too low? I thought anything over 1.0 was acceptable depending on what you are trying to optimize for.

Depending on ISP (which translates into how fast you’re moving propellant mass through the engine), “too low” is the point at which your vehicle loses more delta-V to gravity than it can afford to lose. Launch from Earth and you are fighting two things: gravity and atmospheric drag. If your T/W ratio is too low, you’re wasting prop by not accelerating quickly. Too high and you end up with huge drag losses and aerothermal heating in the dense lower atmosphere.

Like all engineering, there’s a lot of balancing and trade-offs involved.

Propulsive landings change the dynamics too.  For Falcon 9 or Super Heavy, any extra tankage is all the more mass you have to turn around and bring back to the landing site.  This is very different from the expendable model where you can tolerate 80% gravity losses for a portion of the flight and still come out ahead because the dry mass for a booster doesn't matter so much - it will always have a heavy second stage on top of it.

So the dry mass of Super Heavy matters a lot more than an expendable stage, and thus they can't go as far down the road of diminishing gains lengthening the stage to gain a small bit of extra boost at the cost of low TWR wasting most, but not all, of the gains.

Also, with high T/W, the staging happens faster and you have moved less horizontal distance when that staging happens, while you still have about as much velocity, so you stay in the air for equal total time (more remaining time)

And when you have less distance but more time for your way back, that means you can travel back at considerably slower horizontal speed => smaller boostback burn needed, less fuel spent for the boostback burm.

[Keldor]

According to a cursory Google search, the Saturn V had a 1.2 T/W. So my question is how low is too low? I thought anything over 1.0 was acceptable depending on what you are trying to optimize for.

Depending on ISP (which translates into how fast you’re moving propellant mass through the engine), “too low” is the point at which your vehicle loses more delta-V to gravity than it can afford to lose. Launch from Earth and you are fighting two things: gravity and atmospheric drag. If your T/W ratio is too low, you’re wasting prop by not accelerating quickly. Too high and you end up with huge drag losses and aerothermal heating in the dense lower atmosphere.

Like all engineering, there’s a lot of balancing and trade-offs involved.

Propulsive landings change the dynamics too.  For Falcon 9 or Super Heavy, any extra tankage is all the more mass you have to turn around and bring back to the landing site.  This is very different from the expendable model where you can tolerate 80% gravity losses for a portion of the flight and still come out ahead because the dry mass for a booster doesn't matter so much - it will always have a heavy second stage on top of it.

So the dry mass of Super Heavy matters a lot more than an expendable stage, and thus they can't go as far down the road of diminishing gains lengthening the stage to gain a small bit of extra boost at the cost of low TWR wasting most, but not all, of the gains.

Also, with high T/W, the staging happens faster and you have moved less horizontal distance when that staging happens, while you still have about as much velocity, so you stay in the air for equal total time (more remaining time)

And when you have less distance but more time for your way back, that means you can travel back at considerably slower horizontal speed => smaller boostback burn needed, less fuel spent for the boostback burm.

Yes, you can benefit from a higher arcing trajectory, even with greater gravity losses.  This of course lowers your payload vs. expendable, but it makes the boostback cheaper, which ends up a win for a reusable rocket trajectory.  Of course, this means that TWR is all the more important.

Boostback is a high delta-V maneuver in any case.  The only reason it's feasible at all is because you shed half the mass (or much more for Starship!) of the rocket at staging.  From the booster's point of view, the second stage is entirely dry mass, and so shedding it gives it a huge boost in delta-V to get back to the landing site.

This is also why reusable rockets benefit from a much larger second stage than expendable.  The lighter the booster is, the easier it is to get back to the landing site, so the optimal balance is different.  For Starship, the wet mass of the second stage is somewhere around 5 times that of the empty booster, whereas for Atlas V, to pick an expendable rocket at random to compare, the ratio is more like 1:1.  This seems to vary widely between rockets, though, and it's hard to find rockets than don't have strap on boosters in most configurations (Falcon 9 doesn't count because the booster is resuable, so again, different dynamics), which would effective make them 2 1/2 stage rockets which probaby skews the numbers.

Rockets with just the booster stage have different dynamics too.  Here you get to balance the delta-V cost of the boostback against the pricetag of the expended upper stage.  This also influences whether ASDS or RTLS landings are preferred when the launcher gets to pick the payload size for optimal launch costs (i.e. whether it's better to launch fewer Starlink satellites at a time and return to land or launch more at a time and land on the drone ship).  Ships are expensive to operate, but compared to the price of a brand new second stage, not *that* expensive.  Starship is probably different in this regard, since firstly any truely rapid launch cadance can't wait half a week for the drone ship to be brought to the landing zone and then back to shore, but also because the second stage isn't expended.

[Alberto-Girardi]

There was a post, that I can't find, in which was sais that a wors expantion ratio could be a option, maybe temporaney,  to gwt more theust from a Raptor. "Worse" means higher 1:e (e is the exit area in throat area unit) or lower? Why is considerable worse?

[rsdavis9]

There was a post, that I can't find, in which was sais that a wors expantion ratio could be a option, maybe temporaney,  to gwt more theust from a Raptor. "Worse" means higher 1:e (e is the exit area in throat area unit) or lower? Why is considerable worse?

Might of been me.
So given musk said:
300bar
300 tonne force
I worked up in rpa-lite an engine with the same exit diameter.
So the only thing you can change is the throat diameter.
The sl raptor has a throat diameter of .2216m from the KSC EIS doc.
So I played around with the throat diameter until I got 300tf.

         * no gimbal/throttle booster engine
         Pc         = 30MPa
         ER          = 24
         diam throat = .2653m
         diam exit   = 1.3 m
         Isp sl      = 332.1s
         Isp vac     = 347.7s
         thrust sl   = 291 t-f
         thrust vac  = 305 t-f
         mass flow    = 876.4kg/s
         Pe         = .1466 MPa

Here is the sl raptor for comparision:

         * current SL Raptor
         Pc         = 30MPa
         ER          = 34.4
         diam throat = .2216 m
         diam exit   = 1.3 m
         Isp sl      = 332.9 s
         Isp vac     = 355.3 s
         thrust sl   = 203.7 t-f
         thrust vac  = 217.4 t-f
         mass flow    = 611.9 kg/s
         Pe         = .092 MPa

[Redclaws]

There was a post, that I can't find, in which was sais that a wors expantion ratio could be a option, maybe temporaney,  to gwt more theust from a Raptor. "Worse" means higher 1:e (e is the exit area in throat area unit) or lower? Why is considerable worse?

So, a higher expansion ratio is, absent other concerns, better, so a lower expansion ratio is worse.  There are trade offs to making engine bells bigger and it is not always possible to do, etc, but a higher expansion ratio extracts more thrust from the exhaust.  In atmosphere, maximum expansion ratio is severely restricted by the atmosphere, but in vacuum in theory an engine with an *infinite* expansion ratio would be best.

[CorvusCorax]

There was a post, that I can't find, in which was sais that a wors expantion ratio could be a option, maybe temporaney,  to gwt more theust from a Raptor. "Worse" means higher 1:e (e is the exit area in throat area unit) or lower? Why is considerable worse?

So, a higher expansion ratio is, absent other concerns, better, so a lower expansion ratio is worse.  There are trade offs to making engine bells bigger and it is not always possible to do, etc, but a higher expansion ratio extracts more thrust from the exhaust.  In atmosphere, maximum expansion ratio is severely restricted by the atmosphere, but in vacuum in theory an engine with an *infinite* expansion ratio would be best.

Yeah. There is allways multiple ways to look at the equation:
1. You leave flow rate, chamber pressure and throat area identical. You increase the nozzle size
This increases ISP - with little change to thrust (slightly higher thrust at higher ISP), but large change to area and weight
2. You leave the nozzle size and the chamber pressure but change the throat size
This leaves area identical but an increase in ISP will come with a decrease in flowrate and as such thrust. it will also reduce thrust to weight ratio
3. You leave throat and nozzle but change chamber pressure
with higher chamber pressure, ISP will only increase marginally, but thrust and thrust2weight will increase significantly

now of course you can run combos. you could
4, change both chamber pressure and throat size together to keep thrust constant. this will give the highest ISP gains at constant nozzle size, and little to no change in thrust2weight

or you can do a mix of 3 and 4 to get an increase in both ISP and thrust. Increase chamber pressure, reduce throat size a little bit, but not much, so the flow rate is still higher but not as high as it would be with no change.

I think this is possibly the best route to go if you want an overall better engine, but have an area limit at the bottom of the stage.

I can see why SpaceX went for super high chamber pressure