After watching the amazing footage of the Apollo 13 capsule parachuting down on the ocean, I was intrigued by how NASA predicted when and where it would splashdown on the vast ocean's surface with amazing precision. The time and location had to be predicted far in advance, considering they had to account for the travel time of the slow moving recovery vessel. This is especially impressive considering the capsule was in free-fall without any guidance system, leaving its eventual landing location at the mercy of the elements, such as wind and turbulence.
-
1$\begingroup$ Do you want to know specifically about Apollo 13, or about the Apollo missions generally? The splashdown locations were planned in advance and guidance was used to aim the re-entry at those locations. $\endgroup$– Erin AnneCommented Sep 11 at 22:45
-
12$\begingroup$ " capsule was in free-fall without any guidance system," Nope, It had a guidance system. See ENTRY TRAJECTORY-CONTROL MODES in ntrs.nasa.gov/citations/19720013191 also space.stackexchange.com/a/49555/6944 $\endgroup$– Organic MarbleCommented Sep 11 at 22:50
-
5$\begingroup$ It didn't start at NASA, it came from the need to send nuclear weapons to the USSR on an ICBM. It had to work well because it was an demonstration to the Soviets of how good US missile guidance technology was. $\endgroup$– user71659Commented Sep 12 at 19:26
3 Answers
At high altitude, the effects of the air on the falling capsule are minimal. The main parachutes don't deploy until around 2 miles altitude, and the capsule is still coming down at about 22 miles per hour at splashdown, so it's easy to see that even in very high winds it couldn't be blown off course by more than a small number of miles.
The spacecraft doesn't maneuver much on the return trip from the moon; once its final midcourse correction is made (two days before splashdown, for Apollo 11) the trajectory can be projected fairly accurately (although, for Apollo 13, venting from the LM's thermal management system did push it off course, and they had to adjust their trajectory estimates as they went).
That leaves maneuvering during the reentry phase as the primary source of uncertainty in the location of landing. Apollo flies a lifting trajectory once it begins to enter the atmosphere, in order to keep from descending too steeply into dense air and suffering a high-g deceleration. In this phase, the capsule, under manual control or automatic guidance, rolls to adjust the amount of lift -- if the capsule is oriented vertically in the roll plane, it gets maximum lift, "going long"; if the rolled towards the horizontal, it descends more steeply (while pulling slightly to the right or left). Depending on slight variations in how steep the approach starts, and how it flies this phase, there can be a fair amount of variation in where it ends up.
According to Apollo By The Numbers, the crewed Apollo flights all splashed down within 3 nautical miles of the target point. Apollo 14 hit closest, at 0.6 miles, Apollo 16 furthest at 3.0.
-
1$\begingroup$ Ah, perhaps not! The sim scene from Apollo 13 may be looming too large in my mind. I have edited my answer to create tactical ambiguity. $\endgroup$ Commented Sep 12 at 1:04
-
2$\begingroup$ Re: the bit you deleted: I suspect you're thinking of the EMS, which contains the EMS scroll (a sort of paper nomogram of velocity vs. acceleration for a nominal reentry) and RSI (which compares the current lift vector with the EMS's idea of whether lift up or down is needed). They did train for a manual reentry (to the extent of not burning up and not getting lost in space, not a pinpoint landing) but it was a contingency that no one ever wanted to use and no one ever did. $\endgroup$– hobbsCommented Sep 12 at 9:35
-
$\begingroup$ Is the apollo command module body axis aligned such that the attitudes maintained for a lifting reentry constitute roll changes? It seems more likely they would be pitch or yaw changes. Additionally, is it true that the capsule oriented "vertically" gives the most lift? I would expect an angle of attack of ~30-45 degrees to produce a more favorable lift coefficient. $\endgroup$ Commented Sep 12 at 12:33
-
6$\begingroup$ @AMcKelvy The aerodynamics precluded direct pitch or yaw control during reentry. The CM was stable in an attitude slightly tilted from "blunt end forward" because its center of mass was offset from the symmetry axis, but it was free to roll (roll is rotation about the symmetry axis, don't let the image of an upright capsule make you think that's "yaw"). By rolling they could point the lift vector generated by that slight tilt up, down, or sideways. $\endgroup$– hobbsCommented Sep 12 at 13:16
-
1$\begingroup$ @AMcKelvy, here's another question on reentry: space.stackexchange.com/questions/43762/… $\endgroup$ Commented Sep 12 at 23:17
I am not sure that the existing answers really address the question asked, namely
How did NASA figure out when and where the Apollo capsule would touch down on the ocean?
The short answer is that NASA targeted a landing point and set up the entry guidance software to fly to that point.
They knew where it was going to land because they sent it there!
For the details of how they targeted it, see Apollo experience report: Mission planning for Apollo entry. Read the chapter titled ENTRY TARGETING. It starts out
Two types of targeting are used for the Apollo entry: velocity and flight-path angle V, y targets at the entry interface and entry range. The V, y targets are selected to ensure aerodynamic capture by the atmosphere of the earth while entry-ranging capability and acceptable aerodynamic heating conditions are maintained. The target entry range is chosen to be compatible with GNCS performance and to enhance the entry monitoring and backup control capabilities.
The entry-ranging capability is not used for control of the landing position relative to the surface of the earth. That is, the landing latitude is essentially at the lunar antipode at the time the CM enters the sphere of influence of the earth. This restriction occurs because the transearth trajectory must pass over the lunar antipode once the gravitational potential of the earth becomes the predominant force field, and the relatively short entry ranges result in the landing point always being located near the antipode. The landing longitude is controlled by varying the transearth injection time and by varying the transearth transit time to permit the earth to rotate to a favorable position relative to the transearth and entry trajectories. Therefore, the primary use of the variable entry-ranging capability is to make relatively small adjustments to the planned landing point during the mission to avoid bad weather conditions that may develop in the landing area.
Once the targets were set up and the vehicle was at Entry Interface, the automatic guidance system flew the capsule to the landing site.
From the Apollo 11 Crew Technical Debrief:
We gave spacecraft control over to the computer after we passed all our pitch attitude cross-checks. We gave it to the computer shortly before 400,000 feet. I don't recall exactly when, but a matter of seconds before 400,000 feet. We stayed in CMC, AUTO for the rest of the entry. The computer did its usual brilliant job at steering. We just sort of peered over its shoulder and I made sure that the spacecraft was responding to the bank angles that the computer commanded, and that those bank angles made sense in light of what we saw on the EMS and through other bits and pieces of information.
Acronymology:
- CM: Command Module
- CMC: CM Computer
- EMS: Entry Monitor System
- GNCS: Guidance, Navigation, and Control System
As answered here, all of the Apollo missions had accurate splashdowns. This level of accuracy actually started with Project Gemini several years earlier. The Gemini capsule was the first spacecraft which had the ability to maneuver to this level of precision during reentry.
The first three Gemini missions had various problems which led to splashdowns that were well off target. Reminiscent of some of the embarrassing mistakes seen in the past few years in spaceflight, during the Gemini 5 reentry the computers worked perfectly, however someone had inadvertently entered into the flight software the Earth's rotation as 360 degrees per 24 hours instead of 360.98, causing the capsule to miss the splashdown target by over 100 kilometers.
However the next several flights had increasingly accurate splashdowns, with the most accurate of the Gemini program being the splashdown of Gemini 9A less than a kilometer from its target.
Mission | Date | Kilometers | Miles | Nautical Miles |
---|---|---|---|---|
Gemini 3 | 3/23/1965 | 84 | 52 | 45 |
Gemini 4 | 6/7/1965 | 80 | 50 | 43 |
Gemini 5 | 8/29/1965 | 129 | 80 | 70 |
Gemini 6A | 12/15/1965 | 18 | 11 | 10 |
Gemini 7 | 12/18/1965 | 12 | 7 | 6 |
Gemini 8 | 3/17/1966 | 2 | 1.2 | 1.1 |
Gemini 9A | 6/6/1966 | 0.7 | 0.43 | 0.38 |
Gemini 10 | 7/21/1966 | 5.6 | 3.5 | 3 |
Gemini 11 | 9/15/1966 | 4.5 | 2.8 | 2.4 |
Gemini 12 | 11/15/1966 | 4.8 | 3 | 2.6 |
Gemini 9A splashdown (NASA)
-
2$\begingroup$ A more accurate value of the ERA (Earth Rotation Angle) is 360.9856°/24 hours, which equates to ~109.5 km at the equator. $\endgroup$– PM 2RingCommented Sep 12 at 2:36
-
1$\begingroup$ @PM2Ring - that's pretty close to the amount of error they had. It makes me wonder if someone didn't have the exact number handy at the moment, so they entered in 360 as a placeholder, intending to add the precise value in later. We know how that usually goes. $\endgroup$ Commented Sep 12 at 3:56
-
4$\begingroup$ Well, it's exactly 360° in 24 hours of sidereal time. ;) But I suspect some programmer simply goofed up, and didn't understand the sidereal vs mean solar time issue. Everyone knows the Earth rotates once per day. The astronomers know that it's once per sidereal day, and may not have even realised that that's not obvious to a non-astronomer programmer. $\endgroup$– PM 2RingCommented Sep 12 at 4:10
-
3$\begingroup$ @PM2Ring - Possibly. It seems hard to imagine someone that deep into orbital mechanics not knowing that, but I guess not impossible. These wouldn’t have been just outsourced programmers like we are used to today. Many of the high profile embarrassing mistakes are not as dumb-dumb as they are often portrayed, they were just mistakes. Not an exact example but the Mars Climate Orbiter mistake is often portrayed as supposedly smart people not knowing that the measurement system that you are using makes a difference. When in reality a conversion step was missed, and it went undetected in testing. $\endgroup$ Commented Sep 12 at 14:19
-
$\begingroup$ Seems pretty dumb to leave such domain expertise to a "non-astronomer programmer." I am terrible with math, and even with simple calculations, I get the accountant or bookkeeper to sign off on my code. $\endgroup$– ATL_DEVCommented Sep 12 at 17:19