Meadville Space Center
Welcome, Guest. Please login or register.
May 25, 2020, 11:15:10 PM

Login with username, password and session length
Search:     Advanced search
Gemini 060615 released!
25068 Posts in 2094 Topics by 2266 Members
Latest Member: twa517
* Home Help Search Login Register
+  Meadville Space Center
|-+  Project Apollo - NASSP
| |-+  Project Apollo - NASSP Development (Moderators: movieman, Tschachim, Swatch, lassombra)
| | |-+  LM Landing questions/report
« previous next »
Pages: [1] 2 3 ... 47 Print
Author Topic: LM Landing questions/report  (Read 128458 times)
LazyD
Project Admin
Sr. Member
*****
Posts: 412


View Profile Email
« on: May 16, 2005, 04:11:12 PM »

Hello,

I'm trying to get started getting P63-P66 working for the NASSP LM.  I have some questions someone might answer to help me get going:

1.  The landing site latitude and longitude seem to come from whoever creates the LM.  For Apollo 11, there are lat/lon numbers, for all other missions, only 0,0,0.  Should I proceed assuming these numbers will already be initialized?  (SetLanderData is the routine that gets called)

2.  I assume I should add my code to the autoland.cpp file?

3.  For attitude control should I use control.cpp or engine gimballing?  Gimballing would be more authentic, and any clues how to do this would be appreciated.

4.  I understand it's low-class to save class variables between time steps, but this is the way quadratic guidance works.  It calculates a descent profile, then flies it trying to arrive at the target correctly.  So I will need to do some saving and restoring, but it's not excessive.  My MFD version of this does this without problems.  I'm assuming the variables should be added to LEMcomputer.h?

I would welcome any input about how P63-P66 should behave.  I will probably start by just trying to fly the vessel properly, then get on with making it as authentic as anyone can stand.  

I am really excited about getting to work on this!

D.
Logged
movieman
Moderator
Hero Member
****
Posts: 1710



View Profile
« Reply #1 on: May 16, 2005, 04:44:57 PM »

The location should be specified in the scenario files. If they're not, you need to add them.

Yeah, autoland.cpp is where the code was meant to go.

I seem to remember having a long discussion about gimballing a while ago: the conclusion seemed to be that gimballing _wasn't_ really used much for attitude control.

You'll also need to add your variables into the 'erasable memory' functions so they automatically get saved and restored in the scenario file. But ideally you want to maintain as few as possible between time steps.

Incidentally, you may want to add a few 1201 and 1202 alarms to the Apollo 11 landing program just to keep people on their toes Happy.
Logged
LazyD
Project Admin
Sr. Member
*****
Posts: 412


View Profile Email
« Reply #2 on: May 16, 2005, 06:01:54 PM »

Great to hear from you, movieman!

The issue with gimballing is mostly this:  My MFD needs it turned off to use RCS, otherwise it counteracts anything I try to do.  Then, if we get to P66, the gimballing gives a perfect simulation of "attitude hold", which is exactly what we want at that stage.  If I use only RCS, I would like to be able to disable gimballing, then if a user selects P66, enable it again.  The LM is a LOT easier to fly manually with gimballing enabled.

I'll look at eraseable memory to save what I need to.  It would be cool if that could be done by default.

Thanks for the reply, I'll probably have a lot more questions, and hope you don't mind.

What you guys have already done is great!

D.
Logged
LazyD
Project Admin
Sr. Member
*****
Posts: 412


View Profile Email
« Reply #3 on: May 17, 2005, 04:39:15 PM »

Okay - I finally found landing site lat and lon are possed in EMEM072 and 73.  If you look at A15Delco, noun 89 is:

R1 latitude/1000 (+ is north)
R2 (longitude/2)/1000  adds a digit of precision and takes a factor of two
R3 altitude in nm (who cares for now)

I want to use the authentic representation because it gives 5 times more precision for lon, and 10 times for lat.  Make sense?

After a PRO response to lat and lon, P63 is supposed to display V06N61.  It is:

R1 time to go (braking phase)
R2 time from ignition
R3 Crossrange

I know how to calculate these numbers, but can you give me a clue where to make the display happen on the DSKY?

RCS control during descent and ascent used only downward-firing jets.  Is there a way to accomplish this with the NASSP LM?

And finally, the AGC used a 2 sec cycle to update displays and control inputs.  I want to reproduce this authentically - any advice pro or con?

Sorry if I seem to be an idiot.  I understand the guidance pretty well, but I'm new to C++, and fairly disoriented in NASSP at present.  Any help and advice will be appreciated, and I want to do this as authentically as possible.

Thanks for listening,

D.
Logged
movieman
Moderator
Hero Member
****
Posts: 1710



View Profile
« Reply #4 on: May 17, 2005, 05:35:16 PM »

Quote
can you give me a clue where to make the display happen on the DSKY?


You just need to add a new noun to the list that can be displayed. There's a big switch statement, AFAIR, with all those nouns in it.

Quote
Is there a way to accomplish this with the NASSP LM?


Probably not at the moment, but you could add code to the LEM that the AGC could call.

Quote
And finally, the AGC used a 2 sec cycle to update displays and control inputs


I thought it was 0.5 seconds? Anyway, that's easy: the program is called with the current mission time, so you just check that it's two seconds greater than last time before doing your processing... most of the programs already do that.

BTW, have you read the programming info?

http://www.ibiblio.org/mscorbit/phpBB2/viewtopic.php?t=40
Logged
LazyD
Project Admin
Sr. Member
*****
Posts: 412


View Profile Email
« Reply #5 on: May 18, 2005, 02:27:54 PM »

Thanks movieman,  I found the noun display stuff in lemcomputer, and it is very straghtforward - thanks!  I have added nouns 61 and 62, and got it counting down to PDI.  Things are making much more sense, and I thank you for your help.

Can someone tell me what was the function of "KEY REL"?  There is one in the P63 checklist, and I don't know what to do.

At this stage I'm just going to skip all the landing radar stuff and concentrate on getting the powered descent working properly.  Although for realism if it's Apollo 11 I could make a bunch of program alarms just to make it authentic!

D.
Logged
movieman
Moderator
Hero Member
****
Posts: 1710



View Profile
« Reply #6 on: May 18, 2005, 03:23:00 PM »

KEY REL releases the DSKY display for the computer to use. When you start pressing keys, the keypad takes priority, and if you then don't want to complete the entry you press KEY REL and the computer takes over again.
Logged
Tschachim
Project Apollo - NASSP
Administrator
Hero Member
****
Posts: 3700


nassp.sf.net


View Profile WWW
« Reply #7 on: May 19, 2005, 05:28:24 AM »

Hi LazyD,

sorry for my delayed answer, too much to do the last weeks and now I have a flue...
I added you as a developer to Project Apollo, I hope you have no problems with TortoiseCVS and the CVS stuff, otherwise please ask we will help you. Again welcome to Project Apollo - NASSP!  Very Happy

Because of P63: Unfortunately I have only little knowledge about the LM, but if you are looking for more informations about using P63 (or another LGC program) you probably want to take a look at the "R-567:  Guidance System Operations Plan for Manned LM Earth Orbital and Lunar Missions Using Program Luminary 1C (Rev. 131)" ( http://www.ibiblio.org/apollo/links.html ), it contains "flowchart" like descriptions of the programs and their usage. Especially this section should be useful: http://www.ibiblio.org/apollo/NARA-SW/R-567-sec4-rev7-P40-P69.pdf

... and another point/issue/proposal: RMaia did some changes of the AGC because of the ascent autopilot but unfortunately he was not able to merge and commit it to CVS. At the moment it's available as a separate download ( http://www.ibiblio.org/mscorbit/phpBB2/viewtopic.php?t=130 ). Perhaps you (or movieman) can take a look at it and merge it into CVS when you implement the descent programs. Otherwise it's no problem of course, perhaps I can try it later.

It's really great that you do all this AGC programs, this is a great improvement of Project Apollo!  Very Happy  Very Happy  Very Happy

Thanks and Greetings
Tschachim
Logged

movieman
Moderator
Hero Member
****
Posts: 1710



View Profile
« Reply #8 on: May 19, 2005, 06:15:28 AM »

I was trying to remember what had happened to those changes... Happy.
Logged
LazyD
Project Admin
Sr. Member
*****
Posts: 412


View Profile Email
« Reply #9 on: May 19, 2005, 09:27:14 AM »

Hi Tschachim!

I'm still finding my way around the AGC and DSKY code, haven't figured out CVS yet.  I'm just making changes to 6.4.1 for now, but my changes are very contained and isolated, so far I have only changed LEMcomputer and autoland.  (I am using Rodrigo's variables for now - I haven't added any class variables yet)  I think it will be easy to splice this in to the current version when it is working.

I'm starting out doing things self-contained - using my own attitude-control code and vector stuff etc, but later after things work I will try to wire it into existing LM systems.

Thanks for that document!  That's the kind of detail I need.

D.
Logged
LazyD
Project Admin
Sr. Member
*****
Posts: 412


View Profile Email
« Reply #10 on: May 19, 2005, 04:56:13 PM »

Another day, more questions...

I'm getting close to starting the PDI, but there are some things I don't understand:

1.  We start P63, PRO the landing coordinates and I switch to a V06N61 display, works fine, counting down to PDI and showing crossrange.

2.  Next in the sequence shows a N33E, which displays the time of ignition in R1=hh R2=mm R3=seconds.hundredths - coded the noun, and it works.  I assume the user keyed this in?

3.  Next shows a KEY REL followed by a PRO, which then flashes a V50N18, which is attitude angles for orientation.  Here's where I get confused - how do I detect the KEY REL, and what do I do when I get it, and then how do I detect the PRO?  This seems kind of out of the pattern of DSKY interaction I am used to.  Advice will be appreciated.


Another issue - I've been reading the Apollo Lunar Surface Journal trying to learn as much as I can about how this was done.  It is clear that Apollo 11 initiated PD "face-down", then yawed "face-up" at about 3:30 into the descent.  It seems from ALSJ that the rest of the missions went to PLI "face-up".  Does anyone know anything more about this?  I want to do it authentically even if different missions need to be different.

Okay, one more question.  At -7 seconds before PDI, the LM fired "down" RCS for ullage.  I find no information about how the astronauts knew this had occurred.  Does the DSKY do anything to indicate the ullage burn has happened?

I've been checking timings between the landing journals and my MFD, which is trying to mimic the Apollo guidance, and things are reasonably close.  I am doing throttledown about 35 seconds early, and P64 starts about 4 seconds late.  The descent profiles look very similar.

I want to do this right, and any information or advice is really welcome.

D.
Logged
movieman
Moderator
Hero Member
****
Posts: 1710



View Profile
« Reply #11 on: May 19, 2005, 06:57:42 PM »

Basically, they're showing the user entering NOUN 33 ENTER so they can monitor the time to ignition. Then they press KEY REL to release the keyboard and then they press PRO to move on.

So all you actually have to do is wait for them to press PRO and then go to the next stage in the program. They say to press ENTER a bit later on, I'd change that to pressing PRO instead.

Not sure about up/down, but you can get the mission number from the ApolloNo variable.

Quote
At -7 seconds before PDI, the LM fired "down" RCS for ullage. I find no information about how the astronauts knew this had occurred.


I would imagine they'd hear it pretty well from inside the LEM Happy.
Logged
LazyD
Project Admin
Sr. Member
*****
Posts: 412


View Profile Email
« Reply #12 on: May 19, 2005, 07:10:58 PM »

Thanks movieman,

I think my confusion involves the verb/noun displays that the program displays automatically, and what the user/astronaut can request.  I'll keep working on this and try to get the landing working without getting too anal about every detail of the DSKY interaction.  We can work that out later.

Also, please disregard my face up/down question in the post above.  I now believe that all Apollo landings did a yaw maneuver at +3 minutes from face-down to face-up.

So far, so good!  This is becoming fun!

D.
Logged
Coussini
Guest
« Reply #13 on: May 20, 2005, 11:33:48 AM »

[They are flying with engine forward and windows down. After they finish landmark tracking to confirm their trajectory during the first few minutes of the powered descent, they will yaw the spacecraft (a rotation around the thrust axis) to put themselves in a windows-up orientation. This will be done so that, as the spacecraft pitches toward an upright orientation for the landing, they will be looking out the windows at the landing site. During a post-flight press conference, Neil indicated that they passed over the crater Maskelyne W about three seconds late. In a September 24, 1969 memo, Howard Tindall discussed factors which make pinpoint landings difficult. He cited "thrust from the LM water boiler" resulting in a 6,000 foot miss distance, or about 1/4 of the 4 miles by which Apollo 11 overflew its target. Readers should note that Paul Fjeld is doubtful about this conclusion. He writes: "The vent for the sublimators was angled up to a deflector which should have sprayed the gas evenly with little net thrust to the spacecraft. As far as I can find out, the design for the vent was never changed for the later, accurate LMs."] Wink
Logged
LazyD
Project Admin
Sr. Member
*****
Posts: 412


View Profile Email
« Reply #14 on: May 20, 2005, 01:59:42 PM »

Here's how P63 is set up at present - I welcome any input or suggestions.

V37N63E  Loads P63

Flashing V06N89
Displays:
R1=Latitude in degrees*1000
R2=Longitude/2 in degrees*1000
R3=Landing Altitude (zero for now)

PRO to accept and proceed

Flashing V06N61
Displays:
R1=Braking phase burn time
R2=Time from Ignition mm ss
R3=Crossrange distance km*10

PRO to accept and proceed

Flashing V06N33
Displays:
R1=Ignition hour hh
R2=Ignition minute mm
R3=Ignition second*100 ss.ss

PRO to accept and proceed (instead of Key Rel)

Flashing V50N18
Displays:
R1=Pitch for PDI degrees*100
R2=Roll for PDI
R3=Yaw for PDI

PRO to perform automaneuver to this attitude

V06N18
Displays:
R1=current pitch
R2=current roll
R3=current yaw

When attitude is achieved:

Flashing V50N18
Display: same as above

PRO to accept retrograde face-down attitude

V06N62
Displays:
R1=Absolute value of velocity
R2=Time from ignition mm ss
R3=Delta V accumulated (zero until ignition)

At t-35 seconds, the DSKY blanks, then resumes above display

At t-5 seconds,

Flashing V99N62
Displays same as above

PRO to proceed with ignition

Flashing V06N63
Displays:
R1=Delta Altitude
R2=Altitude rate
R3=Computed Altitude

Does anyone know what Delta Altitude and Computed Altitude are?  And is there a PRO I'm supposed to be waiting for and quit flashing?

I'm skipping the V57E for accepting landing radar data for now.  This is the point where A15Delco runs out of information.  It was my impression that the astronauts could have the DSKY display a variety of nouns related to their descent.

Also, when I blank the DSKY at t-35, when I call SetVerbNounAndFlash, the display doesn't show program 63 any more.  How can I fix that?

Looking down the road a bit, I'm going to need to select some keys for P64 and P66 to use.  

For P64 target redesignation,  I plan to use keypad 2, 4, 6 and 8 for closer, left, right and further respectively.  We also need a pair of keys for P66 rate-of-descent control - maybe + and -.

I know less than nothing about panels, but it would be really nice for P66 if the velocity cross-pointers could be made to work!

Sorry for the big post, but I'm about to wire the guidance into this, and it's easier to make major changes now than it will be later.  Also I'd like some feedback - I am trying to do this as authentically as I reasonably can, but do think this is too much work for the Orbiter users?  It would be easier for me to just P63 PRO and land the damn thing without any further interaction but it seems to me NASSP appeals to those who are fascinated with the details of the missions.  Please advise if you think I'm going down the wrong road.

D.
Logged
Pages: [1] 2 3 ... 47 Print 
« previous next »
Jump to:  

Powered by MySQL Powered by PHP Powered by SMF 1.1.10 | SMF © 2006-2009, Simple Machines LLC Valid XHTML 1.0! Valid CSS!