Apollo 15 Simulated Mission results

[ggalfi]

Hi All,

I went through the full Apollo 15 mission with the goal to stick to the Flight Data File (that means 90% of cases the flight plan, but the individual checklists as well) as much as possible. Actually I did this with the missions 12 and 13 as well, but meanwhile I improved my technique as much as I think my notes, calculations and pad loads are worth to share. To go through the whole mission it took me 10 months, between January and October. During that period I made a few version upgrade (including a commit with the overhaul of ECS). Even with running the update script on the saved scenarios, I had some minor glitches in the ECS:

• continuous need stir up H2 tanks in every 10 minutes to avoid CRYO PRESS warning.
• H2 was consumed significantly faster than O2, even latter was expended also by LM pressurization and with the deep space EVA. At 295 hours, when SM Sep happened, H2 was down to 2%.
• After a while I wasn't able to refill/pressurize neither the surge tank nor the emergency O2 tank more than 200 PSI.

But likely these were caused by scenario-NASSP version inconsistencies. Apart from these, I was merely on the abundant side of the consumables, despite the NASSP LM doesn't have those extras which were available on the J-missions (lunar battery, third O2 tank on CSM). Actually I executed all the prescribed depress/repress cycles (on CSM: one LM pressurization after T/D and the EVA; on LM: SEVA, EVA1-2-3, PLSS&crap dump), and I've never been close to run out of O2. The only thing were I've had to cheat was the battery power of LM: a few hours before the lunar liftoff, the voltage went down to a level that it began to cause failing devices. I decided to top up the descent batteries with 10 hours of allowance, that was the only case where I made a significant "cheating" by manually editing a scenario file.

My primary goal was to fly the mission as close to the nominal trajectory as possible. RTCC gave a great help in it, however I had to do a lot of calculation on my own (even it isn't trivial to answer the question, what is the nominal trajectory). I made a Libreoffice Calc workbook with all the necessary historical values, my own calculations, RTCC settings, and also filled PAD sheets. You may download it from here: www.absimp.org/orbitersim/apollo15/apollo_15_pad_sheets.ods . I've written it originally only for myself, so it could be difficult to decipher it sometimes, and I don't have time to fully document it. However I'm happy to answer questions about it, and here I give you some hints, what you'll find in it:

• The first sheet (Planning) is the most important one. It contains most of the trajectory calculations and parameters for RTCC. To be able to see how much I diverged from the real life values, I organized the numbers into three columns:
  • FP, which are mostly the values I found in the Flight Plan or occasionally in the Operational Trajectory document (these two are not 100% consistent with each other!)
  • PAD/MR which are the historical values either in the mission transcripts (available here and here) or in the Mission Report.
  • NASSP as you may guess is the column with the number relevant for the actual simulation.
• Also on this sheet you may see some side calculations. One is related to the different selenographical coordinate system used in Apollo Era than today. I tried different methods to determine the transformation between the two, but finally I simply looked at the preflight images/maps of the planned landing location and I searched the crater formation in LROC imagery - I used that coordinates in targeting. Additionally I put some effort to determine the accurate parameters for the descent orbit. For that I used nominal PDI conditions given in LM padloads.
• The "Pad Loads" sheet are containing some selected subset of values from the LM padloads. As mentioned above, I used them to get the accurate descent trajectory, and also to calculate PDI time. Latter was only 1.34 sec earlier than the nominal PDI point, so I was able to hit the planned descent orbit pretty accurately.
• GDC align: I made this sheet for the Apollo 13 mission which used a star pair for backup GDC aligment unsupported by RTCC. So this sheet allows you to copy their any pair of stars' coordinates, and also a REFSMMAT and it provides the angles for alignment. On Apollo 15 I rather used it for managing my REFSMMATs and remind myself where they were stored (don't forget, PTC REFSMMAT is reutilized after 6 days in lunar orbit!)
• RTGO: again something needed for earlier versions of RTCC, I calculated the RTGO on my own. However in this simulation I more relied on RTCC's Entry PADs.

The remaining sheets are connected to the individual events of the mission. It has absolutely no scientific value, but I found some aesthetical pleasure in fancying some PAD sheets and making them as authentic looking as Libreoffice allows it. Feel free to use them in any other NASSP mission.
Also I want to share the actual mission scenario files, here: absimp.org/orbitersim/apollo15/Apollo%2015%20mission%20scenarios.zip. All of the scenario files has the Ground Ellapsed Time in their names, and sometimes I added some other info as well. However, as I managed to remain within only a +-2 minutes of the original flight plan, what you see in the FP likely you will get when you start up the corresponding scenario file. What was really amazing for me, when the Flight Plan said that at X hours, crater Y is visible in, say, CM5 window, then it was exactly there in the simulation

 

[rcflyinghokie]

Not bad! Sounds like the issues you had were from the mid-mission updates and not using the J mission branch of NASSP which has the increased consumables?

All the ECS glitches you listed were from updating mid mission, and consumables would be from using the base branch not the J mission branch.

 

[ggalfi]

Not bad! Sounds like the issues you had were from the mid-mission updates and not using the J mission branch of NASSP which has the increased consumables?

All the ECS glitches you listed were from updating mid mission, and consumables would be from using the base branch not the J mission branch.

I've only used the Orbiter2016 branch. At the beginning my main focus was on trajectory and decided that I would disregard the tighter consumables due to the non-J configuration. I had thought that I would address serious ECS-problems by my "divine influence" which ultimately needed only in the case of the descent batteries. I didn't aim for a proper testing of the ECS-simulation, the only reason why I mentioned the glitches is that I don't know which one was caused by the version updates and which one is a permanent issue. For example, the faster H2 depletion is suspicious for me that some hard-wired flow rates caused that, however I'm not sure, as I haven't looked into the FC simulation.
I have to add that the last NASSP commit I used during the mission was 5bbc885 from May (I was afraid that the new IMU drift model would ruin things), so any development after that didn't have any effect on the simulation.

 

[rcflyinghokie]

I've only used the Orbiter2016 branch. At the beginning my main focus was on trajectory and decided that I would disregard the tighter consumables due to the non-J configuration. I had thought that I would address serious ECS-problems by my "divine influence" which ultimately needed only in the case of the descent batteries. I didn't aim for a proper testing of the ECS-simulation, the only reason why I mentioned the glitches is that I don't know which one was caused by the version updates and which one is a permanent issue. For example, the faster H2 depletion is suspicious for me that some hard-wired flow rates caused that, however I'm not sure, as I haven't looked into the FC simulation.
I have to add that the last NASSP commit I used during the mission was 5bbc885 from May (I was afraid that the new IMU drift model would ruin things), so any development after that didn't have any effect on the simulation.

Yeah everything you described is your older version when starting and not using the J mission branch, so nothing to worry about in the current iteration.

 

[n72.75]

There are a few "opportunities for improvment" with respect to the cryo fluids themselves. The big one right now is liquid bulk-moduli. They are currently fixed value. I've been working on making them dependant on pressure and temperature, but it requires transcribing an enormous table and I haven't had the time for a few months to work on it.

 

[ggalfi]

A small, but interesting thing which I forgot to mention: as in the real mission, I did the LM deorbit through Erasable Memory Program. I struggled quite a lot with P99, till the point I looked up the most recent available EMP document (here: https://www.ibiblio.org/apollo/Documents/LUM211R1-DM_text.pdf), and used both the checklist and the EMP given there. The latter was slightly bigger than what is available in the NASSP source (see the attachment). I don't know whether this newer EMP or the fact that I very strictly stuck to the checklist helped, but finally I managed to setup the P99 and got a very nice and accurate automatic RCS burn. The LM missed the target only by a few kilometers. When I say stricly sticking to the checklist I mean it literally: there is the part, where the commands are uplinked from Houston, that I typed in the LM's DSKY (I'm not aware of an LM Telemetry client). During that, even the temptation is big, don't use the PRO key, instead type V33E! I think, using PRO was the cause of all my P99 problems, and not the older EMP version. However, I suggest to include the attached EMP in the NASSP source, as this version is more likely been used during the real mission.

 

[rcflyinghokie]

A small, but interesting thing which I forgot to mention: as in the real mission, I did the LM deorbit through Erasable Memory Program. I struggled quite a lot with P99, till the point I looked up the most recent available EMP document (here: https://www.ibiblio.org/apollo/Documents/LUM211R1-DM_text.pdf), and used both the checklist and the EMP given there. The latter was slightly bigger than what is available in the NASSP source (see the attachment). I don't know whether this newer EMP or the fact that I very strictly stuck to the checklist helped, but finally I managed to setup the P99 and got a very nice and accurate automatic RCS burn. The LM missed the target only by a few kilometers. When I say stricly sticking to the checklist I mean it literally: there is the part, where the commands are uplinked from Houston, that I typed in the LM's DSKY (I'm not aware of an LM Telemetry client). During that, even the temptation is big, don't use the PRO key, instead type V33E! I think, using PRO was the cause of all my P99 problems, and not the older EMP version. However, I suggest to include the attached EMP in the NASSP source, as this version is more likely been used during the real mission.

Yeah the telemetry clients are very useful for this. We also have all the EMPs with RTCC now. I think your version predates all of those additions.

You might have missed specific steps like the V96 before initiating things. The V33 for PRO is only because the TM client keyboard does not have a PRO button like the DSKY.

You really ought to join us on discord sometime https://discord.gg/n255GFyNHW

 

[rcflyinghokie]

CSM TM Client

LM TM Client