SpaceX Falcon 9 v1.1 - Dragon - CRS-5/SpX-5 -Jan. 10, 2015 - DISCUSSION

[JamesH]

According to interview with Musk, the fins ran out of fluid with a minute to go. Quite a long time...

[Kabloona]

According to interview with Musk, the fins ran out of fluid with a minute to go. Quite a long time...

Thinking out loud about margins...

He said the hydraulic fluid was sized for 4 minutes of flight control, and that they ended up 10% short on fluid quantity.

One minute out of 4 is 25% short, not 10% short. But they no doubt expected to land with some fluid left in the tank, in addition to the 10% that would have been needed to land successfully. So 10% needed to land, and 15% margin left in the tank. Sounds reasonable...

[JamesH]

According to interview with Musk, the fins ran out of fluid with a minute to go. Quite a long time...

Thinking out loud about margins...

He said the hydraulic fluid was sized for 4 minutes of flight control, and that they ended up 10% short on fluid quantity.

One minute out of 4 is 25% short, not 10% short. But they no doubt expected to land with some fluid left in the tank, in addition to the 10% that would have been needed to land successfully. So 10% needed to land, and 15% margin left in the tank. Sounds reasonable...

Or that the duty cycle on the fins was much higher than expected, so they used more than expected in the first three minutes, in lots of small chunks, rather than the usage being linear over the whole time period.

[Kabloona]

Or that the duty cycle on the fins was much higher than expected, so they used more than expected in the first three minutes

Yes, that's the most logical explanation. As has been remarked before, that was the problem with the Conestoga TVC failure. Unexpected noise in the guidance loop caused the TVC duty cycle to be excessive and the fluid ran out.

But my comment about margins wasn't about why the fluid ran out. I was simply comparing the 25% duration shortage with the 10% fluid shortage and saying there's not necessarily a discrepancy in those numbers, since they would have expected to have some fluid left over for margin.

But however you slice it, obviously they underestimated the duty cycle.

[sghill]

Here's a gif showing the retroburn on the monitor, for easier watching.

This is very neat but does not show the structure visible in at least two of the previous hypersonic retro-propulsion burns, including the first for CASSIOPE.
These involved quasi-periodic patterns, like repeated bars of flame, although they were only present for parts of the burns.

I don't follow you, can you elaborate please?  Are you talking about perceived flame and exhaust structures?  The first burn in the Orbcomm landing video shows flames all over the place at the end of the retro-propusion burn (which is where this snippet comes from) even though there are nice patterns at the start of the burn when it's still going supersonic.  The CRS-5 video is high contrast and shot at night, so we're going to see fewer smoke patterns and more of the flame dancing about.

As others have noted, the color here is blue, where in previous burns it was decidedly yellow.
I wonder what is different this time.

I wouldn't read anything into that.  It's a nighttime video, so we're losing colors due to the white balance.  Plus, TV's tend to shift colors to the blue end of things, and we're looking at an animated gif., made from compressed Internet video, taken with a webcam, of a tv across a room under florescent lighting, showing a video taken with (essentially) a webcam, in the dark with a dazzling, shifting light source, and streamed from a spotty data connection.  Honestly, I'm surprised we see color at all let alone the miracle of the video itself!

[llanitedave]

Plus, TV's tend to shift colors to the blue end of things, and we're looking at an animated gif., made from compressed Internet video, taken with a webcam, of a tv across a room under florescent lighting, showing a video taken with (essentially) a webcam, in the dark with a dazzling, shifting light source, and streamed from a spotty data connection. 

Yeah, but why aren't the colors accurate?   

[sghill]

Plus, TV's tend to shift colors to the blue end of things, and we're looking at an animated gif., made from compressed Internet video, taken with a webcam, of a tv across a room under florescent lighting, showing a video taken with (essentially) a webcam, in the dark with a dazzling, shifting light source, and streamed from a spotty data connection. 

Yeah, but why aren't the colors accurate?   

This is off topic, but the short answer is that each element in the above chain (plus your own eyes) is trying to simultaneously balance the color temperature without correcting for previous errors.  It's like a game of "Telephone" for cameras.  The guys in launch control likely saw much more color on the monitor with their eyes than we do now in the recordings of the recordings.

http://www.apogeephoto.com/july2004/jaltengarten7_2004.shtml

[russianhalo117]

The video posted earlier in this thread of the Brazilian YouTube CRS-5 launch video has some interesting shots of the SpaceX launch control as others have already pointed out.  I didn't see anyone else post it (forgive me if someone did), but one of the shots clearly shows end of the retro-propulsion burn with the grid fins deployed in the rocketcam shot.

You can see the video feed over the technician's shoulder up on the wall monitor on the far left, and also projected on the far wall to the right.  He turns around and notices the webcam right after the burn ends followed by a shot of lots of clapping in Hawthorne- at which point, we don't see that launch control room anymore for the remainder of the video.

 

Fast forward to 7:40 in the video (T+ 7:02 for the launch itself) to see the retro-propulsion burn.  At 8:33 in the video, everyone's attention in Hawthorne quickly jerks towards their monitors, and then we see more the engine bell and lox shots for a while with a bummed out looking-Elon leaning back in his center-front Captain's chair for the remainder of the launch.

I added it in the update thread as an update and was posted day after launch.

[macpacheco]

On the instantaneous launch window for ISS (and other) missions, couldn't SpaceX perform the whole countdown (and associated functions) ahead of time and then hold @ a few minutes prior to launch, giving itself room to handle some glitches (replacing the hold with a recycle if something go wrong and can be fixed) ? Of course they would probably have to hold at a point where propellants are still being topped off.
Of course long term we'd hope those aborts and recycles will go away (but not the topic of the question).

[yg1968]

This video is cool:

[Aerospace Dilettante]

This video is cool:

Not 100% accurate, but definitely cool!

[hpras]

This video is cool:

Second stage firing when Dragon separates worries me a little (lot).  Seem to recall they had a problem with this (kinda, residual thrust) with Falcon 1.

[Comga]

Here's a gif showing the retroburn on the monitor, for easier watching.

This is very neat but does not show the structure visible in at least two of the previous hypersonic retro-propulsion burns, including the first for CASSIOPE.
These involved quasi-periodic patterns, like repeated bars of flame, although they were only present for parts of the burns.

I don't follow you, can you elaborate please?  Are you talking about perceived flame and exhaust structures? 

Check out the videos of previous hypersonic retro burns.  For extended moments the flames arrange themselves into a series of bright and dim bars.  Presumably these are perpendicular to the line between the firing engines.  The shape is common to interference effects from three point sources.  It's really fantastic.

These structures were not seen for the full duration of the burn, so their absence for a clip as brief as this is not proof that it wasn't the retroburn.   (Did you follow all of those negatives? )  But it is curious.

There is no common mechanism that would take a yellow flame and make it bright blue.  Something else is going on in that video.  I don't think it portrays a grid fin silhouetted in front of the retroburn.

[deruch]

Here's a gif showing the retroburn on the monitor, for easier watching.

This is very neat but does not show the structure visible in at least two of the previous hypersonic retro-propulsion burns, including the first for CASSIOPE.
These involved quasi-periodic patterns, like repeated bars of flame, although they were only present for parts of the burns.

I don't follow you, can you elaborate please?  Are you talking about perceived flame and exhaust structures? 

Check out the videos of previous hypersonic retro burns.  For extended moments the flames arrange themselves into a series of bright and dim bars.  Presumably these are perpendicular to the line between the firing engines.  The shape is common to interference effects from three point sources.  It's really fantastic.

These structures were not seen for the full duration of the burn, so their absence for a clip as brief as this is not proof that it wasn't the retroburn.   (Did you follow all of those negatives? )  But it is curious.

There is no common mechanism that would take a yellow flame and make it bright blue.  Something else is going on in that video.  I don't think it portrays a grid fin silhouetted in front of the retroburn.

As for why there's no interference pattern, besides the short length of the clip that you rightly identify it's entirely possible that the 3 engines burning are rotated 90^o from what previous burns had.  In the attached drawing (sorry for poor quality), the green arrow shows where the camera is.  IMO, in previous reentry burns, the 3 engines used were those marked with red.  In this reentry, I think it might have been the blue ones.  That could also explain why there wasn't an interference pattern.

It's very clear that it is video of an extended gridfin during a burn (see attached frame; if it were any clearer, you could see the hat and microphone ).  Based on mission time, the reentry/retroburn is what fits most closely.

[ugordan]

These structures were not seen for the full duration of the burn, so their absence for a clip as brief as this is not proof that it wasn't the retroburn.   (Did you follow all of those negatives? )  But it is curious.

Precisely. If you go back to the Orbcomm video and watch carefully through the ice buildup, at the end the interference pattern is rapidly and violently disturbed, and it still does not show the end of the burn. Additionally, it's obvious the actual flame is overexposed so any pattern is likely to be drowned out in that sea of white.

There is no common mechanism that would take a yellow flame and make it bright blue.  Something else is going on in that video.  I don't think it portrays a grid fin silhouetted in front of the retroburn.

You are underestimating the difference in appearance a difference in lighting conditions and exposure duration will produce. You don't need to look any further than comparing daylight plume appearance immediately pre-MECO from CRS-3 to nighttime appearance of CRS-4. Just because some stuff (like bluish plasma glow?) is not evident in daylight video because it's faint, doesn't mean a night exposure won't bring it out, and make the normally yellow engine flame turn more of a white due to auto white balancing. Didn't you notice that in all the nighttime launches the yellow flame of the 1st stage exhaust actually looks white?

In fact, one thing that I noticed in that gif matches what I noticed in the video I posted above, the overall brightness drops suddenly about a second or so before the burn is complete. I am speculating that both these videos show the same thing - the reentry burn - and that the burn is finished with center engine burning only for increased aimpoint accuracy (due to lower g). That also would, obviously, terminate any interference pattern in that last second.

[Jdeshetler]

Check out the videos of previous hypersonic retro burns.  For extended moments the flames arrange themselves into a series of bright and dim bars.  Presumably these are perpendicular to the line between the firing engines.  The shape is common to interference effects from three point sources.  It's really fantastic.

These structures were not seen for the full duration of the burn, so their absence for a clip as brief as this is not proof that it wasn't the retroburn.   (Did you follow all of those negatives? )  But it is curious.

There is no common mechanism that would take a yellow flame and make it bright blue.  Something else is going on in that video.  I don't think it portrays a grid fin silhouetted in front of the retroburn.

As for why there's no interference pattern, besides the short length of the clip that you rightly identify it's entirely possible that the 3 engines burning are rotated 90^o from what previous burns had.  In the attached drawing (sorry for poor quality), the green arrow shows where the camera is.  IMO, in previous reentry burns, the 3 engines used were those marked with red.  In this reentry, I think it might have been the blue ones.  That could also explain why there wasn't an interference pattern.

It's very clear that it is video of an extended gridfin during a burn (see attached frame; if it were any clearer, you could see the hat and microphone ).  Based on mission time, the reentry/retroburn is what fits most closely.

Here is a stabilized video from Astronomy Live, the first 30 seconds is ascending and rest of it is decenting.

[Lars-J]

There is no common mechanism that would take a yellow flame and make it bright blue.  Something else is going on in that video.  I don't think it portrays a grid fin silhouetted in front of the retroburn.

Have you ever heard of white balance? If you've ever taken pictures or video with any kind of manual settings, you know that incorrect white balance can make a red color look blue, and vice versa. And if this is a picture of a room, where you are looking at an image on a monitor... Yeah... Don't trust the colors.

[gospacex]

Russian news site:

http://www.vesti.ru/doc.html?id=2265465

Title: "Dragon launch: fiasco of American engineers".

[ArbitraryConstant]

There is no common mechanism that would take a yellow flame and make it bright blue.  Something else is going on in that video.  I don't think it portrays a grid fin silhouetted in front of the retroburn.

Here's a picture of my stove. Option 1, I heated it two white hot. Option 2, that's what infrared looks like on a CCD that's not properly IR filtered. Which do you think is more likely?

[woods170]

Russian news site:

http://www.vesti.ru/doc.html?id=2265465

Title: "Dragon launch: fiasco of American engineers".

Typical case of ill-informed reporting and the resulting opinionating. Best to ignore it.