12
$\begingroup$

During powered descent a 1202 alarm occurred when LM Pilot Buzz Aldrin asked the Apollo Guidance Computer (AGC) for the difference between computed and measured altitude above ground by typing V16N68 (this is well explained here).

I am analyzing the raw telemetries of the landing; the part showing registers status is not clear to me at 102:38:27, at first 1202 alarm:

table

The above source states that "V90N50E" is an "Aldrin input to stop the alarm", but there are some problems here:

  • The recorded squence is V50N90E, not V90N50E
  • Same author says in another site that the sequence was V05N09E
  • I read that verb 50 means "Please perform", i.e. it's an AGC request to the operator.
  • I also read that verb 50 means user request to AGC
  • Noun 90 does not even exist

So it's a mess...

Possibily 50 and 90 have a different meaning w.r.t. the "cheatsheets", being them associated to the alarm, just like R2 and R3 have no meaning during the alarm, but where to find some documentation about this topic?

As a side note, here it is what astronauts were trying to see "by numbers", not being available any graphical monitor:

altitude

Delta altitude

(data taken data from Register 3 (R3) of Apollo 11 LM-5 AGC (=LGC), recorded here, pages 1-10, and from "APOLLO MISSION 11 TRAJECTORY RECONSTRUCTION AND POSTFLIGHT ANALYSIS - VOLUME 1", data transcripted here). Register 3 contents becomes 00000 upon alarms 1201/1202 and DeltaH when V16N68E is working. Data table: https://github.com/jumpjack/Apollo11LEMdata/blob/master/registers-verb-noun-prog-altitude.tsv )

Per Apollo GN&CS guide:

landig radar availability statement

Per Timeline book:

timeline book

Telemetries:

telemetries

Improve this question
$\endgroup$
6
  • $\begingroup$ I found that the 0000 in R2 and R3 are not meaningless: the 3 register contain FAILREG, FAILREG+1 and FAILREG+2 registers, but in case of 1202, FAILREG +1 and +2 are empty. $\endgroup$
    – jumpjack
    Feb 17, 2022 at 17:46
  • 1
    $\begingroup$ I have been also suggetsed that probably there is a bug in the telemetry: those "50" and "90" do not read as V50N90E, but probably as V5N9E, typed in place of V05N09E, which is the command for "show alarm code" to be passed to AGC in case PROG lamp gets lit: program number displaying is probably not automatic, it must be requested by user; and the N separator would make it legal to type single digits without 0s. $\endgroup$
    – jumpjack
    Feb 17, 2022 at 17:52
  • 1
    $\begingroup$ I think you've at least partially answered your own question. Armstrong calls the program alarm at 102:38:26. At this time according to your sheet, Aldrin is already trying to find out what the alarm code is by entering V5N9. Armstrong sees the 1202 on the display and calls it down to Houston at 102:38:30. Why the 5 and 9 get changed to 50 and 90 in the recorded log, while verb 6, at least, doesn't become 60, is a good question, but it may be very difficult to find an answer. $\endgroup$
    – Russell Borogove
    Feb 19, 2022 at 20:34
  • 1
    $\begingroup$ It's kind of neat to cross reference your sheet with the annotated flight journal. At 102:38:42 Armstrong says "Let's incorporate" (referring to the landing radar data); Aldrin instantly enters V57. A bit later at 102:39:14, Aldrin calls down "same alarm, and it appears to come up when we have a 16/68 up" -- the alarm returned when he went back to the V16N68 display, and he's made the association. $\endgroup$
    – Russell Borogove
    Feb 19, 2022 at 20:39
  • 2
    $\begingroup$ I did this crossreference: user-images.githubusercontent.com/1620953/… I wonder if astronauts typed V5N9 rather than V05N09 because they were in hurry. The V6N63 sequence is the one previously active, probably the display autoswitched back to it automatically. $\endgroup$
    – jumpjack
    Feb 19, 2022 at 21:18

1 Answer 1

Reset to default
4
$\begingroup$

Apparently the answer is "They don't mean anything, it's a telemetry bug".

More precisely, that "Verb = 50, Noun = 90" actually reads "Verb = 5, Noun = 9".

i.e., Aldrin probably typed V5N9E; but computer expected two digits per number, so it actually did "see" V - 5 - blank - N - 9 - blank" sequence.

But, to save precious memory locations (each single bit required a complex wiring of core memories, and it increased weight), a very peculiar hardware/software algorithm (link for real nerds: link) was invented to represent the 18 digits of the 3 registers (5 digits plus sign), which eventually resulted in this truth table:

  • BLANK: 00000
  • 0: 10101
  • 1: 00011
  • 2: 11001
  • 3: 11011
  • 4: 01111
  • 5: 11110
  • 6: 11100
  • 7: 10011
  • 8: 11101
  • 9: 11111

Notice the first two lines!

A "blank" digit (all segments turned off) corresponds to value 00000 sent to ground!

Instead, a "0" digit (6 segments turned on) is represented by octal number 10101.

Hence, 5+blank, 9+blank resulted in 11110+00000, 11111+00000, which was erroneously decoded into 50, 90.

Typing V5N9E is perfectly legal, given that there is the "N" which separates verb from noun and prevents amibiguity. Usually the astronauts typed single-digit verbs and nouns as two digits (0 followed by digit); in this specific case, probably due to hurry caused by alarm, they just typed the "short version", which resulted in the telemetry bug.

This is better explained here: https://github.com/virtualagc/virtualagc/discussions/1166

The sources

These are excerpts from LMA790-3-LM - APOLLO OPERATIONS HANDBOOK:

handbook

It's not exactly for LM-5, hence alarm codes are slightly different, 31201 and 31202 rather than 01201 and 01202, but it explains well the V05N09E sequence.

Another possible confirmation that V50N90 is an user input rather than LGC output comes from timing and flashing; in the below transcription of raw telemetries we can see that:

  • 1201/1202 codes are displayed only after 50/90 (V5N9E) appears
  • 50/90 (5/9) do not flash, although verb/noun flashing is recorded by LGC in bit 6 of channel 11, and as far as I can understand from documentation available around, verb+noun flashing means "expecting user input"
  • R1 is filled with 1201/1202 only 2 to 6 seconds after the alarm code is loaded in FAIL REGISTER, and after couple 50-90 appeared in verb/noun telemetry.

Hence the sequence is:

  1. Astronaut types V16N68E
  2. LGC overflow
  3. 1201/1202 appears in FAILREG+0
  4. astronaut types V5N9E (V50N90 is sent to ground)
  5. FAILREG+0 content is displayed in R1

fail sequence

Full log of the alarms:

all alarms

legenda

Full image of telemetries: https://github.com/jumpjack/Apollo11LEMdata/blob/master/registers-verb-noun-prog.png

Data file for telemetries: CSV, TSV

Improve this answer
$\endgroup$
1

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge that you have read and understand our privacy policy and code of conduct.

Not the answer you're looking for? Browse other questions tagged or ask your own question.