Apollo 11 Fuel Level at touchdown

[Zakalwe]

It's an oft-repeated "fact" that the Eagle had only about 20 seconds fuel remaining when it touched down. It certainly adds to the tension of the story and adds a little of the "spice" that journalists so love.  However, I am sure that I have seen it somewhere that the fuel level was later re-calculated and they had a lot more to spare. The design of the anti-slosh baffles allowed the fuel to slosh around and casued the instruments to under-read. I also seem to remember seeing that the anti-slosh baffles were later re-designed.

Is my recollection correct or did I dream this? If my recollection is correct, do you have a verifiable source for it?

Thanks in advance.

[BertL]

I think I heard this from In the Shadow of the Moon, but the "20 seconds left" was an indicator that if they couldn't find a landing spot in that time the mission was aborted and Aldrin and Armstrong were to rendezvous with the CSM again.

[Zakalwe]

Arrgghh...as luck would have it...20 seconds after I hit "Post" I found the information here: http://www.hq.nasa.gov/alsj/a11/a11.landing.html

"Post-flight analysis indicated that Neil landed with about 770 pounds of fuel remaining. Of this total, about 100 pounds would have been unusable. As indicated in an unnumbered figure ( 63k ) from page 9-24 in the Apollo 11 Mission Report ( 5.7 Mb ), the remainder would have been enough for about 45 seconds, including about 20 seconds for an abort."

". Slosh uncovered the Quantity Gauge, latching the light early, losing the crew half a minute of flight time. It also made the LPD unreliable. Apollo 12 flew with the same configuration and its Quantity light came on early"

Grumman redesigned the baffles to prevent the problem:


The engineers were able to demonstrate retro-fitting the new design through a 2"  Quantity System hole (maybe they missed their calling as gynaecologists!)

[Echnaton]

I have wondered how precise was the knowledge the controllers had about the error in their readings.  They could have gotten information from the earth and lunar orbit testing, but there was no way to look into the tank to see how much gas was really left, so to speak.  As disciplined as Armstrong was, I also wonder if he would have ignored an abort order under circumstances where he was almost there.  IIRC Aldrin said that he thought Armstrong would have and it is hard to tell a test pilot how to fly his craft from a quarter of a million miles away. 

These were guys of great skill with nerves of steel.  I still feel a sadness with the loss of Armstrong.  The world seems a little less bright. 

[Zakalwe]

I think I heard this from In the Shadow of the Moon, but the "20 seconds left" was an indicator that if they couldn't find a landing spot in that time the mission was aborted and Aldrin and Armstrong were to rendezvous with the CSM again.

From the same source that I linked to:

"The quantity light latched at 102:44:31, and indicated that 5.6% of the original propellant load remained. This event started a 94-second countdown to a 'Bingo' fuel call which meant 'land in 20 seconds or abort.' So if the count gets down to zero, Neil will have 20 seconds to land, if he thinks he can get down in time. Otherwise, he will have to abort immediately. If you're 50 feet up at 'bingo fuel' with all of your horizontal rates nulled and are coming down to a good spot, you could certainly continue to land. With your horizontal rates nulled at 70 to 100 feet, it would be risky to land - perhaps giving you a landing at the limiting load of the landing gear. At anything over 100 feet, you'd punch the abort button, say goodbye to the moon, and stew for the rest of your life!"

The Contact Light came on 55 seconds after the Fuel Quantity light, which lit at 102:44:45. CAPCOM called 60 seconds to the "bingo" call at 102:45:02. But  the lamp lit early and the post-flight analysis showed that they had enough for 45 seconds of flight after they had touched down.

[ka9q]

At about 7:45 in the descent Charlie Duke calls up "Eagle, Houston. It's Descent 2 fuel to Monitor. Over."

There were two fuel quantity indicators, and I believe the mission rules called for monitoring whichever provided the lowest reading. That may also have biased the fuel estimates on the low side.

Gene Kranz had this to say in Failure Is Not An Option:

Carlton's [LM CONTROL] backroom controller, Bob Nance, using a paper chart recorder, is mentally integrating throttle usage by the crew and giving Carlton his best guesstimate of the hover time remaining before the fuel runs out. During training, he got pretty good at it and could hit the empty point within plus or minus ten seconds, but I never dreamed we would still be flying this close to empty and depending on Nance's eyeballs. I wait for Carlton's next call.

I see stuff like this all through Apollo and I still find it astonishing. Everything was so manual, so analog, so ad-hoc...so...primitive. So much was done in people's heads, most of it very esoteric and requiring a detailed understanding of a system's smallest details. Extensive crew and controller training and mission rules writing had almost the same role that real-time software development and testing does today -- working out the appropriate responses for each condition, examining all the corner cases, improving reliability and minimizing response time. The controllers and their support people almost (or in Nance's case,  literally) became computers, each one doing a very specific job.

So much of that effort could be so easily automated today, but that just wasn't practical in 1969. All that intense manual effort -- and the fact that it worked so well -- makes the Apollo achievement even more impressive.

[Noldi400]

I see stuff like this all through Apollo and I still find it astonishing. Everything was so manual, so analog, so ad-hoc...so...primitive. So much was done in people's heads, most of it very esoteric and requiring a detailed understanding of a system's smallest details. Extensive crew and controller training and mission rules writing had almost the same role that real-time software development and testing does today -- working out the appropriate responses for each condition, examining all the corner cases, improving reliability and minimizing response time. The controllers and their support people almost (or in Nance's case,  literally) became computers, each one doing a very specific job.

So much of that effort could be so easily automated today, but that just wasn't practical in 1969. All that intense manual effort -- and the fact that it worked so well -- makes the Apollo achievement even more impressive.

You know, I think that fact probably contributes to the HB's mind-set - at least the younger ones. They just can't understand how it could be possible to do the complex and difficult things Apollo did without modern computing power.

[ka9q]

As disciplined as Armstrong was, I also wonder if he would have ignored an abort order under circumstances where he was almost there.

The guy I wonder about is Shepard. As you probably know, Apollo 14's landing radar didn't lock onto the surface at the expected time. Eventually they got it going by cycling power (which cures 99% of all electronics problems even today) but there's been a lot of speculation as to what Shepard would have done if the radar never came up. When asked that question he answered "You'll never know!" (Maybe that's because he didn't know the answer himself.)

Kranz's take based on knowing Shepard was that he would have tried for a landing but would certainly have been forced to abort as the propellant margins were simply too tight.

[Inanimate Carbon Rod]

Kranz's take based on knowing Shepard was that he would have tried for a landing but would certainly have been forced to abort as the propellant margins were simply too tight.

I've seen Ed Mitchell being asked the same question, and he believed that Shepard would have gone for it.

[ka9q]

There seems to be some disagreement on the details here.

The purpose of the landing radar was to remove integrated errors in the IMU. Every inertial guidance system has inaccuracies that build with time due to errors in the accelerometers, drift of the stable platform, or errors in the local gravity model. These showed up as biases in altitude and velocity that were usually too great for the later phases of a lunar landing to be safe.

But I see different figures on the delta-H (difference in height) reported by the Apollo 14 landing radar and IMU before the radar was used to update the IMU. One report has it at several thousand feet, within spec but far too great to land safely without the radar. Another has it at only 300 feet, which may well have been close enough to permit Shepard to land by the seat of his pants if he was willing to violate mission rules and risk it.

[ka9q]

As discussed in the ALSJ at 108:09:39, the discrepancy regarding delta-H on Apollo 14 is between Gene Kranz's account in Failure Is Not An Option and the Apollo 14 mission report. Kranz says it was almost 4,000 feet, which is why Shepard would have been forced to abort due to lack of fuel. But the mission report says that once the radar settled down, the delta-H was more like 300 feet.

I'm inclined to go with the mission report. By Gene's own admission he was "weak on the trajectory", having a better understanding of spacecraft systems than of orbital mechanics. He did not work the Apollo 14 mission, and he might have been using a dated memory and/or misunderstood what he had heard from others.

[ka9q]

Speaking of powered descents, while playing with the 1970s-era "lunar lander" computer game I soon learned that the most fuel-efficient technique was to land at full throttle, timing your burn to decelerate to a stop the instant you contact the ground. This minimizes gravity loss, the bane of any powered descent. Or launch.

In other words, ideally you conduct a launch in reverse.

The problem is that there's absolutely no margin for error. The slightest error in timing or in the thrust or performance of your engine and you either hit the ground with residual speed or come to a stop above the surface, run out of gas, and fall the rest of the way.

You also see this problem in landing on Mars. Although you can use a heat shield and a parachute to remove much of your unwanted velocity early in the landing, the atmosphere is so thin that the final landing must be done on rockets - a powered descent. Although radars on the Mars Exploration Rovers (Spirit and Opportunity) gave exact altitude and velocity, uncertainty in the exact thrust and burn time of the solid fuel landing rockets still made it impossible to come to a stop just as the payload contacted the surface. The system was instead designed to come to a stop 10-15m above the surface and fall the rest of the way on airbag cushions.

Unlike solids, liquid rockets can be shut off and restarted and their performance measured and adapted to in real time. LRO has provided high resolution maps of many potential lunar landing sites, eliminating the need to search visually for a safe landing spot during approach. The two GRAIL spacecraft are about to provide a very high quality lunar gravity model. So everything needed to approach the ideal full-power landing is now coming together and future lunar landings should be possible with less propellant. But it will still be a hell of a ride.

Another idea that occurred to me is to give yourself a comfortable propellant margin for landing and then pump any residual propellants to the ascent stage. Then your margin wouldn't be wasted; the more you have after landing, the greater the mass of samples you could carry with you when you return.

[smartcooky]

I see stuff like this all through Apollo and I still find it astonishing. Everything was so manual, so analog, so ad-hoc...so...primitive. So much was done in people's heads, most of it very esoteric and requiring a detailed understanding of a system's smallest details. Extensive crew and controller training and mission rules writing had almost the same role that real-time software development and testing does today -- working out the appropriate responses for each condition, examining all the corner cases, improving reliability and minimizing response time. The controllers and their support people almost (or in Nance's case,  literally) became computers, each one doing a very specific job.

So much of that effort could be so easily automated today, but that just wasn't practical in 1969. All that intense manual effort -- and the fact that it worked so well -- makes the Apollo achievement even more impressive.

You know, I think that fact probably contributes to the HB's mind-set - at least the younger ones. They just can't understand how it could be possible to do the complex and difficult things Apollo did without modern computing power.

I get that often with younger people I employ. One of them would be doing some kind of price calculation,

Youngster: "What's twelve times nine?"

Me: "108"

Younger: "How did you do that so quickly without a calculator?"

I shake my head in disbelief.