[snip]I understand the efficiency issue in terms of using as much "free" drag delta-v as possible but...[snip]
but what I don't know is what would the effect (on ISP) be of short (80% to 100% thrust) bursts?
Quote from: meadows.st on 05/26/2014 01:37 am[snip]I understand the efficiency issue in terms of using as much "free" drag delta-v as possible but...[snip]The real efficiency issue is not getting as much "free drag delta-v as possible", but in avoiding gravity losses.Reduction to the absurd: If the capsule hits zero velocity at some elevated altitude, then it has to do an additional burn to stop after it starts falling again. That's a gravity loss.Every additional second the capsule spends off the ground is one second times one g of acceleration needed from the engines. So the most fuel efficient burn is the shortest, which translates to high acceleration, low altitude. It's just not the most comfortable, reassuring, or most flexible if something goes wrong. The goal is to get all of those things without greatly increasing the amount of fuel needed.<snip>
This is my uneducated guess:Do a test fire at an altitude where it is still safe to deploy parachutes if the engines fail. Probably two short bursts of the two largely independend sets of four engines. Decide if it is ok to go for powered landing if one of the two sets does not perform or both sets have to work for redundancy at that point. If necessary go for (thrust assisted) parachute landing.If the engines are ok at that point it is very very likely they will fire successfully again a few seconds later for the landing burn.
Your table has terminal velocity of just over 70 m/s.
What if the trunk remains attached for these landings? Weight maybe as much as doubled, terminal velocity higher, burn times longer due to both extra mass and higher initial velocities. This might get us up toward those 5 second burns in the FAA doc.
I agree that a permanently attached trunk is unlikely but for a different reason. Dragon is passively stable during re-entry, meaning that its CG is located low enough in the structure that it will always passively present its heatshield first at the first sign of atmosphere resistance. That dynamic completely flips if a now-empty trunk is still attached at re-entry. With the CG now substantially higher in the structure, Dragon will passively flip around and re-enter nose first, providing a bad day for everyone aboard.
Quote from: sublimemarsupial on 05/21/2014 05:01 pmQuote from: Lars_J on 05/21/2014 04:49 pmQuote from: JBF on 05/21/2014 04:42 pmQuote from: Jarnis on 05/21/2014 04:33 pmFreefall, 5 sec burn at the last moment, soft touchdown.Whoever will ultimately ride aboard that kind of flight profile (obviously not during these tests) is a brave man Can anyone model the G-Force curve on that? It sounds like a rough ride.If anyone can guess the terminal velocity of the capsule, it should be easy to calculate. If the G-load is evenly spread over 5 seconds, it might not be too bad.For a very first order estimate I get a terminal velocity of ~150 m/s given a total mass of 17,000 lbs (dry mass plus 3000 lbs of fuel) and a drag coefficient of 0.8.A 5 second burn to brake from 150 m/s would result in an average of ~3Gs of deceleration.
Quote from: Lars_J on 05/21/2014 04:49 pmQuote from: JBF on 05/21/2014 04:42 pmQuote from: Jarnis on 05/21/2014 04:33 pmFreefall, 5 sec burn at the last moment, soft touchdown.Whoever will ultimately ride aboard that kind of flight profile (obviously not during these tests) is a brave man Can anyone model the G-Force curve on that? It sounds like a rough ride.If anyone can guess the terminal velocity of the capsule, it should be easy to calculate. If the G-load is evenly spread over 5 seconds, it might not be too bad.For a very first order estimate I get a terminal velocity of ~150 m/s given a total mass of 17,000 lbs (dry mass plus 3000 lbs of fuel) and a drag coefficient of 0.8.
Quote from: JBF on 05/21/2014 04:42 pmQuote from: Jarnis on 05/21/2014 04:33 pmFreefall, 5 sec burn at the last moment, soft touchdown.Whoever will ultimately ride aboard that kind of flight profile (obviously not during these tests) is a brave man Can anyone model the G-Force curve on that? It sounds like a rough ride.If anyone can guess the terminal velocity of the capsule, it should be easy to calculate. If the G-load is evenly spread over 5 seconds, it might not be too bad.
Quote from: Jarnis on 05/21/2014 04:33 pmFreefall, 5 sec burn at the last moment, soft touchdown.Whoever will ultimately ride aboard that kind of flight profile (obviously not during these tests) is a brave man Can anyone model the G-Force curve on that? It sounds like a rough ride.
Freefall, 5 sec burn at the last moment, soft touchdown.Whoever will ultimately ride aboard that kind of flight profile (obviously not during these tests) is a brave man
Quote from: clongton on 05/26/2014 11:02 amI agree that a permanently attached trunk is unlikely but for a different reason. Dragon is passively stable during re-entry, meaning that its CG is located low enough in the structure that it will always passively present its heatshield first at the first sign of atmosphere resistance. That dynamic completely flips if a now-empty trunk is still attached at re-entry. With the CG now substantially higher in the structure, Dragon will passively flip around and re-enter nose first, providing a bad day for everyone aboard.Good point. Pretty much kills the idea.Thanks.
....Obviously I'm not talking about a slight modification of current trunk. This would be a new phenotype.
Quote from: AncientU on 05/26/2014 12:24 pmQuote from: clongton on 05/26/2014 11:02 amI agree that a permanently attached trunk is unlikely but for a different reason. Dragon is passively stable during re-entry, meaning that its CG is located low enough in the structure that it will always passively present its heatshield first at the first sign of atmosphere resistance. That dynamic completely flips if a now-empty trunk is still attached at re-entry. With the CG now substantially higher in the structure, Dragon will passively flip around and re-enter nose first, providing a bad day for everyone aboard.Good point. Pretty much kills the idea.Thanks.I'm not sure it's been fully dismissed.1) CG issue is only what it is designed to be. Why assume the trunk would be empty? Perhaps the more dense cargo would preferentially be stowed there (potentially removes center of grav issue).
Quote from: go4mars on 05/26/2014 02:19 pm....Obviously I'm not talking about a slight modification of current trunk. This would be a new phenotype.Indeed. This is complete speculation with little evidence connecting it to what we know at the moment about the DragonFly tests.
The FAA license application(impact statement) talked about landing with trunk -- that's what got me thinking along this path.
The DragonFly RLV is the Dragon capsule with an integrated trunk (which may or may not be attached during a DragonFly operation) and up to four steel landing legs.
I ran the simulation again and made a new chart. This attachment includes graphs and data for 0% and 20-100% thrust profiles. I am not running a cosine loss for 10 degrees since it would only be a 1.5% difference which is less than my current error rate.If I can locate my dragon reentry simulation then I will try to run a more realistic simulation with horizontal velocity to cancel out.