Meadville Space Center
Welcome, Guest. Please login or register.
June 19, 2013, 08:07:53 PM

Login with username, password and session length
Search:     Advanced search
Welcome to the new Meadville Space Center forums!
20011 Posts in 1930 Topics by 2250 Members
Latest Member: mikegk
* Home Help Search Login Register
+  Meadville Space Center
|-+  Project Apollo - NASSP
| |-+  Project Apollo - NASSP Development (Moderators: movieman, Tschachim, Swatch, lassombra)
| | |-+  Apollo 7 Rendezvous Procedure
« previous next »
Pages: 1 2 [3] Print
Author Topic: Apollo 7 Rendezvous Procedure  (Read 3877 times)
jarmonik
Full Member
***
Posts: 130


View Profile Email
« Reply #30 on: July 07, 2011, 03:23:15 PM »

I didn't found the parameters of the orbit following the NCC1, so I don't know its period. But due to the fact that it seems that the Ap is rather high (the CSM overfly the SIVB more than 40 NM above) the period should be quite long, more than 90 min, so you can guess that the travel angle was less than 360° between NCC1 and NSR. In this case, conditions are satisfied for the NCC1 to adjust the Pe to the NSR height. But if I’m sure that the NCC1 shift the Lpe (it is mathematically almost mandatory) I’m not sure that the line of apside of the 2 vessels are aligned at that time. I’d rather think that this was done during the NSR.
Yes, I agree. The NSR is the one that must align the apsides but shifting LPe is a property that is required in both burns NCC1 and NSR. I made a small RDV test using the Target Intercept of IMFD. The offset was set 10Nm below the target and 85Nm behind the target and it did give a similar solution that is also rotating the apsides. Also, it wouldn't be possible to archive that kind of flight profile, as shown in a several PFD documents, without rotating the apsides.

Once again, I have the quite the same conclusion: In my mind the NCC1 places the CSM in the insertion orbit: the orbit from which the CSI occurs if we refer to the standard model. And the NCC2 is the CSI but due to the fact that this RDV is not following a lunar liftoff, there are no constraints about the insertion orbit. So they managed to compute a maneuver, the NCC1, so that the insertion orbit places the CSM in the proper way to directly perform the CDH. The result is that the CSI (NCC2) is nominally 0 and was eventually to be done only to null possible dispersion.
I agree.

I spent most of the day (while I was at work!!! Embarassed) thinking on this problem and I remember having the same questions 2 years ago while trying to put this feature in EOCA. I remember having concluded that one of the standard maneuver was dropped from the Apollo 7 RDV but which one?
Well, I suppose the CSI is the best candidate. But if the distance to the S-IVB really was only about 70Nm as shown in the documents then they could have dropped everything and executed the direct TPI right-away. But meik84 said something that they increased the distance since it wasn't sufficient for a meaningful RDV test.
Logged
Christophe
Project Team Member
Hero Member
****
Posts: 1072


View Profile Email
« Reply #31 on: July 07, 2011, 04:19:28 PM »

I made a small RDV test using the Target Intercept of IMFD. The offset was set 10Nm below the target and 85Nm behind the target and it did give a similar solution that is also rotating the apsides. Also, it wouldn't be possible to archive that kind of flight profile, as shown in a several PFD documents, without rotating the apsides.

In my mind the line of apsides orientation is the trickiest thing to match because there is a terrible lack of information about this in the NASA doc.


I didn’t remember IMFD allow to set an offset for interception.
May be I should look carefully at it.
In this case may be we can make all the RDV with it?
Do you think it’s possible?

I read a lot of doc today and I found some specifications for the different maneuvers. For example the CSI is always a horizontal (in local frame) burn. That mean that the delta V component is almost along the velocity vector (with low eccentricity) .
May be, applying those constraints we can compute all the maneuvers with IMFD?
Logged
jarmonik
Full Member
***
Posts: 130


View Profile Email
« Reply #32 on: July 08, 2011, 04:31:26 AM »

In my mind the line of apsides orientation is the trickiest thing to match because there is a terrible lack of information about this in the NASA doc.
It's a bit tricky because the non-spherical gravity is causing the orbital elements to jump up and down. That shouldn't be a problem after I implement a mean elements display in LTMFD. Also the advanced op-mode of the LOI program of LTMFD is almost exactly the CDH. It allows to set PeA, ApA and LPe. Check the page 7 from LTMFD manual. BTW, what constraints are there for CDH ?


In this case may be we can make all the RDV with it?
Do you think it’s possible?
I don't think so. The IMFD is based on a conic solution and it's not accurate enough. Also, the IMFD can't handle transfer angles greater than 360. There is a numeric lambert solver in the LTMFD but currently it is only used as an sub program for the other programs. There is no user interface for it but it shouldn't be difficult to create one.
Logged
meik84
Project Team Member
Sr. Member
****
Posts: 299



View Profile
« Reply #33 on: July 08, 2011, 07:58:13 AM »

Quote
But meik84 said something that they increased the distance since it wasn't sufficient for a meaningful RDV test.
Don't get me wrong: the first phasing maneuver was calculated so that at 26:25:00 the CSM would have been 75 nm ahead of the SIVB. However, the SIVBs orbit decayed more rapidly then expected (I guess they underestimated its drag), and so they had to do that second burn to restore those conditions.
Logged
Christophe
Project Team Member
Hero Member
****
Posts: 1072


View Profile Email
« Reply #34 on: July 09, 2011, 06:36:08 AM »

It's a bit tricky because the non-spherical gravity is causing the orbital elements to jump up and down. That shouldn't be a problem after I implement a mean elements display in LTMFD.

That’s sounds great because it’s really needed.
BTW that was one of the goals of EOCA. I’m not sure I did that very well according to some posts here on this thread, reporting some differences between the EOCA versus DSKY Pe and Ap values. At least I think that the computations I’ve implemented was not the same than the AGC uses.

An other point is the major axis secular shift due to the non spherical gravity field as a function of J2, Rt, a, i and e.
This seems to be a ignored by NASA doc. In my mind, this is because the target and the chaser are on similar orbits: Inclinations i are equals, of course Rt too, a and e are very closed and may be they consider that the 2 major axis shift at the same rate so they are fix relatively.
What do you think about that?   

Also the advanced op-mode of the LOI program of LTMFD is almost exactly the CDH. It allows to set PeA, ApA and LPe. 

I noticed that! Thumbs Up
 
what constraints are there for CDH ?

AFAIK the CDH constraints are:

- 180° away from the CSI: since the CSI delta V is always horizontal (+/- X in the local horizontal frame) it means that the CSI occurs at an apside cross (Pe or Ap) otherwise the  X impulse would make an undesired major axis shift.
So the CDH should occurs at the next apside cross after CSI (180°)
More exactly, the CDH occurs at a number N of apside cross after CSI. BTW it’s an input the astronauts enterd into the AGC during P32, V06N55, line 1.
N could be:
. 1: (1 half of orbit: 180°) or 3 (1 orbit and one half: 540°) for a so called “unconstrained delta H” CSI
. 2 or 4 (1 or 2 full orbits: 360° or 720°) that means that the CDH occurs at the same place than the CSI and CSI must be done at the coelliptic delta H. That’s why such a CSI is called “constrained delta H” CSI. Note that in this case, CSI is just a phasing. 

- At least 10 minutes after the CSI (with a travel angle of 180 ° it’s not a problem) and 10 minutes before the TPI.

- The CDH set Ap and Pe to the target one’s minus the delta H and shift the chaser’s orbit major axis parallel to the target’s orbit major axis

I can’t guarantee I’m totally right because I discovered yesterday that I’ve made a mistake in the RDV with EOCA: I thought the CDH occurs when the chaser crosses the target’s line of apsides and that was wrong.
So if I have made one mistake I can make another one!  Ashamed


The IMFD is based on a conic solution and it's not accurate enough. Also, the IMFD can't handle transfer angles greater than 360. There is a numeric lambert solver in the LTMFD but currently it is only used as an sub program for the other programs. There is no user interface for it but it shouldn't be difficult to create one.

So may be a spreadsheet for the RDV would be useful in order to present a synoptic page on a graph in the curvilinear coordinates. It would help the simmer to build the whole sequence.
Then, to perform the burns he would only use the AGC and or IMFD/LTMFD without requiring to go back again to any spreadsheet.
Logged
jarmonik
Full Member
***
Posts: 130


View Profile Email
« Reply #35 on: July 10, 2011, 05:41:16 PM »

This seems to be a ignored by NASA doc. In my mind, this is because the target and the chaser are on similar orbits: Inclinations i are equals, of course Rt too, a and e are very closed and may be they consider that the 2 major axis shift at the same rate so they are fix relatively.
What do you think about that? 
Yes, it is typical that errors like that tends to cancel each other in nearly equal orbits, therefor a conic calculations can be pretty accurate. But the AGC is using a numeric trajectory propagator for both active and passive vehicle orbits. The non-spherical gravity is often notified in calculations in the AGC. 


. 2 or 4 (1 or 2 full orbits: 360° or 720°) that means that the CDH occurs at the same place than the CSI and CSI must be done at the coelliptic delta H. That’s why such a CSI is called “constrained delta H” CSI. Note that in this case, CSI is just a phasing. 
That is exactly the issue that has been disturbing me. The delta H can't be adjusted/set during the CSI burn in that kind of condition, it must be already set before the CSI. This is also the case with the NCC1.

I thought the CDH occurs when the chaser crosses the target’s line of apsides and that was wrong.
Looks like it's done in the periapis of the active vehicle or N*180deg from CSI if the eccentricity is small than 0.000488.


Logged
Christophe
Project Team Member
Hero Member
****
Posts: 1072


View Profile Email
« Reply #36 on: July 20, 2011, 04:15:22 AM »

I've been at work in real life for some days so sorry for not having reply meanwhile.

However I found time to work on the new spreadsheet dedicated to RDV and I'm glad to say that I get some good results.
But I've had to almost entirely redo the old spreadsheet I've started a long time ago.

Now I have to fix a few bugs, to improve stability and to implement a friendly user interface and... above all, to write a doc for the user, that is not the smallest thing to do!

It will take some time but I'm able to promize that a beta would be released ASAP.
Logged
pattersoncr
Project Team Member
Full Member
****
Posts: 162


View Profile Email
« Reply #37 on: January 05, 2013, 12:30:02 PM »

Has there been any progress on how to calculate RDV burns?
Logged
Pages: 1 2 [3] 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!