Jump to content

  •  

CNers have asked about a donation box for Cloudy Nights over the years, so here you go. Donation is not required by any means, so please enjoy your stay.

Photo

How Much Computational Power to Gain the Moon?

EAA
  • Please log in to reply
6 replies to this topic

#1 Geo.

Geo.

    Vendor - Nexstar Parts

  • *****
  • Vendors
  • topic starter
  • Posts: 5078
  • Joined: 01 Oct 2008
  • Loc: Upstate NY

Posted 14 August 2019 - 09:44 AM

Good stuff:   https://www.wired.co...out-of-control/


  • bobharmony likes this

#2 bobharmony

bobharmony

    Mariner 2

  • *****
  • Posts: 265
  • Joined: 28 Oct 2017
  • Loc: Connecticut, USA

Posted 14 August 2019 - 10:15 AM

I remember clearly when that first program alarm 1202 was called during the descent.  It gave me chills, and the length of time to get a response from Mission Control, along with none of the TV "experts" explaining what was going on, just heightened the drama.  Great read, thanks for the link!  Talk about seat-of-the-pants flying in a stressful situation - I guess Armstrong was the right man for the job!

 

Bob



#3 bobzeq25

bobzeq25

    Hubble

  • *****
  • Posts: 17118
  • Joined: 27 Oct 2014

Posted 14 August 2019 - 11:14 AM

I read a somewhat different version of this, I think in the Armstrong biography, "First Man".

 

NASA was doing a simulation only a few weeks before the launch, with the backup crew.  The leader of the sim threw some computer error codes at the crew.  That had never been done before.  He knew what they didn't, that they were non-critical.  They aborted, and everyone was debriefed. 

 

So, when the lander reported the same computer error codes, Houston immediately replied "We're go on that code".  Simulations win again.

 

Another interesting thing.  NASA's major concern about Armstrong, well supported by his history, was that he was so committed to the mission that he wouldn't abort, in circumstances where they'd want him to abort.  The head of NASA took the crew aside before the launch.  Told them if they had to abort he'd shuffle the crews, and they'd get the very next try.  Who knows if he was telling the truth?  Who knows if they believed him?

 

Here's one way that played out.  Armstrong the engineer knew the LEM could fall unpowered from 50 feet in lunar gravity, and if it didn't hit too crooked, survive, the hydraulic legs could absorb the shock.  So while everyone's concerned about the fuel (which proved to be not an issue), Armstrong was less so.  He figured he didn't need enough fuel to land, he just needed to get to 50 feet.  <smile>

 

In the reality he landed so gently the hydraulic legs didn't collapse enough.  Why he had to make a big jump to the surface.  Later missions deliberately came down harder.


Edited by bobzeq25, 14 August 2019 - 11:29 AM.


#4 EJN

EJN

    Skylab

  • *****
  • Posts: 4416
  • Joined: 01 Nov 2005
  • Loc: Between what is and what's not there

Posted 14 August 2019 - 12:30 PM

I've read the above in Andrew Chaikin's "A Man on the Moon" and Gene Kranz's "Failure is not an Option." Kranz had an entire chapter called "SIMSUPS Win the Final Round" (SIMSUP = simulation supervisor).

IIRC the LM computer had 32K of memory.

At one time you could get scans of the source code from MIT, I don't know if it is still available. It was a custom version of assembler. I found the alarm codes in it.

The reason they got the alarms is because Aldrin left the rendezvous radar on in case that they had to abort. With the descent radar also on, the data stream was overloading the computer. Aldrin had done this in sims without a problem, but none of the simulations had been designed to provide data from both radars at once.

Edited by EJN, 14 August 2019 - 12:46 PM.


#5 BillP

BillP

    Hubble

  • *****
  • Posts: 18795
  • Joined: 26 Nov 2006
  • Loc: Spotsylvania, VA

Posted 15 August 2019 - 07:57 PM

I read a somewhat different version of this, I think in the Armstrong biography, "First Man".

 

NASA was doing a simulation only a few weeks before the launch, with the backup crew.  The leader of the sim threw some computer error codes at the crew.  That had never been done before.  He knew what they didn't, that they were non-critical.  They aborted, and everyone was debriefed.

 

Unfortunately, in today's world of heavy automation, there are many instances when there is no manual alternative to fly it yourself.  Case-in-point would be the recent 737-Max problem that caused the airliner to crash.  The pilots did disable the autopilot per procedures.  However, as I understand it another independent program re-invoked it causing the plane to go down. The days of being able to revert to manual processes are almost at an end, if not already. 


Edited by BillP, 15 August 2019 - 07:59 PM.


#6 bitnick

bitnick

    Sputnik

  • -----
  • Posts: 29
  • Joined: 21 Feb 2019
  • Loc: Sweden

Posted 18 August 2019 - 11:52 AM

IIRC the LM computer had 32K of memory.

At one time you could get scans of the source code from MIT, I don't know if it is still available. It was a custom version of assembler. I found the alarm codes in it.

The Apollo Guidance Computer had 2048 15-bit+parity words of RAM (magnetic core memory) and 36864 words of ROM (core rope memory), the latter in six exchangeable modules. There were two of these computers in each mission: one in the Command Module and one in the Lunar Module.

This amazing video series from CuriousMarc of a restoration (and usage for simulated moon landings, including the 120x alarms!) of an actual AGC is well worth a watch: link.

Transcriptions of original AGC software (as well as emulators for the AGC and some peripheral devices) can be found here and here.

Mike from the restoration video above has also created a gate-exact FPGA replica of the AGC and recreated the AGC Monitor, used by NASA for hardware debugging of the AGC: link.

Edited by bitnick, 18 August 2019 - 11:55 AM.


#7 bitnick

bitnick

    Sputnik

  • -----
  • Posts: 29
  • Joined: 21 Feb 2019
  • Loc: Sweden

Posted 18 August 2019 - 12:20 PM

From the Wired article linked by Geo.: "With almost no fuel to spare, the lander dropped, in slow motion, to kiss the surface upright, and the particles of moondust hung suspended in the sunlight until the gentle lunar gravity pulled them back to rest."

I think this is a bit of unfortunate reporter poetry. One of the astronauts who walked on the moon (I believe it was Armstrong, but I can't remember for sure) said in an interview that one of the strangest things on the Moon was the fact that the very fine moondust didn't form dust clouds - there is no atmosphere to slow it down or keep it suspended once it has been kicked up! So the particles kicked up by the rockets just disappeared over the horizon, and the particles kicked up by boots just fell back without forming any dust clouds.


CNers have asked about a donation box for Cloudy Nights over the years, so here you go. Donation is not required by any means, so please enjoy your stay.


Recent Topics





Also tagged with one or more of these keywords: EAA



Cloudy Nights LLC
Cloudy Nights Sponsor: Astronomics