IR movie of falcon 9 boost back and reentry by NASA

[Jet Black]

Enjoy

http://www.nasa.gov/press/2014/october/new-commercial-rocket-descent-data-may-help-nasa-with-future-mars-landings/#.VEHbqqakqJI

[malu5531]

I'm quite shocked this actually works, Wow!

[Jet Black]

I really like the boost back at 1:05

[Dalhousie]

What velocity is the F9 1st stage travelling at at different times?

[FutureSpaceTourist]

Very cool video. Elon has tweeted it:

https://twitter.com/elonmusk/status/523290693612367872

@elonmusk: Infrared video of Falcon rocket reentry captured by @NASA tracking cameras http://t.co/GQLCFLlrUC

I believe F9 1st stage separation occurs at about Mach 6.

[sdsds]

 OK, forget the ponies. For Christmas may I please have a 75-channel imaging spectrometer in the 2.0 - 6.4 μm IR bands?

Please?

That's some serious imager!

[Dalhousie]

I believe F9 1st stage separation occurs at about Mach 6.

So strictly speaking this is hypersonic retropropulsion?

[Norm38]

It's evident from that video how the sides of the first stage got blackened on reentry.  (As shown by the clean white areas covered by the legs).  The plume is quite narrow.

But very close to expected.  Not too far off from sci-fi imaginings.

[Herb Schaltegger]

That high-altitude plume imagery is fantastic!

[Comga]

That high-altitude plume imagery is fantastic!

It is, indeed!
The plume looks very turbulent. It would seemime a strong ACS would be needed to keep the stage stable.

[Jcc]

I believe F9 1st stage separation occurs at about Mach 6.

So strictly speaking this is hypersonic retropropulsion?

I think the boost back may have cut some of the velocity, and that happens in vacuum, but yes.

From what we have seen the reentry burn doesn't start until it begins entering the atmosphere, so there is already some drag, and it manages the drag and heating as the stage gets into thicker atmosphere by bringing the velocity down below Mach 3 (my guess).

[Targeteer]

Anyone have more details on the sensor package used? I can't find a likely candidate on the NASA list of WB-57 payloads.

http://jsc-aircraft-ops.jsc.nasa.gov/WB57/instrumentsandlinks.html

[Dalhousie]

I believe F9 1st stage separation occurs at about Mach 6.

So strictly speaking this is hypersonic retropropulsion?

I think the boost back may have cut some of the velocity, and that happens in vacuum, but yes.

From what we have seen the reentry burn doesn't start until it begins entering the atmosphere, so there is already some drag, and it manages the drag and heating as the stage gets into thicker atmosphere by bringing the velocity down below Mach 3 (my guess).

Thanks.  Is there a velocity-altitude plot available?

[sdsds]

Anyone have more details on the sensor package used?

I'm guessing it's this one:
Airborne Remote Earth Sensing (ARES) Program: an operational airborne MWIR imaging spectrometer and applications

http://proceedings.spiedigitallibrary.org/proceeding.aspx?articleid=1021464

[Jdeshetler]

Repackaged this cool video by NASA:
 - 2X close-up
 - enlarged NASA's time clock
 - video stabilized
 - 3X speed up to match the real-time (close as possible)

www.youtube.com/embed/yaCFNvyNol8

Cheers,

[Proponent]

At 0:24 in the video, the first stage is described as maneuvering out of the 2nd stage's plume, and it looks as though there are multiple pulses from a rocket engine(s).  Does this mean the Merlins are re-ignited many times post-separation (multiple maneuvering pulses, the boost-back burn near apogee, the 70-to-40-km deceleration burn [I presume that's what the "re-entry burn" is], and the landing burn)?

[guckyfan]

At 0:24 in the video, the first stage is described as maneuvering out of the 2nd stage's plume, and it looks as though there are multiple pulses from a rocket engine(s).  Does this mean the Merlins are re-ignited many times post-separation (multiple maneuvering pulses, the boost-back burn near apogee, the 70-to-40-km deceleration burn [I presume that's what the "re-entry burn" is], and the landing burn)?

That looked really weird. The bursts seem much too short for Merlin ignitions. But I already mentioned that I would expect not to see cold gas thruster firings in the infrared imaging. I even wondered if they have replaced cold gas with Draco when they "beefed up" RCS.

[MikeAtkinson]

At 0:24 in the video, the first stage is described as maneuvering out of the 2nd stage's plume, and it looks as though there are multiple pulses from a rocket engine(s).  Does this mean the Merlins are re-ignited many times post-separation (multiple maneuvering pulses, the boost-back burn near apogee, the 70-to-40-km deceleration burn [I presume that's what the "re-entry burn" is], and the landing burn)?

That looked really weird. The bursts seem much too short for Merlin ignitions. But I already mentioned that I would expect not to see cold gas thruster firings in the infrared imaging. I even wondered if they have replaced cold gas with Draco when they "beefed up" RCS.

The cold gas hits the airstream / Merlin Vac exhaust and is heated.

[Llian Rhydderch]

Repackaged this cool video by NASA:
 - 2X close-up
 - enlarged NASA's time clock
 - video stabilized
 - 3X speed up to match the real-time (close as possible)

www.youtube.com/embed/yaCFNvyNol8

Cheers,

Very helpful edits, Jd! 

Nice to have a version with the speed approximated to the many videos we've been looking at recently of the other parts of the controlled-descent flight test profile, and the visible time clock helps tool.

[rockettrey]

This is absolutely incredible!  The boost back burn is amazing and somewhat easy to comprehend, but what is the reentry burn video actually showing?  I thought the boost back burn did its thing way up in the vacuum after separation, and the stage only re-ignighted just before landing. Is there another burn of the Merlins during re-entry, or is that just atmospheric heating searing the entire stage that causes the fireball?

[Lars-J]

This is absolutely incredible!  The boost back burn is amazing and somewhat easy to comprehend, but what is the reentry burn video actually showing?  I thought the boost back burn did its thing way up in the vacuum after separation, and the stage only re-ignighted just before landing. Is there another burn of the Merlins during re-entry, or is that just atmospheric heating searing the entire stage that causes the fireball?

Yes, a burn is started just as the stage hits the upper atmosphere to slow it down. You can see that burn from the points of view of the stage itself here: (it looks less violent in visible light)



The first stage does *3* restarts after separation:
1. Boost-back burn
2. Braking burn
3. Landing burn

[simonbp]

The first stage does *3* restarts after separation:
1. Boost-back burn
2. Braking burn
3. Landing burn

Where the second is called the "reentry burn" in the video, and is the most difficult/Mars-relevant, as the rockets are firing directly into a supersonic (incompressible) flow. That's the part that JPL has been scared to do for decades, and why no one has attempted this type of booster recovery before.

I distinctly recall Rob Manning telling me in 2006 that if anyone figures out how to do supersonic retropropulsion, they'd be rich...

[rpapo]

I distinctly recall Rob Manning telling me in 2006 that if anyone figures out how to do supersonic retropropulsion, they'd be rich...

I'm not an aerospace engineer (though I wanted to be at one point), and so never had any significant schooling in fluid dynamics, let alone supersonic fluid dynamics, but it strikes me that the key to this comes down to two facts:

(1) The oncoming flow is supersonic, and theoretically incompressible, but relatively weak.
(2) The exhaust flow is also supersonic, but quite strong.

Something has to give, and that IR video shows that happening.  This is not a case of the irresistible force versus the immovable object.

Somebody who knows better can now shoot me down . . . or not.

[Lars-J]

I think part of the problem with modeling hypersonic retro-propulsion is that there has been so little data to use as a basis for a model. Wind tunnels have a hard time simulating the exact conditions.

At these speeds, the outcome might not always be what you expect. (for example a low thrust creates a bubble that lowers the friction of the craft, thus making it fall *faster* than it would have otherwise)

There is a thread here that discusses hypersonic retro-propulsion issues: http://forum.nasaspaceflight.com/index.php?topic=33006.0

[blazotron]

<snip>
Where the second is called the "reentry burn" in the video, and is the most difficult/Mars-relevant, as the rockets are firing directly into a supersonic (incompressible) flow.
<snip>

<snip>
(1) The oncoming flow is supersonic, and theoretically incompressible, but relatively weak.
<snip>

Supersonic flow is highly compressible, not incompressible.  Compressibility is one of the factors that distinguishes flow above about Mach 0.3.  However, the general idea here that the oncoming flow is "weak" (i.e. low density) during the high altitude entry burn is the right way to think about it.  The other important thing to keep in mind is that the freestream gas is cold, while the exhaust exiting the engines is hot.  Because speed of sound scales as the square root of temperature and gas velocity scales as Mach number times speed of sound, even if both gases were at say Mach 3, the rocket exhaust would have a much higher velocity.  The plume penetrates into the freestream for some distance after leaving the rocket engines, but at some point it is turned back in the opposite direction along the rocket body (from the perspective of the rocket) by transferring momentum to the freestream such that the rocket basically catches up to and passes to its own exhaust. 

[llanitedave]

I distinctly recall Rob Manning telling me in 2006 that if anyone figures out how to do supersonic retropropulsion, they'd be rich...

Turns out you have to already be rich to do it.

[Norm38]

But the beauty of what SpaceX did is figure it out using spent first stages. So they didn't spend much extra, and the potential for reuse means it should pay for itself. That changes the equation. 
I just love that industry is finally progressing to the point where it's breaking bariers.

[Llian Rhydderch]

<snip>
Where the second is called the "reentry burn" in the video, and is the most difficult/Mars-relevant, as the rockets are firing directly into a supersonic (incompressible) flow.
<snip>

<snip>
(1) The oncoming flow is supersonic, and theoretically incompressible, but relatively weak.
<snip>

Supersonic flow is highly compressible, not incompressible.  Compressibility is one of the factors that distinguishes flow above about Mach 0.3.  However, the general idea here that the oncoming flow is "weak" (i.e. low density) during the high altitude entry burn is the right way to think about it.  The other important thing to keep in mind is that the freestream gas is cold, while the exhaust exiting the engines is hot.  Because speed of sound scales as the square root of temperature and gas velocity scales as Mach number times speed of sound, even if both gases were at say Mach 3, the rocket exhaust would have a much higher velocity.  The plume penetrates into the freestream for some distance after leaving the rocket engines, but at some point it is turned back in the opposite direction along the rocket body (from the perspective of the rocket) by transferring momentum to the freestream such that the rocket basically catches up to and passes to its own exhaust.

That is a very helpful explanation, Blazotron.

Thank you for taking the time to offer it up for all of us to benefit from. 

[Proponent]

This exchange in another thread makes me wonder, just how tough was it for SpaceX to perform a supersonic retro burn?

The video certainly makes it look easy, but it seems to have been something that had been regarded with apprehension for years.  Do we know of any unusual features of the Merlin 1D that make the burn possible?  Did SpaceX luck out in that some big guesses panned out, or was it pretty sure thing from the beginning that there were no fundamental problems with its approach?

To put it another way, what was the TRL before (and after) CRS-4?

[Dudely]

Man, they've got some immensely huge brass balls to even try this. Mad props to those engineers.

[R7]

Do we know of any unusual features of the Merlin 1D that make the burn possible?

Ability to do an air start. AFAIK booster engines usually can't, relying on GSE during start-up sequence.

[guckyfan]

Do we know of any unusual features of the Merlin 1D that make the burn possible?

Ability to do an air start. AFAIK booster engines usually can't, relying on GSE during start-up sequence.

That question was related to supersonic retropropulsion, I believe.

No, Merlin does not have any special feature that enables supersonic retropropulsion. What was not well known was the interaction of the engine exhaust with the supersonic stream. This knowledge could only be gained by doing it and study the interaction. Which is why this video is so valuable.

The gathered data can probably help to do even better analysis of the Red Dragon concept.

[Llian Rhydderch]

Do we know of any unusual features of the Merlin 1D that make the burn possible?

Ability to do an air start. AFAIK booster engines usually can't, relying on GSE during start-up sequence.

Yes, the air start was just one of the technologies needed to make it work.  In my view, it wasn't the technologies themselves that prevented this from being done before now, or did not seem to be possible.  Each of those technologies, individually, is an eminently doable engineering project.

Rather, the critical element was the entrepreneurial idea to put them together and fund the requisite development project so that the details could be engineered, bugs worked out, and the tech could be refined and finish development.  Musk did that, and he did that as a private entrepreneur who is the principal residual claimant on the success or failure of such an undertaking. 

The ideas to do this might have been there for many years.  Heck, the ideas date back to, and were widely illustrated in, Buck Rogers shortly after Robert Goddard was flying small gasoline rockets.   But the incentives faced by people in government  entities are quite different, and no government ever made the sustained commitment to make it happen.


^{Edit:  fixed link to provide a list of the various SpaceX reusability programme technologies}

[MP99]

This exchange in another thread makes me wonder, just how tough was it for SpaceX to perform a supersonic retro burn?

The video certainly makes it look easy, but it seems to have been something that had been regarded with apprehension for years.  Do we know of any unusual features of the Merlin 1D that make the burn possible?  Did SpaceX luck out in that some big guesses panned out, or was it pretty sure thing from the beginning that there were no fundamental problems with its approach?

To put it another way, what was the TRL before (and after) CRS-4?

You can look at it in a different way.

SpaceX has produced a successful expendable launcher. Merlin 1D vac needs the ability to perform restarts, and there is much commonality between that and the SL version used on the first stage.

The actual risk to SpaceX in performing these tests was relatively low, since they could be done post-MECO on flights that they were being paid for anyway.

Of course, if these tests had failed then SpaceX's future direction would have been very different (Elon will happily tell you that such a future would be a failure).

To some extent, this particular mode of recovery is a matter of how late you can leave the reentry burn (for a greater payload), or very early (with a bigger hit to payload, but less difficult conditions for the burn).

But, the great thing is that they have been able to run these tests without risking the future of the company on the outcome.

The ballsier thing is their overall reusability programme - Grasshopper, et al, and an attempt to grow the overall launch market.

Cheers, Martin

[RanulfC]

Do we know of any unusual features of the Merlin 1D that make the burn possible?

Ability to do an air start. AFAIK booster engines usually can't, relying on GSE during start-up sequence.

Yes, the air start was just one of the technologies needed to make it work.  In my view, it wasn't the technologies themselves that prevented this from being done before now, or did not seem to be possible.  Each of those technologies, individually, is an eminently doable engineering project.

Rather, the critical element was the entrepreneurial idea to put them together and fund the requisite development project so that the details could be engineered, bugs worked out, and the tech could be refined and finish development.  Musk did that, and he did that as a private entrepreneur who is the principal residual claimant on the success or failure of such an undertaking. 

The ideas to do this might have been there for many years.  Heck, the ideas date back to, and were widely illustrated in, Buck Rogers shortly after Robert Goddard was flying small gasoline rockets.   But the incentives faced by people in government  entities are quite different, and no government ever made the sustained commitment to make it happen.

Pardon my bolding the relvent above The main problem has been a lot of myopic "vision" on how things were supposed to be by a lot of people over the years. Elon's been one willing to fund a more focused vision that did NOT fall to most of the assumptions that have been developed over the decades since Tsiolkovsky stated the basic problems and provided sugested solutions. (Note: This doesn't mean Elon hasn't had his own issues with vision-eering to reality, he has)

Everything from "RTLS has to be airplane-like, so requires wings-and-wheels" to "Once we have LH2/LOX engines we will have SSTO-airplane-like-Launch-Vehicles" was out there as "common knowledge" even though evidence to the contrary was ALSO out there in the common-knowledge pool

Sometimes it really does take someone NOT listening to the "commen-sense" answers and going their own way to get the job done

Randy

[rpapo]

Sometimes it really does take someone NOT listening to the "commen-sense" answers and going their own way to get the job done

Somebody else said it very succinctly: Question Everything, Assume Nothing. 

There is a time and a place to throw a whole bunch of bright kids at a problem.  They don't know yet what they cannot do, and sometimes discover that it can be done.

[llanitedave]

But there are still those here who claim SpaceX doesn't innovate.

[RanulfC]

But there are still those here who claim SpaceX doesn't innovate.

Depends on how you mean "innovate"

Randy

[Dalhousie]

This exchange in another thread makes me wonder, just how tough was it for SpaceX to perform a supersonic retro burn?

The video certainly makes it look easy, but it seems to have been something that had been regarded with apprehension for years.  Do we know of any unusual features of the Merlin 1D that make the burn possible?  Did SpaceX luck out in that some big guesses panned out, or was it pretty sure thing from the beginning that there were no fundamental problems with its approach?

To put it another way, what was the TRL before (and after) CRS-4?

IMHO it was one of those things that has been known to be feasible for decades but nobody tried it.  the longer people didn't try it the harder it seemed.

[Jcc]

I think part of the problem with modeling hypersonic retro-propulsion is that there has been so little data to use as a basis for a model. Wind tunnels have a hard time simulating the exact conditions.

At these speeds, the outcome might not always be what you expect. (for example a low thrust creates a bubble that lowers the friction of the craft, thus making it fall *faster* than it would have otherwise)

There is a thread here that discusses hypersonic retro-propulsion issues: http://forum.nasaspaceflight.com/index.php?topic=33006.0

If as stated, the exhaust stream creates a buble that reduces friction, this may actually be advantageous during reentry as it would reduce heating of the skin and shear forces, until the stage slows down enough so it can use friction to slow down further without burning up.

[Lars-J]

I think part of the problem with modeling hypersonic retro-propulsion is that there has been so little data to use as a basis for a model. Wind tunnels have a hard time simulating the exact conditions.

At these speeds, the outcome might not always be what you expect. (for example a low thrust creates a bubble that lowers the friction of the craft, thus making it fall *faster* than it would have otherwise)

There is a thread here that discusses hypersonic retro-propulsion issues: http://forum.nasaspaceflight.com/index.php?topic=33006.0

If as stated, the exhaust stream creates a buble that reduces friction, this may actually be advantageous during reentry as it would reduce heating of the skin and shear forces, until the stage slows down enough so it can use friction to slow down further without burning up.

This is not necessarily better. Ideally you want to bleed if your velocity during a longer time. If you reduce your friction and then hit the lower (and denser) atmosphere faster, you peak G loads and heating will likely be higher.

[Jcc]

I think part of the problem with modeling hypersonic retro-propulsion is that there has been so little data to use as a basis for a model. Wind tunnels have a hard time simulating the exact conditions.

At these speeds, the outcome might not always be what you expect. (for example a low thrust creates a bubble that lowers the friction of the craft, thus making it fall *faster* than it would have otherwise)

There is a thread here that discusses hypersonic retro-propulsion issues: http://forum.nasaspaceflight.com/index.php?topic=33006.0

If as stated, the exhaust stream creates a buble that reduces friction, this may actually be advantageous during reentry as it would reduce heating of the skin and shear forces, until the stage slows down enough so it can use friction to slow down further without burning up.

This is not necessarily better. Ideally you want to bleed if your velocity during a longer time. If you reduce your friction and then hit the lower (and denser) atmosphere faster, you peak G loads and heating will likely be higher.

Maybe ideally but that obviously doesn't work for the first stage flight profile, otherwise they would do it. If the stage re-enters at Mach 6 with no TPS and no retro propulsion it would burn up (or break up). With retro propulsion it doesn't break up. Ideally you would want the maximum friction the stage can take in addition to propulsive braking, but not breaking up is a good second choice.

[Torbjorn Larsson, OM]

There was speculation a few years ago, on the Red Dragon thread I believe, that having the superdraco plumes pointed outward along the rim would cause meaningful atmospheric concentration below the heat-shield as it passes through; increasing the atmospheric friction rather than trying to slip through it.  Adding surface area (ballutes, tabs, expandable heat shields, etc. would increase the friction further, as more atmosphere interacts with greater surface area. 

I'll speculate here; that the main challenge on Mars entry is maximizing the atmospheric drag while the main challenge on Earth is minimizing the friction on initial atmospheric impact.                             

Larry Lemke, lead on the Red Dragon project, described RD EDL and retrorocketry @ SETI Talks in June.



The detailed retrorocket segment is 50:00 - 55:00, with references.

The short of it is that:

- Low retrorocket thrust will blow out the shock front and increase drag, whether central or peripheral. Peripheral thrust will maintain that at higher thrust.

- For the RD and F9R S1 EDL, the thrust levels blow a hole in the shock front. They turn off the drag in the simulations since its effect is insignificant.

- RD will do an aerobrake at 5 km above the surface where the atmosphere is densest. (Segment starts at about 30:00.) I checked as best I could at the time, and IIRC that atmosphere height is roughly equivalent to the pressure/drag regime Apollo used when returning from the Moon.

To do that efficiently RD will use attack angle to dive as fast as possible.  Then use that control to maintain height (no Apollo skips, so more like 2-3 min at low g's I think) while drag slows RD until loss of loft at about Mach 3 (IIRC; you may want to check that). Finally do a supersonic brake to hover landing with 10 mt of fuel.

The brake trajectory is ~ 2500 km long, so you may want to start at Vastitas Borealis until you know what you are doing. VB is 8000 km+, so in the beginning drag coefficient is once again not much of an issue. [  http://en.wikipedia.org/wiki/North_Polar_Basin_(Mars) ]

[guckyfan]

Very interesting. While watching it an idea entered my mind. As there were pepople much smarter and knowledgeable than me at it they likely have discounted it already. But anyway here it is. Since it is Red Dragon specific I post it in the appropriate thread.

http://forum.nasaspaceflight.com/index.php?topic=33596.msg1274793#msg1274793