V8 Release Work Thread

Status
Not open for further replies.

indy91

Addon Developer
Addon Developer
Joined
Oct 26, 2011
Messages
1,224
Reaction score
582
Points
128
Just to see if it was feasible, and because my method to make this mission flyable at all was very buggy, I have modified the Apollo 5 mission so that it consists of stages docked together:

LBa6uOt.png


Apollo 5 was a good test case for this as breaking old scenarios won't have a big impact on many NASSP users.

So this is S-IB-4, S-IVB-204, LM-1 and the AS-204 Nose Cone docked together. I definitely think this has many advantages and only a few disadvantages over every stage being one vessel in Orbiter. You can really treat each stage on its own instead of having one giant Saturn class which has everything for the CSM and the Saturn stages. And it was even worse for Apollo 5, as previously I had to build a Saturn IB around the LM class, so the LM.dll included all the IU and S-IVB systems as well. What gets a bit more complicated is sending signals between stages, docking port management (docking ports being deleted after staging so that you don't accidentally re-dock) and vehicle aerodynamics that have to be adjusted with each docking event. Other than that I quite like this solution.

I might try to get this working for Apollo 7 next. And being rather flexible with the stages that are connected will also make for an easier implementation of AS-203 (no CSM or LM, not even a SLA, just a nose cone) and Skylab and any fictional mission or combination of stages you can imagine.

As before, there is an T-1 hour scenario under WIP Scenarios, which has a powered down LM. And the scenario which I recommend people to use is the T-1 minute scenario which is under Apollo - Mission Scenarios\Apollo 5. In that scenario you just need to enjoy the show, as the mission is fully automatic.
 

n72.75

Move slow and try not to break too much.
Orbiter Contributor
Addon Developer
Tutorial Publisher
Donator
Joined
Mar 21, 2008
Messages
2,687
Reaction score
1,337
Points
128
Location
Saco, ME
Website
mwhume.space
Preferred Pronouns
he/him
Well, I haven't been able to connect to the IRC channel recently (life, work, etc.), so I'll ask here.

It appears as though the Apollo 7 flightplan checklist is missing the P27/P30/P40 procedure for the NCC2 burn. For me at least the NCC2 burn was approximately 40 minutes before the NSR burn, which was plenty of time for me to call the appropriate checklists manually.

If I'm wrong, disregard. Otherwise I can add it.
 

indy91

Addon Developer
Addon Developer
Joined
Oct 26, 2011
Messages
1,224
Reaction score
582
Points
128
I think it's missing because the NCC2 maneuver is not normally done. The MCC uses 10 ft/s DV as the criterion to do the burn or not. Usually it shouldn't be that high. I think in reality they would have rather replanned the NSR maneuver a bit to avoid NCC2. So I guess the solution for this is to add a NCC2 checklist group with the P27/P30/P40 and that checklist group is called based on user input, PRO or FAIL. A bit like the decision to redo a P52 if the angle difference is too large.
 

n72.75

Move slow and try not to break too much.
Orbiter Contributor
Addon Developer
Tutorial Publisher
Donator
Joined
Mar 21, 2008
Messages
2,687
Reaction score
1,337
Points
128
Location
Saco, ME
Website
mwhume.space
Preferred Pronouns
he/him
I think it's missing because the NCC2 maneuver is not normally done. The MCC uses 10 ft/s DV as the criterion to do the burn or not. Usually it shouldn't be that high. I think in reality they would have rather replanned the NSR maneuver a bit to avoid NCC2. So I guess the solution for this is to add a NCC2 checklist group with the P27/P30/P40 and that checklist group is called based on user input, PRO or FAIL. A bit like the decision to redo a P52 if the angle difference is too large.

Okay, well that makes sense now. I dont think I nulled my residuals very well for the NCC1 burn if I recall.

The good news is I have about a million quick saves that I was going to rename and sort and then post here.
 

n72.75

Move slow and try not to break too much.
Orbiter Contributor
Addon Developer
Tutorial Publisher
Donator
Joined
Mar 21, 2008
Messages
2,687
Reaction score
1,337
Points
128
Location
Saco, ME
Website
mwhume.space
Preferred Pronouns
he/him
Because I doubt most of the average OF users watch the IRC channel like it's the world cup, and even fewer of you read my clever git commit comments, I thought I would document a few of the new (small) features I've added. over the last few months.

CSM High-Gain S-band Antenna Tracking and Signal Strength.

A schematic of the HGA system is shown below, taken from the CSM systems handbook.
uc


Previously, only the manual mode of the HGA worked and the signal strength was fixed at 75%. Now, both the auto track mode, and the reäcquire mode work.

There is a table in the image above that describes the antenna system logic based on the mode and beam width switches:

uc


There are three servo axes through which the antenna is driven. Antenna pointing vector elevation is less than 45°, the antenna is just moving in the azimuth and elevation directions (A and C servo respectively), but occasionally the B axis servo will step in "look left or right" as the target (Earth) tracks near the pole(antenna boom arm vector). This eliminates the need for extremely high A axis angular rate. If the elevation servo is above 45°, than the servos act in an X-Y mode with the B-C servos. All the while however, there is a negative feedback gain applied to the B servo from the A-axis' error, the goal of which is to reduce the "look left or right" angle to zero, by means of rotating the azimuth servo.

Reacq mode is similar to the auto mode, with the exception that upon signal loss, it reverts to manual mode and drives the antenna to the rotary switch direction.


For signal strength, we're now using a Friis equation-based signal strength model. for the S-band gauge in the CSM, 0-100% is -130dBm to -50dBm. Signal strength is now a function of: distance, frequency, power, tx and rx gains.

Tracking is accomplished by an error signal, which is generated by difference in the array elements, which is turned into servo commands by means of a proportional controller (with appropriate feedback).

CSM OMNI S-band Antenna and Signal Strength.

Also updated to use a Friis-based model, with a much more realistic antenna radiation polar.

LM Steerable S-band Antenna and Signal Strength.

This is essentially the same as the CSM but with only two axes and no beam or alternate auto modes.

uc
 

indy91

Addon Developer
Addon Developer
Joined
Oct 26, 2011
Messages
1,224
Reaction score
582
Points
128
Two long term projects of mine are going to be released today. One is a version of Sundance, the Apollo 9 LM software, is becoming available. I'll make a separate post about that as it's a very special topic. The other one is a fairly major LVDC overhaul, making it possible to run certain functions with the proper timing.

During boost, the LVDC continually runs a major loop, which does navigation and guidance. This takes about 1.7 seconds to complete, so it is comparable to the AGCs guidance cycle of 2 seconds during powered flight. Per interrupt the LVDC also runs a minor loop, which mainly does attitude control and this is done 25 times per second. In NASSP we already had the major loop running with the correct timing, but the minor loop (attitude control) was just done on every timestep, so it depends on frame rate, time acceleration etc. how often the minor loop is actually done.

This did not allow us to implement some features of the LVDC that allows for a much smoother steering. In NASSP the major loop calculates a new attitude to steer to. And in each minor loop a new attitude error is calculated from that and sent to the Flight Control Computer. In effect this lead to a large attitude error when the major loop had just calculated a new attitude and an attitude error closer to 0 just before the next desired attitude update. This was quite notable in our Saturn IB, moreso than in the Saturn V. You get an attitude rate spike every 1.7 seconds and in general the attitude rate goes up and down and isn't constant at all. In reality this is of course bad for loads on the rocket, aerodynamics etc.

In the real LVDC, and what I have now implemented, the major loop calculated a small increment for the minor loop to apply every time the minor loop is run, based on the knowledge how many minor loops are going to be run per major loop. So the desired attitude in the minor loop is continually incremented and ideally reaches the actually desired attitude just before a new attitude is calculated in the major loop. This of course leads to a much smoother steering.

One thing I had to do to make this possible is to run the LVDC timestep on a regular basis instead of once per Orbiter timestep/frame. We do the same for the Virtual AGC, so run X cycles of it per second, so our LVDC is now a bit more of a LVDC hardware simulation than before.

I have only implemented this in the Saturn IB LVDC so far, but it shouldn't be a big problem to implement the same thing for the Saturn V. There are some things the two LVDC versions have in common, so in the mean time the Saturn V LVDC might have worse performance in time acceleration, as a bit of its code is now run 100 times per second and not once per timestep. Probably is no big deal and it will only be this way until it gets the same update as the Saturn IB got. Just have to move that code to a place where it belongs in terms of timing requirements (major or minor loop).

Another thing, I don't know when it started, but NASSP seems to be (even) buggier than before. Might be caused by some bugs in this update, or not. The weirdest thing is a Service Module panel occasionally not being shown. I've only seen this during launch. Haven't been able to figure out what causes it, as it's fairly rare and inconsistent. If you also see this let me know.

So anyway, enjoy the smooth steering of the Saturn IB, and soon the Saturn V!
 

x170doom

New member
Joined
Jan 4, 2020
Messages
11
Reaction score
0
Points
1
so in the mean time the Saturn V LVDC might have worse performance in time acceleration, as a bit of its code is now run 100 times per second and not once per timestep. Probably is no big deal and it will only be this way until it gets the same update as the Saturn IB got.

just tested the Saturn V launch with Apollo 11, not sure if this change is the cause but about half way through the s-IV/b stage's first burn the LV GUIDE light came on and the spacecraft began violently pitching nose down. is it possible this was caused by the change to the launch guidance system, (sadly I wasn't able to see if the issue was corrected by setting the launch vehicle guidance to cmc since I didn't notice fast enough and the spacecraft was already 90 degrees of course when i realized)
 

n72.75

Move slow and try not to break too much.
Orbiter Contributor
Addon Developer
Tutorial Publisher
Donator
Joined
Mar 21, 2008
Messages
2,687
Reaction score
1,337
Points
128
Location
Saco, ME
Website
mwhume.space
Preferred Pronouns
he/him
just tested the Saturn V launch with Apollo 11, not sure if this change is the cause but about half way through the s-IV/b stage's first burn the LV GUIDE light came on and the spacecraft began violently pitching nose down. is it possible this was caused by the change to the launch guidance system, (sadly I wasn't able to see if the issue was corrected by setting the launch vehicle guidance to cmc since I didn't notice fast enough and the spacecraft was already 90 degrees of course when i realized)

I've been out of the dev and IRC loop for a bit, so @indy91, correct me if I'm wrong, but afaik the current LVDC update applies just to the Saturn Ib.

Do you have any quick save scenarios before or during launch you could share so we could test, that's probably the easiest way to answer what's going on.
 

x170doom

New member
Joined
Jan 4, 2020
Messages
11
Reaction score
0
Points
1
sadly no, however this happened with the apollo 11 t-20 minutes scenario ill check the t-20 seconds and see if the same issue arises

edit: i have the current state file here, this is a few moments after the guidance failed
 

Attachments

  • current state.zip
    26 KB · Views: 192
Last edited:

x170doom

New member
Joined
Jan 4, 2020
Messages
11
Reaction score
0
Points
1
just finished a launch using the t-30 seconds launch scenario, no error this time (although i had my hand on the lv guidance switch just incase) however upon seco the s-iv/b aps ullage motors fired and then when i tried to open the next checklist orbiter suffered a ctd. not sure if the aps firing was intentional and it was just a set of coincidences though

seems it was a one off error, however now the s-IV/b lm sep switch isnt working
 
Last edited:

indy91

Addon Developer
Addon Developer
Joined
Oct 26, 2011
Messages
1,224
Reaction score
582
Points
128
NASSP 8 has been updated to use Orbiter Sound 5, so make sure you are installing that version before updating to the latest NASSP release!

seems it was a one off error, however now the s-IV/b lm sep switch isnt working

That is now fixed.
 

AliDarwesh

New member
Joined
Sep 20, 2020
Messages
22
Reaction score
4
Points
3
Location
Turkey
I get a CTD when trying to abort (THC CCW) before liftoff. Abort works fine in all other circumstances after liftoff.
 

indy91

Addon Developer
Addon Developer
Joined
Oct 26, 2011
Messages
1,224
Reaction score
582
Points
128
I get a CTD when trying to abort (THC CCW) before liftoff. Abort works fine in all other circumstances after liftoff.

Yep, seems to be some unsafe code for communication between Mobile Launcher and Instrument Unit. Shouldn't be too difficult to fix I think.

EDIT: That bug should be fixed now. The NASSP auto builds on Github are still broken though as it isn't using the Orbiter Sound 5.0 files yet.
 
Last edited:

AliDarwesh

New member
Joined
Sep 20, 2020
Messages
22
Reaction score
4
Points
3
Location
Turkey
Yep, seems to be some unsafe code for communication between Mobile Launcher and Instrument Unit. Shouldn't be too difficult to fix I think.

EDIT: That bug should be fixed now. The NASSP auto builds on Github are still broken though as it isn't using the Orbiter Sound 5.0 files yet.

Any idea what's going wrong with my P22's in Apollo 7? https://www.orbiter-forum.com/threads/p22.36200/
 

jalexb88

Addon Developer
Addon Developer
Joined
Aug 11, 2008
Messages
152
Reaction score
154
Points
43
Location
Canada
LM VC update: The latest state has been pushed to the main branch. Panels 5,6,8,11,12,14,16 (including all CBs) are now functional. Also included is Jordan's latest work-in-progress: ECS panels which are static for now, panel detailing and other misc. equipment.

This update brings a new views system which aligns with the default Orbiter method for changing views in the VC. Ctrl-Arrows to change between neighboring view positions and from each position you can use Ctrl Alt-left/up/right to "lean" to a sub-view, then Ctrl Alt-down to return to the current view position.

LPD view: To access the LPD view for use during the approach/landing phases hit Ctrl-Down from the default CDR view position. You can then lean to quickly glance at the DSKY (for LPD angle) or flight instruments with Ctrl Alt-Arrow keys.

See screenshots here: https://www.orbiter-forum.com/threads/orbiter-screenshot-thread.8/page-392#post-572549
 
Last edited:

ggalfi

Well-known member
Joined
May 31, 2020
Messages
74
Reaction score
131
Points
48
Location
Budapest
I just want to tell that this LM virtual cockpit is absolutely awesome! I have flown some PDIs and a lunar ascent and it was working beautifully. Actually for the activities during PDI I've found the VC more convenient than the 2d panels. LPD also seems to me reasonably accurate.
 

jalexb88

Addon Developer
Addon Developer
Joined
Aug 11, 2008
Messages
152
Reaction score
154
Points
43
Location
Canada
Command Module Virtual Cockpit update:

The CM VC now has working panels 1-3! The only thing left to animate on panel 1 is the altimeter, work on that and the side panels will start soon.

Like the LM, the CM VC Now uses Ctrl-arrows to change between neighbouring view positions.
 

Thymo

I like breaking things
Addon Developer
Joined
Jun 26, 2016
Messages
120
Reaction score
148
Points
58
Website
nassp.space
Clogging for the fuel cells has been enabled again. If you have updated in the last two weeks it's suggested you update to the latest snapshot.
With today's patch the clogging has been tuned a bit so you don't have to do a purge every couple of hours.
If you forget to purge you will eventually get a master alarm due to a main bus undervolt, ignoring that the fuel cell will eventually fail due to being completely clogged,

n7275 is also doing a complete redo of the ways our fuel cells work using proper physics and chemistry to get the behaviour even closer to a real fuel cell so that, eventually, we can do things like fuel cell activation on the pad and fuel cell failure due to the electrolyte melting or solidifying.
 
Status
Not open for further replies.
Top