Search Ares V

[cymrych]

Columbia42-- Sorry this took so long, but after reviewing a small mountain of the most recent data regarding Ares V (from October-ish 09), I had to do some serious number crunching. A fair amount of data has changed since the Point of Departure vehicle upon which your current model is based.

I made a number of minor changes to your cfgs, and even wrote an explanation of why I made them based on my analysis of the data from NASA and other sources. Unfortunately, my write-up is over 4 pages long. Instead of killing a page of this thread with my analysis, I'll attach it as a document.

In the final section, I list each modification to the configs I made.

On the guidance front, I know diddly about writing guidance routines. I do however, have a basic pitch program (using UAP) in place that hits the ascent profile marks pretty accurately. You still have to manually pilot the EDS to LEO however.

EDIT After testing some more, I discovered you'll need to convert the ISP number in the EDS config back to 4394 and change. Putting it to the actual Isp of 448 burns off all the fuel at an insane rate, so not sure what's going on with the model there. I'm just guessing, but I take it that term in the config doesn't mean the same thing it does in real life?

On a positive note, I have a pitch program which works well for the Ares V. Leaves a dV of about 2.0k for the EDS to take care of. Then it's just raise the nose of the EDS about 12 or so degrees AoA, and follow the Apa til you get to 240km.

 

[PhantomCruiser]

cymrych-
As far as the guidance file goes. I've got it plotted out every ten seconds for the duration of the burn (up until burnout at 297 seconds). I did have it for every 5 seconds but decided it just wasn't worth the headache.
Updating the file now basically inputs what pitch angle at what time. Then testing (over and over) until it's tweaked to the altitudes desired.
I'll input the data you provided and see how much the guidance file changes.

edit-Just tweaked the guidance file with the data you provided. The technical term can't be written out here... Bill the Cat would say "Ack!"

 

[cymrych]

Yep, not too sure why, but changing the Isp value to real-world value in the EDS config really screwed me up too. First time I stopped tinkering with my pitch program (which also utilizes the try, try, try again methodology, lol) and actually tried to orbit the EDS, I made it about half way there and realized I was almost out of propellant! Switching that line in the config back to Columbia42's orignal number works, but I'd still like to know what's going on.

But that's juts my OCD coming out. I expected to have about 45% fuel left when I circularized at 240k .... but I don't and it's gonna drive me mad until I figure it out!

So, for the programming-challenged ... how and where is your guidance program written? Is it something enbedded within one of the model's files, or is it a secondary, stand-alone application? It sounds like one of those ridiculously tedious, exacting challenges I seem to enjoy so much for some damned reason, lol!

Oh yeah, I too discovered early on that shuttle guidance just ain't gonna do it! Shuttle starts it's roll/pitch program something like 10 seconds in, then drops the nose in 110 seconds or so. Ares V, to get it anywhere even remotely close to the flight profile in the publications, you gotta get the nose down faster. From 90 to about 33 degrees by about T+70, hang there to let the direction marker drift 10-12 degrees down past the nose, then a slow drop down until burnout, ending up with nose about 13 degrees above horizon, orbital velocity about 5900, and at 133km (Apa roughly 190km or so at core burnout). If the nose doesn't drop, you end up with way too much altitude too quickly, your velocity sucks at MECO, and Apa is just seconds away instead of over 100 because you've launched a very large, lobbing lawn-dart!

---------- Post added at 02:12 PM ---------- Previous post was at 04:15 AM ----------

So I figured out what the "ISP" setting in the EDS config is all about. I think it is actually a measure of Ve, exhaust velocity (Ve=Isp x g), used in the computation of delta-V with loss of mass due to fuel comsumption. (Equation is deltaV = Ve x ln(M0/M1) where M0 is initial mass at start of burn, and M1 is mass at end of burn.)

I'll save you all the gory details, since it involves a few logarithmic functions, unless someone is actually interested, in which case I'll happily write it all up! But the number I have entered into the "Isp" line of of the EDS config is 4008.1, the result of my base Isp throttled back 91.2% times g. (448 x 0.912 x 9.81).

Knowing the deltaV you need to move the EDS into stable orbit at 240.8km, you can find M1, which in turn gives you fuel consumption (roughly 48.1%). From this, you can calculate the M0 and M1 masses for the TLI burn, and figure out the available deltaV left to do it (in this case, 3507m/s.)

First experiments seem to conform pretty accurately to these expected results, at least for LEO insertion.

Also think I figured out the guidance programming. I'm working on a mock-up know. Will let you know how it goes.

 

[PhantomCruiser]

Yes, please let me know when you've got some new numbers. I've spent many a launch here just to begin a streak of blue language right around T+290.

 

[cymrych]

Hell, I'm still trying to figure out how to get her to roll to heading and not just keep going straight up! So it may be a bit, lol

 

[Columbia42]

First of all, thanks for all the help, you guys. Secondly, to convert real-world ISP into Orbiter ISP, just multiply by 9.81. As for the guidance files, try reading the multistage2 doc on them. (good luck understanding it, however.)

 

[Urwumpe]

First of all, thanks for all the help, you guys. Secondly, to convert real-world ISP into Orbiter ISP, just multiply by 9.81. As for the guidance files, try reading the multistage2 doc on them. (good luck understanding it, however.)

Please don't call measuring specific impulse (Isp) in seconds the real world. It is complete crap, ruining US engineers already for decades (Resulting in you needing to import all important rocket engineers from liberated countries. ). Such measurements are the reason why you can't work with them properly without knowing that the seconds are crap. Specific impulse is impulse per propellant mass. The correct unit for specific impulse is "lb*s/slug". "lb * s" is the measure of impulse, slug is the measure of mass, defined as "The mass that accelerates at 1 ft/s² when subject to a force of 1 lb". One slug is equal to 14.5939 kg or 32.17405 pound-mass (which caused the error then).

Some people in the past just did the error of replacing propellant mass by propellant weight force (also in lb), resulting in "lb* s/lb = s". But this is wrong. If you would throw the value in seconds into the rocket equation, the number would make no sense, you would get the result in seconds as well, without any relation to the velocity change. With slugs, things make more sense.

Orbiter uses actually the correct version, in the SI unit system, it measures the specific impulse in "N * s/kg", which can also be reduced to "m/s" then meaning "average exhaust velocity".

 

[Columbia42]

Cymrych, I read your analysis of documented Ares v 09 thing and changed my configs accordingly. The modifications work but I do have a question:
What are the values for sea level and vac. thrust for the SRBs and core stage? (I ask this because I think I've found a way to simulate the change of thrust in Orbiter and your document wasn't all that clear on this point.)

 

[cymrych]

OK, for some reason the forum keeps timing-out on me tonight, and I'm getting freakin tired of typing this same post for the 3rd time.

So, Columbia42:

SRB (each)

SL 15340279N

Vac 16800000N

Core (each engine)

Sl 3123900N

Vac 3546650N

Urumpe: Thanks for the units clarification. I remember having a lecture about how measurements get all out of whack under a system which has a "pound" or a "kilogram" as both a unit of mass and a unit of force (weight.)

Actually, looking at some simple isp formula to "follow the units" reminded me that isp or exhaust velocity are independant of any throttle changes in the motor, since both thrust and fuel consumption are proportionally reduced.

In other words, Columbia and Phantom, I made an error.:shrug: That "isp" line in the EDS should go back to 4394.9!

---------- Post added at 01:06 AM ---------- Previous post was at 12:45 AM ----------

Although, try as I might, I still get a unit of time for Isp. Ex: Total impulse divided by weight = Isp. For the EDS, that is 1193220N (or kgm/s^2) x 927.8s burn over (251900kg x 9.81m/s^2) = 1107069516kgm/s over 2471139 kgm/s^2 = 448s. Basically, I see a ratio of thrust to weight of fuel flow, so what's the problem with having units of seconds?

---------- Post added at 02:03 PM ---------- Previous post was at 01:06 AM ----------

Hey Phantom, you have any luck getting her to roll to heading? I can get good pitch performance, but only if I modify the heading in the scenario file to exactly what I need, in this case, 90 degrees due east. If I try to get her to roll to heading, she VERY sluggishly turns the compass (as in, it'll take 60+ seconds to roll from 45 to 90), but it's twitchy, like the thrusters are fighting each other or something.

Is there a way to, say, turn off yawing during the guidance program phase? The manual says about turning off pitch and roll, but not yaw. I get the impression that the low roll rate is because she tries to yaw left while rolling right to keep the nose dead-centered on the pitch target. Or better yet, is there a way to just have her piroutte at pitch of 90 degrees, and THEN begin pitching onces she's on heading?

 

[PhantomCruiser]

Mmm, I've taken a look at the Universal AP and watched how that launched. I made an attempt to integrate that performance into the Multistage guidance file. It helped slightly. The UAP has her lauch, roll, then pitch (to what would be a head-up it it were Shuttle). I've been keeping it "head-down" because she has less roll to do that way.
I've been stopping the roll at 107 degrees, which would be pretty close to a lunar inclination (and less time to get tothan 73 degrees). I really screwed up the roll program a few times (but it looked impressive). I figure IMFD, TransX or the new UAP can get us to the moon, I'm just trying to point us in the right direction.

 

[Bj]

OK, for some reason the forum keeps timing-out on me tonight,

...as in you keep getting logged off because of being inactive for 15 minutes? Try clicking 'remember be' check box after you log in next time and that fixes your 'problem'.

 

[cymrych]

Not sure what was going on, but I keep that checked. Think it may have just been Saturday evening server log-jam, probably on my provider's end. Small-ish company, that gives me issues every now and again.

Phantom - I'm gonna just put her on the pad facing 90 degrees for now, to work on an idealized ascent hitting the flight markers in the documentation. I'll post the numbers for you to plug in once I work out the modest errors. Then, hopefully we can figure out the roll thing with multistage.

Meanwhile, in the UAP, try this. It works pretty well, but is a bit ham-fisted to get what we need. Anyhow, it's my best ascent profile at the moment, but I know there is some mose dV to be had.

Set up UAP with 2 Lift-Off sequences. In the first, change Heading to 90, Target pitch to 35, Target duration to 65.97, and Off-set (or Delay, cannot remember what they termed it; bottomest variable you can change on the left!) duration to 3. In the sencond sequence, set Heading to 91, Target pitch to 12.44, Target duration to 228, and Off-set to 0. Then, wait for a nice round number on the mission clock to hit go! Don't forget to pop the Fairings at T+295. Once EDS fires, aim for about a 10-12 degree AoA (I'm still optimizing this phase) and just keep an eye on the Apa as it grows. I found it's really easy to get an orbit that's 600plus x 100km if you overshoot!

Let me know how it flies.

 

[PhantomCruiser]

I think we're getting close, really close... The latest guidance file I've got for us drops out at the end of the burn with an altitude of about 142Km, an ApA of 199 with an ApT of 160 seconds.
Burning the EDS with an attitude of 18 degrees or so I circularized at 218Km with 47.8% fuel remaining. I managed to run this several times and the returns are pretty consistant, so like I said, we are getting close.:thumbup:

I've been dumping the fairing at T+165 or so, but the guidance file will let you do this whenever we want...
The SRB's jettison whenever there empty, but I put a Playsound at T+127 for the separation audio.

 

[Columbia42]

I recently figured out how to simulate the change from sea level thrust to vacuum thrust in multistage2. It involves some changes to both the .ini and guidance file. If you want I can post them.

 

[PhantomCruiser]

Sure, we can see how the performance changes.

 

[cymrych]

Columbia42- That would be awesome! Thanks. Just for my own future reference, how do you build in this change? Is it a linear function of altitude, of a non-linear function of exterior air pressure?

Phantom- Yep, you're getting close. I'm using your numbers as a template, and slowly refining it down to hit those flight data markers. Got SRB sep at about 37.3, just a little high but getting closer.

Quick note: You might want to change your test heading for 90 degrees to maximize the addition of the earth's rotation to the deltaV. You can't really lower your inclination below your present latitude value, so it's cheapest fuel-wise to just wait the quarter orbit til the ship gets closer to the equator.

 

[PhantomCruiser]

I'd thought about that, but I'd played around in AMSO (and payed more attention) to see how they launch the SaturnV. They launched on a heading on 73 degrees, at least for Apollo 11. I'll see how she rolls (so to speak) on a 90 degree. Also I was looking to speed up the roll seqence a bit, if I can do so without an uncontrollable craft.

 

[Columbia42]

Here are the parameters for sea-level vs. vacuum thrust:

For the guidance file (core stage):
Sonce the sea level thrust of the the core stage is about 88% of its vacuum thrust, just put in a command like this: 1=engine(100,88,0.1) to simulate sea level thrust. (The engines have to reach 100% before this command is initiated otherwise the boosters will not activate). Then put a command like this 1.1=engine(88,100,183) to simulate the change in thrust. (It'll take about 185 seconds from launch before the rocket is in a vacuum).

For the .ini file (SRBs):
Since the SRBs' thrust is not regulated by the guidance file, their thrust variance requires that you add a setting called thrust curve. It looks like this:

CURVE_1=(0,91)
CURVE_2=(121.6,100)

The first number is the time value and the second is the thrust value at that specific time. There can be a maximum of 10 settings. The multistage2 module automatically varies the thrust gradually from one setting to the other throughout the flight.

I will attach my modified version of the guidance file in my next post.

 

[Columbia42]

Here's the modified file:View attachment 4353

 

[PhantomCruiser]

I put the Curve stats in the AresV configuration under the Booster_2 and _3 listings (mesh name of AresVsrb1 and 2). Something is off with the guidance file though. I repeatedly end up below 100K altitude at burnout. Also the SRB's separate one after another rather than simultaniously (because of the two commands at 121.6 pehaps?). I'd just let them separate on their own after their fuel is expended...
I have made the roll program a bit more agressive (without catastrophic results). Upload pending after I finish eating...