Author Topic: SpaceX Falcon Mission Simulations  (Read 81461 times)

Offline john smith 19

  • Senior Member
  • *****
  • Posts: 10350
  • Everyplaceelse
  • Liked: 2430
  • Likes Given: 13605
Re: SpaceX Falcon Mission Simulations
« Reply #100 on: 09/18/2017 04:45 pm »
What are you using as the simulator?

The program is called SpaceSim. It is written in C# and you can download it and the source code here: https://github.com/zlynn1990/SpaceSim
Noted.
MCT ITS BFR SS. The worlds first Methane fueled FFSC engined CFRP SS structure A380 sized aerospaceplane tail sitter capable of Earth & Mars atmospheric flight.First flight to Mars by end of 2022 TBC. T&C apply. Trust nothing. Run your own #s "Extraordinary claims require extraordinary proof" R. Simberg."Competitve" means cheaper ¬cheap SCramjet proposed 1956. First +ve thrust 2004. US R&D spend to date > $10Bn. #deployed designs. Zero.

Offline IainMcClatchie

  • Full Member
  • ***
  • Posts: 394
  • San Francisco Bay Area
  • Liked: 279
  • Likes Given: 411
Re: SpaceX Falcon Mission Simulations
« Reply #101 on: 09/18/2017 08:15 pm »
That's some nice work.

I don't mean to look a gift horse in the mouth, but I'll note that all the zooming in and out seems less than ideal.  You might have two views simultaneously, one close up showing the spacecraft attitude, and one far away showing the trajectory.

Offline deruch

  • Senior Member
  • *****
  • Posts: 2422
  • California
  • Liked: 2006
  • Likes Given: 5634
Re: SpaceX Falcon Mission Simulations
« Reply #102 on: 09/22/2017 03:46 am »
<snip>
 But what if the Falcon S2 could be modified (as others have suggested) to have comparable dynamics to the BFS, and still support existing payloads? Here is a speculative 'Frankenstein' approach:

3. Three strakes of a similar form to the BFS strakes, but containing extra nitrogen tanks and thrusters. The strakes could also have the BFS split body flaps mentioned (but not so far visualised) for additional control authority.


Not sure about the rest, but this would invalidate the launch for USAF certification as it would entail a change to "major airframe structure" of the launch vehicle which is one of the things that would require a new vehicle certification.  It's cool as an idea that could potentially achieve 2nd stage reuse, but it certainly won't be the way they attempt on the FH demo mission.  Beyond that, something like it could be done if they are willing to produce multiple vehicle variants, i.e. one upper stage for FH and one for F9.
« Last Edit: 09/22/2017 03:48 am by deruch »
Shouldn't reality posts be in "Advanced concepts"?  --Nomadd

Offline OneSpeed

  • Full Member
  • ****
  • Posts: 1587
  • Liked: 4928
  • Likes Given: 2077
Re: SpaceX Falcon Mission Simulations
« Reply #103 on: 10/28/2017 03:58 am »
With the Falcon Heavy Demo Mission approaching, I thought a 'compari-sim' between CRS-11 and a Falcon Heavy launch to LEO might be of interest.


Offline OneSpeed

  • Full Member
  • ****
  • Posts: 1587
  • Liked: 4928
  • Likes Given: 2077
Re: SpaceX Falcon Mission Simulations
« Reply #104 on: 12/11/2017 10:37 am »
For better or for worse, the Falcon Heavy demo mission payload will be a cherry red Tesla Roadster. I'm still assuming the ASDS will be only 342kms downrange from LC-39A. If so, this demands a very lofted profile, something like Formosat-5, which used a single S2 burn to deploy a very light payload.

Launch opportunities for such a profile are daily for most of the next year, but assuming a launch on Feb 1st 2018, the sim predicts that the two side boosters would stage at 2050 m/s, and perform a 30 second horizontal boostback burn on the way to an apogee of 285 kms. The core booster would stage at 3040 m/s, with a ballistic trajectory that unchecked would have a 485 km apogee and land about 900 kms downrange. In order for it to reduce the downrange to 342 kms, it would need to flip to retrograde for its boostback burn, which could be made more efficient if it utilised 5 rather than the usual 3 Merlin engines.

The S2 burn would commence at an altitude of 190 kms, and SECO would be at 1200 kms. The TMI ΔV requirement at this altitude would be significantly less than that from sea level, so the shutdown velocity would be 10.6 km/s. The transit to aphelion would take 238 days.


Offline OneSpeed

  • Full Member
  • ****
  • Posts: 1587
  • Liked: 4928
  • Likes Given: 2077
Re: SpaceX Falcon Mission Simulations
« Reply #105 on: 04/01/2018 12:52 am »
Here is the latest simulation I've been working on in collaboration with NSF member Zach Lynn. This simulation shows the Falcon Heavy demo flight that took place earlier last month. This is an update from the previous video posted, as the flight profile now matches the webcast telemetry very closely. Additionally, this video shows the predicted Earth flyby in 2091. Simulating the n-body system out this many years into the future requires a more course time-step. Therefore, the exact flyby date and distances predicted by this simulation are just 'best effort'.

Special thanks to NSF member Jay DeShetler for the hi-fidelity launch audio recorded live near the LC-39A HIF.



Offline OneSpeed

  • Full Member
  • ****
  • Posts: 1587
  • Liked: 4928
  • Likes Given: 2077
Re: SpaceX Falcon Mission Simulations
« Reply #106 on: 05/14/2018 12:51 pm »
Here is a 'comparisim' between Bangabandhu-1 and BulgariaSat-1. My recent sims pointed to block 4 being 91% of block 5 thrust, but the Falcon Heavy S1 cores ran at 92% thrust. From this I've concluded the SpaceSim dry mass estimates were too low, and for Bangabandhu I've added 1% of GLOW (about 5.5 mT) to the S1 and S2 dry masses. Even with this additional mass, I've only had to run Bangabandhu at 97% thrust at lift-off, gradually throttling to 90% before reaching 4Gs of acceleration just before MECO. Perhaps SpaceX were conservative on the first block 5 flight, and more performance will be demonstrated in the near future?

Points of interest:

1. Block 5 initial acceleration was 1.43Gs, only slightly less than the Falcon Heavy demo at 1.47Gs.
2. The Bangabandhu-1 S2 ran at a constant 96% of block 4 thrust up to about 4Gs. Again, perhaps there is more performance to come.
3. Bangabandhu-1 was not sent to a super-synchronous GTO, whilst BulgariaSat-1, a slightly heavier satellite was. Is that down to differing customer requirements?

« Last Edit: 05/14/2018 12:52 pm by OneSpeed »

Offline deruch

  • Senior Member
  • *****
  • Posts: 2422
  • California
  • Liked: 2006
  • Likes Given: 5634
Re: SpaceX Falcon Mission Simulations
« Reply #107 on: 05/14/2018 03:24 pm »
3. Bangabandhu-1 was not sent to a super-synchronous GTO, whilst BulgariaSat-1, a slightly heavier satellite was. Is that down to differing customer requirements?

Bangabandhu-1's orbit inclination was ~4.5o less than BulgariaSat-1.  Not sure why they wanted it, but I'm sure it was a customer choice.
Shouldn't reality posts be in "Advanced concepts"?  --Nomadd

Offline leetdan

  • Full Member
  • ***
  • Posts: 370
  • Space Coast
  • Liked: 323
  • Likes Given: 282
Re: SpaceX Falcon Mission Simulations
« Reply #108 on: 05/14/2018 04:08 pm »
These are awesome!  Have you ever considered exporting some of these to KMZ files?  This would be useful to the general public for visualizing each path, as well as planning launch viewing and photography.

Offline OneSpeed

  • Full Member
  • ****
  • Posts: 1587
  • Liked: 4928
  • Likes Given: 2077
Re: SpaceX Falcon Mission Simulations
« Reply #109 on: 06/11/2018 12:57 pm »
Here is a 'comparisim' between Iridium NEXT 4 and Iridium / Grace FO. Iridium NEXT 4 was legless, whilst Iridium / Grace FO had a lighter payload, giving similar performance off the pad. However, Iridium Grace needed to deliver its payloads to two different orbits with a minimum of second stage restarts. By my count, three is as many as have been demonstrated (someone else may have kept better track of this).

The first Iridium NEXT 4 S2 burn placed it in a 182 x 625km orbit, and only required a circularisation burn at apogee before deployment of the 10 Iridium satellites to 625 x 625kms. They subsequently raised themselves to their 780 x 780km operational orbits.

The first Iridium GRACE S2 burn placed it directly into a 505 x 483km orbit, ready to deploy the Grace FO satellites. This required a very lofted profile, with the last few minutes of the S2 burn at a relatively high angle of attack, and hence with significant cosine losses. The losses were minimised by throttling S2 back near 60% throttle over those last few minutes of its first burn, and so extending its duration to over the T+10 minute mark of the flight. By extending the duration, the AoA required to reach the required orbit was slightly reduced, minimising losses. The second S2 burn placed Iridium GRACE in a 482 x 720km orbit, from which the 5 Iridium satellites will have had to raise and circularise themselves.



Offline joncz

  • Veteran
  • Full Member
  • ****
  • Posts: 523
  • Atlanta, Georgia
  • Liked: 296
  • Likes Given: 390
Re: SpaceX Falcon Mission Simulations
« Reply #110 on: 01/12/2019 01:02 pm »
« Last Edit: 01/12/2019 02:12 pm by gongora »

Offline OneSpeed

  • Full Member
  • ****
  • Posts: 1587
  • Liked: 4928
  • Likes Given: 2077
Re: SpaceX Falcon Mission Simulations
« Reply #111 on: 01/20/2019 07:09 am »
I'd assumed Iridium 7 and 8 would have very similar mission profiles, but it looks like there has been a subtle change of approach beginning with the GPS-III mission. For that mission some extra TPS was added to the tip of the fairing halves, and from the clearest shot of the Iridium-8 fairing I can find, the same shaped TPS was added to that as well, although painted white (credit Michael Baylor).

From the three SpaceX webcasts, I've tabulated the velocity and acceleration at one second intervals. For Iridium-7 you can see a traditional 'in the bucket' throttle back, whist for GPS-III and Iridium-8, there is a smaller throttle back, followed by a slow and continuous increase in acceleration towards the end of the S1 burn (although the expendable GPS-III does limit g forces to about 3.5). In other words, one side of the bucket has disappeared.

I've then created simulations of the Iridium-7 and 8 missions in order to infer the throttle settings. From the sims, Iridium-7 throttles back to 78% for Max-Q, beginning at the 46 second mark, and finishing at 68.
For GPS-III, there is a smaller throttle back to 81% at 43 seconds, and then a gradual throttle up to 93%, followed by a gradual throttle down to 66% to limit gs. For Iridium-8 the throttle back is even smaller at 83%, and then gradually up to 92% at the end of the burn, which is three seconds longer than for Iridium-7.

Points of interest:
1. As a consequence of the smaller throttle backs, the peak dynamic stress on the vehicle has increased from 23.7kPa for Iridium-7 to 27.7kPa for Iridium-8. I'm wondering if the TPS added to the fairing makes any structural difference, or are there some other concurrent changes to the fairing structure?

2. Because Iridium-7s S1 burn is three seconds shorter, with higher overall throttle, its velocity is generally faster and at a lower altitude, so heating flux should be slightly greater. This is at odds with the extra TPS on GPS-III and Iridium-8, so I'm not sure what's going on there.

3. From the new Falcon users guide, the block 5 M1D has increased thrust by 1% to 854kN, and the M1DVac by a substantial 5% to 981kN. Using these figures, I was able to run the second stage at 99% thrust and closely match the mission profiles.

4. Also from the new guide, the M1D is throttlable to 57%, and the M1DVac to 64%, so I've applied those limits in the sims.

5. Despite the differences in throttle profile, at the 150 second mark both Iridium-7 and 8 are at 67.2 kms altitude, and around 1900m/s. Because Iridium-8 S1 has burnt three seconds longer, the stress on the payload should be slightly less. Also, Iridium-8 has an 11 second longer S2 burn, with a much longer terminal guidance phase, that limits gs to below 4 rather than the 5 for Iridium-7.


Offline Raul

  • Full Member
  • ***
  • Posts: 352
  • Ústí nad Orlicí, CZECH
  • Liked: 1188
  • Likes Given: 99
Re: SpaceX Falcon Mission Simulations
« Reply #112 on: 01/21/2019 07:14 pm »
2. Because Iridium-7s S1 burn is three seconds shorter, with higher overall throttle, its velocity is generally faster and at a lower altitude, so heating flux should be slightly greater. This is at odds with the extra TPS on GPS-III and Iridium-8, so I'm not sure what's going on there.
I don't think that Iridium-8 used extra TPS. Note that same shape on the fairing tip is typical for any Fairing 1.0., which was used for Iridium-8 mission.
Some other found Fairing 1.0. examples with this detail .. FHDemo, Intelsat35e, or SES9.

In contrast with previous Iridium missions (Iridium-6 and 7), where have been used Fairing 2.0., with circular shape on the tip.
The only exception is GPS-III mission, where Fairing 2.0. had black extra TPS.

Offline RotoSequence

  • Senior Member
  • *****
  • Posts: 2208
  • Liked: 2068
  • Likes Given: 1535
Re: SpaceX Falcon Mission Simulations
« Reply #113 on: 01/21/2019 11:14 pm »
2. Because Iridium-7s S1 burn is three seconds shorter, with higher overall throttle, its velocity is generally faster and at a lower altitude, so heating flux should be slightly greater. This is at odds with the extra TPS on GPS-III and Iridium-8, so I'm not sure what's going on there.
I don't think that Iridium-8 used extra TPS. Note that same shape on the fairing tip is typical for any Fairing 1.0., which was used for Iridium-8 mission.
Some other found Fairing 1.0. examples with this detail .. FHDemo, Intelsat35e, or SES9.

In contrast with previous Iridium missions (Iridium-6 and 7), where have been used Fairing 2.0., with circular shape on the tip.
The only exception is GPS-III mission, where Fairing 2.0. had black extra TPS.

Iridium 8 definitely has a cap piece on top of the fairing in the same place as the GPS-III mission, albeit in white instead of black. Different bolt-on fairing nose parts for different ascent profiles?

Offline Raul

  • Full Member
  • ***
  • Posts: 352
  • Ústí nad Orlicí, CZECH
  • Liked: 1188
  • Likes Given: 99
Re: SpaceX Falcon Mission Simulations
« Reply #114 on: 01/22/2019 04:55 am »
Iridium 8 definitely has a cap piece on top of the fairing in the same place as the GPS-III mission, albeit in white instead of black. Different bolt-on fairing nose parts for different ascent profiles?
And many other missions used Fairing 1.0. definitely has identical piece on tip. Still identical for various ascent profiles.

Offline OneSpeed

  • Full Member
  • ****
  • Posts: 1587
  • Liked: 4928
  • Likes Given: 2077
Re: SpaceX Falcon Mission Simulations
« Reply #115 on: 01/22/2019 09:27 am »
I don't think that Iridium-8 used extra TPS.

I think Raul is correct. I've redone the S1 plots including heating rate, and GPS-III, which flew relatively low and fast, has the highest thermal flux.

Offline OneSpeed

  • Full Member
  • ****
  • Posts: 1587
  • Liked: 4928
  • Likes Given: 2077
Re: SpaceX Falcon Mission Simulations
« Reply #116 on: 02/07/2019 11:32 am »
I've been meaning to post a comparison between the Rocket Lab Electron and Falcon 9 for quite a while, but SpaceX's stunning development pace has kept interrupting my plans. I wanted both simulations to have similar destination orbits, so I've picked two polar launches to LEO: ELaNa-19 and Iridium-8.

We've perhaps gotten used to the luxury of real time velocity and altitude feeds in the SpaceX launches, but the Rocket Lab live data is relatively sparse. The events I've managed to garner from the ELaNa-19 webcast are:

T-0:02   Ignition
T+0:00   Liftoff
T+1:13   Max-Q
T+2:31   MECO
T+2:35   S2 Staging
T+2:38   SESU
T+3:06   Fairing deploy
T+3:35   150kms altitude
T+4:54   3km/s velocity
T+6:27   4km/s velocity
T+7:07   Battery hotswap
T+7:25   5km/s velocity
T+8:55   Orbital
T+8:59   SECO
T+51:16   TECO

The first few times I tried to simulate this, I had trouble getting the kick stage to a feasible transfer orbit. Firstly, I hadn't included the battery hotswap. I didn't realise how important this is, from the sim behaviour it actually appears to mean the difference between making orbit and not. Secondly, I was having trouble with the transfer orbit insertion. I'd been assuming an AoA profile similar to the Falcon 9 S2, but when I went back and had a closer look at the video, I realised it was quite different. The Electron S2 uses a lot of initial AoA, and has a much more lofted trajectory, with an apogee around 220kms approaching the T+6 minute mark. Once I introduced the higher loft, the transfer orbit apogee closely matched the kick stage burn time.

Peter Beck recently explained some of the problems creating a rocket that is around 1/3 the scale of Falcon 9. As he pointed out, a pressure transducer on Falcon 9 is a particular size, and a pressure transducer on Electron is about the same size. Overall though, I'm impressed by the similarities of the two mission profiles are rather than the differences. It will be fascinating to see how much of a commercial success Electron becomes.


Offline OneSpeed

  • Full Member
  • ****
  • Posts: 1587
  • Liked: 4928
  • Likes Given: 2077
Re: SpaceX Falcon Mission Simulations
« Reply #117 on: 03/04/2019 11:55 am »
It was great to see the updated screen format for the recent DM-1 launch webcast. The rotating arc timeline makes excellent use of the screen real estate. Unfortunately, this means my recent efforts to automate the extraction of the flight data have been foiled (at least temporarily), and it's back to old school manually reading the values frame by frame for this mission.

Anyway, here's a comparisim between CRS-16 and DM-1. A major differentiator between the two missions is S1 RTLS for CRS-16 vs ASDS for DM-1. The ΔV gained is partly spent lofting the extra 3 odd tonnes of capsule mass to orbit, but there is some extra margin available, so what has it been spent on?

For this mission, the bucket is back, big time. The sim predicts that although CRS-16 throttled to 74% for Max-Q, DM-1 throttled back to 67%. That is, DM-1 went from 1.75g back to 1.2g at the bottom of the bucket. What this achieved of course, is reduced dynamic pressure on the vehicle, and for the Crew Dragon demo mission, this might have been extremely motivating. The sim suggests that Max-Q was reduced from around 24kPa to 21kPa for DM-1.

I was wondering if the extra loft of DM-1 was intended to reduce peak heating, but the sim suggests that peak flux was about 8.9kW/m2 for both mission profiles, so I'm not sure what requirement the loft fulfilled.

« Last Edit: 03/04/2019 12:41 pm by OneSpeed »

Offline IainMcClatchie

  • Full Member
  • ***
  • Posts: 394
  • San Francisco Bay Area
  • Liked: 279
  • Likes Given: 411
Re: SpaceX Falcon Mission Simulations
« Reply #118 on: 03/05/2019 12:11 am »
They had no crew and just a few hundred kg of cargo.  There should have been literally tonnes of extra margin.  Was it *all* used to reduce MaxQ?

Offline Pelorat

  • Member
  • Posts: 27
  • Liked: 19
  • Likes Given: 3
Re: SpaceX Falcon Mission Simulations
« Reply #119 on: 03/05/2019 11:31 am »
They had no crew and just a few hundred kg of cargo.  There should have been literally tonnes of extra margin.  Was it *all* used to reduce MaxQ?

I believe they had crew mass-simulators in every seat. Have a look at this image:


 

Advertisement NovaTech
Advertisement Northrop Grumman
Advertisement
Advertisement Margaritaville Beach Resort South Padre Island
Advertisement Brady Kenniston
Advertisement NextSpaceflight
Advertisement Nathan Barker Photography
0