Author Topic: SpaceX Reusable Falcon 9 (Grasshopper ONLY) DISCUSSION Thread (3)  (Read 665319 times)

Offline cordor

  • Full Member
  • *
  • Posts: 167
  • Liked: 0
  • Likes Given: 0
Another question is throttle response speed, which I'd think would not be terribly fast for a pump-fed engine?  That would seem to have a significant impact on the control problem.  What would be reasonable for such as M1D for small and large throttle changes?  Small fractions of seconds?  Seconds?

I don't think it throttles down thrust by changing pump rpm, instead just limit the propellant flow into main combustion chamber.


Offline Lars_J

  • Senior Member
  • *****
  • Posts: 6161
  • California
  • Liked: 665
  • Likes Given: 195
Another question is throttle response speed, which I'd think would not be terribly fast for a pump-fed engine?  That would seem to have a significant impact on the control problem.  What would be reasonable for such as M1D for small and large throttle changes?  Small fractions of seconds?  Seconds?

I don't think it throttles down thrust by changing pump rpm, instead just limit the propellant flow into main combustion chamber.

And how do you think it pumps propellant? The pump...

Offline Joel

  • Full Member
  • ****
  • Posts: 533
  • Wisconsin
  • Liked: 42
  • Likes Given: 42
Should raining even be a launch constraint if the aim if for fast-paced space access? IIRC Arianes can be launched into rain from Kourou, and we all know Apollo-12.

Sunny skies are relevant if you also aim for cool youtube videos!

Offline CapitalistOppressor

  • Member
  • Posts: 82
  • Liked: 60
  • Likes Given: 3
Should raining even be a launch constraint if the aim if for fast-paced space access? IIRC Arianes can be launched into rain from Kourou, and we all know Apollo-12.

Sunny skies are relevant if you also aim for cool youtube videos!

Plus, I'm sure getting the soot out of the Space Cowboy is way more difficult when he's wet.

Offline pogo661

  • Member
  • Posts: 38
  • Liked: 14
  • Likes Given: 3
Quote
"From a control perspective, if the flight model is accurate, the only reason you'd need fast changes is to respond to external outputs."

I suspect this was sarcasm :-)

Small plane and gliders are known to drop out of the sky because the 35 knot headwind they had at 150'  at the beginning of final approach, dropped 20 knots as they got closer to the ground, leaving them below stall speed.

I think small plane pilots can compensate by adding 1/2 the estimated wind speed to their approach speed, or more if there are wind obstructions near the ground,  as a safety margin, but the grasshopper will have an additional problem in that the top and bottom of the rocket will experience different wind speeds at the same time!

As it approaches the ground with a strong wind, the wind speed gradient will make it seem like a giant hand is pushing the base of the rocket into the wind.   And it has a single engine with which to compensate for descent, yaw, and translation.   

I can maybe imagine if the landing field was perfectly flat, and the landing struts had giant casters on the ends, you could just focus on staying vertical, and controlling descent, and not worry about translation when you touched down.  But I don't remember seeing the rocket move sideways after landing in wind the last time.

If they manage to do it, that's some nice engineering.   I wonder if they have exact knowledge of what the wind gradients will be, or if they managed to do it with just position, speed and acceleration sensors.


Online meekGee

  • Senior Member
  • *****
  • Posts: 9146
  • N. California
  • Liked: 5514
  • Likes Given: 940
Quote
"From a control perspective, if the flight model is accurate, the only reason you'd need fast changes is to respond to external outputs."

I suspect this was sarcasm :-)

Small plane and gliders are known to drop out of the sky because the 35 knot headwind they had at 150'  at the beginning of final approach, dropped 20 knots as they got closer to the ground, leaving them below stall speed.


... no, not this time.

A winged vehicle is an entirely different beast.  When then wind shifts, it can lose lift because there's a change in airspeed over the wing.

Even a helicopter is less sensitive to wind, since the effective wind speed on the rotor blades is much less dependent on the wind. A helicopter can get buffeted as it hovers, but will not drop down in the way a plane does.

A landing F9R is even less sensitive than a helicopter, since the landing profile is very fast. Unlike a helicopter, it doesn't hover before it lands.  (Actually, I wouldn't be surprised if the correct way to land a helicopter in varying wind is to hover a safe distance above ground and then kill the lift with the collective)

So yes, i think it is safe to say the amount of control input required is low, very much unlike an airplane.
ABCD - Always Be Counting Down

Offline AS-503

  • Full Member
  • ****
  • Posts: 431
  • Orion Fab Team
  • Colorado USA
  • Liked: 163
  • Likes Given: 116
Should raining even be a launch constraint if the aim if for fast-paced space access? IIRC Arianes can be launched into rain from Kourou, and we all know Apollo-12.

Yes, it's that trouble with the tribbles...er, I mean trouble with the triboelectrification. :D

Offline Joel

  • Full Member
  • ****
  • Posts: 533
  • Wisconsin
  • Liked: 42
  • Likes Given: 42
Instead of comparing the Grasshopper flight control with aircrafts or helicopters (an utterly useless exercise IMO), why not compare it with something more similar? Does anyone know how the DC-X handled wind disturbances for example? Or more novel control schemes based on the DC-X design?

I came across this paper for ACC2008, which seems relevant to the control design of Grasshopper:

http://ieeexplore.ieee.org/xpl/login.jsp?tp=&arnumber=4586630&url=http%3A%2F%2Fieeexplore.ieee.org%2Fxpls%2Fabs_all.jsp%3Farnumber%3D4586630
« Last Edit: 02/20/2013 09:41 am by Joel »

Offline R7

  • Propulsophile
  • Senior Member
  • *****
  • Posts: 2738
    • Don't worry.. we can still be fans of OSC and SNC
  • Liked: 957
  • Likes Given: 662
Yes, it's that trouble with the tribbles...er, I mean trouble with the triboelectrification. :D

Darn trilobytes, thought they were extinct!  ;D But interesting point, learned something new today, thanks! Apparently that can be mitigated since the Right Stuff and the French have done wet launches anyway. Beef up the Signal Conditioning Equipment! ;)

Just wanted to raise the question if more rapid, sustained space access is the goal then launch constraints should be reconsidered too, no? Might be dry in Texas most of the time but not so in Florida.

AD·ASTRA·ASTRORVM·GRATIA

Online meekGee

  • Senior Member
  • *****
  • Posts: 9146
  • N. California
  • Liked: 5514
  • Likes Given: 940
Instead of comparing the Grasshopper flight control with aircrafts or helicopters (an utterly useless exercise IMO), why not compare it with something more similar? Does anyone know how the DC-X handled wind disturbances for example? Or more novel control schemes based on the DC-X design?

I came across this paper for ACC2008, which seems relevant to the control design of Grasshopper:

http://ieeexplore.ieee.org/xpl/login.jsp?tp=&arnumber=4586630&url=http%3A%2F%2Fieeexplore.ieee.org%2Fxpls%2Fabs_all.jsp%3Farnumber%3D4586630

Well, I dis-compared them, since pogo661 was wondering.  But it's an important distinction that wind speed does not affect lift.

If you want to think about it some more, consider how the instantaneous gimbal response can counter either the side force exerted by the wind, or the tipping torque - but not both!

The complete response to correct for both is therefore more complicated, more like what a segway does.

From what I figured out, this defines what "terminal approach" is, since at some altitude it becomes impossible to correct for both drift and tilt due to time constants of the rocket, and so from that point onwards the system can correct only for one of them.

But if you assume that terminal approach is in the order of 3 seconds (50 m at 1 g) then the resultant drift or tilt are really small.

We'll see a lot more on the next GH flights  (or a crater, as Elon pointed out)
ABCD - Always Be Counting Down

Offline R7

  • Propulsophile
  • Senior Member
  • *****
  • Posts: 2738
    • Don't worry.. we can still be fans of OSC and SNC
  • Liked: 957
  • Likes Given: 662
consider how the instantaneous gimbal response can counter either the side force exerted by the wind, or the tipping torque - but not both!

Remove the adverse tipping torque and make it work for you, deploy legs with fins to move CP below CG.
AD·ASTRA·ASTRORVM·GRATIA

Offline hrissan

  • Full Member
  • ***
  • Posts: 397
  • Novosibirsk, Russia
  • Liked: 313
  • Likes Given: 1986
consider how the instantaneous gimbal response can counter either the side force exerted by the wind, or the tipping torque - but not both!

Remove the adverse tipping torque and make it work for you, deploy legs with fins to move CP below CG.
With legs length approaching the length of the body, the whole object will resemble grasshopper much better. :)

To be serious, I remember the heated discussion here "are the thrusters at the top of the GH required for canceling wind forces acting on the large and lightweight tank during and after last seconds of landing".

It seems SpaceX thinks that gimballng just single engine is enough. The second flight was impressive - the wind (though small) did not prevent computer to land exactly at the point where it took off, without "fins". Anyway, very large fins could be traded for small set of thrusters?

I do not know what flies first - the next Falcon 9, or GH, hope they both do it soon and without craters...

Online meekGee

  • Senior Member
  • *****
  • Posts: 9146
  • N. California
  • Liked: 5514
  • Likes Given: 940
consider how the instantaneous gimbal response can counter either the side force exerted by the wind, or the tipping torque - but not both!

Remove the adverse tipping torque and make it work for you, deploy legs with fins to move CP below CG.

A flying vehicle that has to cover such a huge range of airspeeds is an exercise in contradictions. If you add aerodynamic surfaces that are effective in 10 m/s wind, they will be the death of you at supersonic speeds.

Since the low velocity stuff can be handled by the rocket, I wouldn't touch any aerodynamic solutions for that phase. Just brute force vertical landing relying solely on the motor.

This is something they can simulate really accurately btw.  I'm sure that in simuland GH has already landed successfully under changing wind conditions
ABCD - Always Be Counting Down

Offline R7

  • Propulsophile
  • Senior Member
  • *****
  • Posts: 2738
    • Don't worry.. we can still be fans of OSC and SNC
  • Liked: 957
  • Likes Given: 662
If you add aerodynamic surfaces that are effective in 10 m/s wind, they will be the death of you at supersonic speeds.

It's going to fly at supersonic speeds with legs folded/stowed, no?
AD·ASTRA·ASTRORVM·GRATIA

Online meekGee

  • Senior Member
  • *****
  • Posts: 9146
  • N. California
  • Liked: 5514
  • Likes Given: 940
If you add aerodynamic surfaces that are effective in 10 m/s wind, they will be the death of you at supersonic speeds.

It's going to fly at supersonic speeds with legs folded/stowed, no?

Yes, that's true.  So perhaps having intentionally large cross drag at the bottom will help somewhat, but a) it can't eliminate the issue, and b) the issue is not really a problem.

ABCD - Always Be Counting Down

Offline mlindner

  • Software Engineer
  • Senior Member
  • *****
  • Posts: 2146
  • Space Capitalist
  • Silicon Valley, CA -- previously in Ann Arbor, MI
  • Liked: 833
  • Likes Given: 339
Watch the discussion people, it's starting to trend in to F9R rather than grasshopper. This thread is not for speculation of what F9R will do, but what grasshopper is (and possible second versions will be) capable of.
Internal combustion engine in space. It's just a Bad Idea.TM - Robotbeat

Offline QuantumG

  • Senior Member
  • *****
  • Posts: 8780
  • Australia
  • Liked: 3785
  • Likes Given: 892
Anyone care to hazard a guess at the total delta-v capability of Grasshopper?

5 km/s?
Human spaceflight is basically just LARPing now.

Online Robotbeat

  • Senior Member
  • *****
  • Posts: 28744
  • Minnesota
  • Liked: 8843
  • Likes Given: 5740
Anyone care to hazard a guess at the total delta-v capability of Grasshopper?

5 km/s?

Not even close. It can't take off with anywhere near a full tank, and it's got a huge hunk of metal for legs (mild steel?). I guesstimate (based on figures from spacelaunchreport) about one-tenth that, 500m/s. Maybe up to 1400m/s depending on the actual dry mass of Grasshopper and what T/W ratio they actually need. But it'd have a lot of aerolosses because of the huge structure at the bottom.
Chris  Whoever loves correction loves knowledge, but he who hates reproof is stupid.

To the maximum extent practicable, the Federal Government shall plan missions to accommodate the space transportation services capabilities of United States commercial providers. US law http://goo.gl/YZYNt0

Offline QuantumG

  • Senior Member
  • *****
  • Posts: 8780
  • Australia
  • Liked: 3785
  • Likes Given: 892
Not even close. It can't take off with anywhere near a full tank, and it's got a huge hunk of metal for legs (mild steel?). I guesstimate (based on figures from spacelaunchreport) about one-tenth that, 500m/s. Maybe up to 1400m/s depending on the actual dry mass of Grasshopper and what T/W ratio they actually need. But it'd have a lot of aerolosses because of the huge structure at the bottom.

Merlin 1D is something like 650 kN thrust, 450 kg mass.. say 855 kg with the thrust assembly. Sea level isp of 282s. Vacuum isp of 311s.

Best gross lift off weight with that much thrust is about 60,000 kg. As you say, no-where near a full tank.

Falcon 9 v1.0 tank is about 7,500 kg.

How much for the legs? 2,000 kg? 3,000 kg?

At the outside, the dry mass of Grasshopper is no more than 12,000 kg.

dv = g * 282 * Math.log(60000 / 12000)

That's still 4,450 m/s.
Human spaceflight is basically just LARPing now.

Online Robotbeat

  • Senior Member
  • *****
  • Posts: 28744
  • Minnesota
  • Liked: 8843
  • Likes Given: 5740
spacelaunchreport lists about 28mT for the dry mass of Falcon 9 v1.1. I think the legs are much more than just a couple tons (I seriously think they add ten tons or more). And you want a T/W of 1.2 or so for stability purposes.

Plus, don't forget the cowboy.
Chris  Whoever loves correction loves knowledge, but he who hates reproof is stupid.

To the maximum extent practicable, the Federal Government shall plan missions to accommodate the space transportation services capabilities of United States commercial providers. US law http://goo.gl/YZYNt0

Tags: