[It's Back!] Another Voyage Ares Mission, and A KSP'ers Crash Course in Realism Overhaul.

[TiktaalikDreaming]

My flights keep doing daft things at about 100m off the ground.  The SAS just goes nuts and decides up is down or retro is sideways.  At first I thought it was bad interactions with the chutes, but I made sure to cut the chutes this time.  If it happened higher up, or just happened once, I think I could recover.  But it's like I have to pass through a field of stupid to get to the ground, and with the ground that close, the only survival method is throttle-up-abort, and then I have to come back through the stupid field.

nerdraaaaaaaaage

 

[TiktaalikDreaming]

My craft still wants to turn upside down at ~100m, but by expecting that, and keeping the ballute I've managed a landing with the craft the pointy end upwards.

 

And the lab decoupler CLS config isn't working.  So no moving kerbanauts to the lab, or an airlock.  :-(  I'll have a go at fixing that.

Gotta go do that going-to-work thing.  But I'll see if I can get onto the surface using the airbrake alternative next (after fixing that decoupler).

PS: Plenty of fuel in the descent stage without extra tanks now that the descent engine thrust isn't lame.  The remainder shown is after quite some hovering balance work going downslope even.

[MaxL_1023]

1 hour ago, TiktaalikDreaming said:

My craft still wants to turn upside down at ~100m, but by expecting that, and keeping the ballute I've managed a landing with the craft the pointy end upwards.

 

And the lab decoupler CLS config isn't working.  So no moving kerbanauts to the lab, or an airlock.  :-(  I'll have a go at fixing that.

Gotta go do that going-to-work thing.  But I'll see if I can get onto the surface using the airbrake alternative next (after fixing that decoupler).

PS: Plenty of fuel in the descent stage without extra tanks now that the descent engine thrust isn't lame.  The remainder shown is after quite some hovering balance work going downslope even.

I notice that if you travel slower than about 1 m/s the navball loses prograde/retrograde - could this be happening? 

[Nittany Tiger]

4 hours ago, TiktaalikDreaming said:

My craft still wants to turn upside down at ~100m, but by expecting that, and keeping the ballute I've managed a landing with the craft the pointy end upwards.

 

And the lab decoupler CLS config isn't working.  So no moving kerbanauts to the lab, or an airlock.  :-(  I'll have a go at fixing that.

Gotta go do that going-to-work thing.  But I'll see if I can get onto the surface using the airbrake alternative next (after fixing that decoupler).

PS: Plenty of fuel in the descent stage without extra tanks now that the descent engine thrust isn't lame.  The remainder shown is after quite some hovering balance work going downslope even.

I started testing the alternative airbrake today and I found that my craft flip and roll like mad in the upper Martian atmosphere at AOAs of 15 degrees or more (which I do for a skip re-entry).  I had to abort the descent before even reaching descent stage ignition.  I'm trying to solve it by placing the airbrake shell under the standard fairing shell and then jettisoning the shell, but I have yet to see the results.

Hopefully, we'll have this issue nailed soon so I can move on to the next part of my Voyage recreation.  I want to nail down the RCS balance and in-orbit assembly.

[TiktaalikDreaming]

3 hours ago, MaxL_1023 said:

I notice that if you travel slower than about 1 m/s the navball loses prograde/retrograde - could this be happening? 

I don't think so.  It seems to happen whether I have SAS set to retrograde or radial out, which shouldn't be affected by losing the definition of prograde etc.  It's like hitting a trampoline.  Coming down nice and gentle, and suddenly the craft is doing cartwheels, usually with about 1(local) G of TWR pointing the wrong way.  If no-one else sees this, I'll put it down to "weird sh!t with my install/bunch of mods". 

And I tried getting MJ to land, but I couldn't tell if it was experiencing the same issue, as MJ is still of the opinion that throttling the descent engines is something that can happen after passing through a few km of regolith.

15 minutes ago, Nittany Tiger said:

I started testing the alternative airbrake today and I found that my craft flip and roll like mad in the upper Martian atmosphere at AOAs of 15 degrees or more (which I do for a skip re-entry). 

Hopefully, we'll have this issue nailed soon so I can move on to the next part of my Voyage recreation.  I want to nail down the RCS balance and in-orbit assembly.

I was getting some strange behaviour with the airbrakes as well.  I need to set the CoL I think.  I thought the control surface modules would do that.  But the part's root is at the hinge, so the drag will centre on that.  That's not so good for either state.  I'll have to check how the stock airbrakes handle it.  Maybe I shouldn't have a collider for the airbrake surface.  Maybe I should have a CoL adjustment.  Not sure.

Also thinking (if I can keep the collider) that I could add a generic animation that has them fold further out and provide ramps.  Especially as I notice Mars' gravity is just enough to make Kerbal rocket packs useless.  And full scale, it's big compared to the little green men

[Nittany Tiger]

On second thought, I'm still getting the tumbling even with the heat shield fairings on.  The whole MEM wants to dive nose-first.  I did add fuel to the MEM, so maybe it's the extra tanks making the CoM crazy.

I've removed them and went back to just the airbrake fairings and doing another test.  Hopefully it doesn't flip this time.

EDIT: After another test, it actually seems like an issue with the airbrakes.  The MEM wants to fall nose-first when they're closed, but they'll stabilize the craft end-first when they're open.

Good news is that the uprated engine seems to make a huge difference with easing landing difficulty.

Edited September 19, 2016 by Nittany Tiger

[TiktaalikDreaming]

4 hours ago, Nittany Tiger said:

On second thought, I'm still getting the tumbling even with the heat shield fairings on.  The whole MEM wants to dive nose-first.  I did add fuel to the MEM, so maybe it's the extra tanks making the CoM crazy.

I've removed them and went back to just the airbrake fairings and doing another test.  Hopefully it doesn't flip this time.

EDIT: After another test, it actually seems like an issue with the airbrakes.  The MEM wants to fall nose-first when they're closed, but they'll stabilize the craft end-first when they're open.

Good news is that the uprated engine seems to make a huge difference with easing landing difficulty.

I was getting the same, re: closed airbrakes=prograde, open=retrograde.  I think they have a size, and thus a drag cube.  And that drag is applied as if the centre is at the part's root, which is the middle of the hinge. I have to figure how to alter that drag centre in a way that's appropriate for an airbrake (ie, it has to move).

[Nittany Tiger]

4 minutes ago, TiktaalikDreaming said:

I was getting the same, re: closed airbrakes=prograde, open=retrograde.  I think they have a size, and thus a drag cube.  And that drag is applied as if the centre is at the part's root, which is the middle of the hinge. I have to figure how to alter that drag centre in a way that's appropriate for an airbrake (ie, it has to move).

Yeah, and the funny thing is that even when I moved the airbrakes into the MEM and put the standard fairings over them, they still drag the MEM around prograde.  I don't think that should happen with FAR, but maybe I'm wrong.

[TiktaalikDreaming]

The airbrakes seem OK for CoM and CoL in the VAB display, although I'm unsure if FAR changes the VAB display along with behaviour though, so I'm not sure if what shows relates to what behaviour to expect.

BUT, the descent base definitely has issues.  The centre of lift appears well below the bottom of the craft.  I'm surprised all MEMs using that aren't facing forward.

EG:

 

I've done a couple of flights using the aero overlay.  Again, I don't know if FAR changes this or if it's just showing stock forces.

But, looks like the aero forces are applied at the hinge point of the airbrakes.  But, the big difference looks to be they provide next to zero forces when not deployed, but also seem to prevent the main body providing any drag.  So the craft is experiencing aero as if it was a flat pancake shaped like the base, but with a centre of mass higher up.  FIIK what causes this, or if I'm really seeing what I seem to be seeing.  It just looks like the closed airbrakes forma protective shell around the craft that prevents any aero forces from covered areas, but doesn't provide any of it's own.

I'm going to go with "deploy airbrakes early".

Edited September 19, 2016 by TiktaalikDreaming

[Nittany Tiger]

I noticed the strange aero drag on the airbrakes as well when using aero overlays.

Good news is that I finally achieved a soft landing on Mars.  No more martian pancakes.  I thinking adding some fuel and firing the descent engine around 1min 30sec before impact seems to do it.

I'll upload pics once Steam cloud works again.

EDIT: Also added a to-do list on the front page for anyone that's interested in what still needs to be done before I do the mission.

 

Edited September 19, 2016 by Nittany Tiger

[Nittany Tiger]

More landing tests today.  Now seeing how well I can hit Mangala Valles.  First few tests were to just verify that I could land the MEM consistently.  Looks like firing the descent engine 1 min 30 sec before impact gives me enough descent fuel to work with on final approach.  Anything earlier and I run out too soon.

Found the airbrakes make getting onto the surface impossible...

...so I made them jettisonable.  Now they get jettisoned before landing.

Valentina gets to simulate a walk and plant the flag, representing York in the real attempt.

The only issue left is to make sure the descent engine heat shield doesn't destroy the descent engine, which happens rarely but it does happen.  The last time it happened, I did abort and rescue the crew, but I don't want to fly all the way to Mars to have the descent engine get destroyed.  I'll likely place some small separation motors on the descent engine heat shield to make sure it doesn't hit the engine on jettison.  Descent testing should be done after that.  I just need to do another test launch of the Ares with the new MEM to make sure it isn't too heavy now with the added fuel, then I can get to RCS placement and rendezvous testing.

Edited September 21, 2016 by Nittany Tiger

[TiktaalikDreaming]

That engine cover has been annoying me for ages.  There's something odd about how the ModuleDecouple functions.  There's plenty of ejection force, but it decouples like there isn't any.  I've had the same issue with other jettisonable parts.  But then others, that I'd expect issues with, like the half cone shell pieces, jettison really nicely.  I'll poke around the forums and see if others have had similar issues.

[Dafni]

1 hour ago, Nittany Tiger said:

Glorious, just glorious. I imagine Phil Stone peering out at her through the hatch.

[TiktaalikDreaming]

I will get to the bottom of this forceless decouple.

Here's my test rig, in stock, because this isn't an RO specific thing.  And in 1.2pre, as it's an issue there as well.

What it's not;

too low ejectionForce, because cranking a 200kg piece with 10tonnes of force should make it move.

Not that the stack nodes were both negativeY, I've adjusted the centre point to a point part way between the stack nodes.  (this change will roll into the next update, as it's a *should have been fixed anyway* fix.

Not the only .15m gap between the stack nodes, I tried with artificially increased nodes .6m apart.

Not that the collider was too big, I shrunk it a bit to test.  (I will double check the plainly confusing, overly complex descent base collider soon)

Not interaction with ablator resource (removed), or in fact anything ablatorish (test removal of the modules all together).

Testrig

 

[Nittany Tiger]

I always thought maybe atmospheric drag was pushing the cap back into the engine.  I would think the decoupler force would be enough to give it a forward push against drag, though.

[TiktaalikDreaming]

1 hour ago, Nittany Tiger said:

I always thought maybe atmospheric drag was pushing the cap back into the engine.  I would think the decoupler force would be enough to give it a forward push against drag, though.

That was my first thought ages back, but an increase in force should improve that.  I've also been staging mine to match the ballute deployment, so it *should* coincide with a chunk of Gs helping it along.  But it still got stuck.  I think I've found the issue.  The base collider is a set of six (duplicated) mesh colliders arranged in a circle so they have a gap.  But even though the whole purpose of that is to compensate for not being able to have concave shapes, I completely forgot about that when making the inside edge.  The inside of the circle was matching the actual circle, and formed a hexagon smaller than the gap when forced into convex shapes.

So, in blender it looked like;

But in Unity with the colliders showing, it looks like;

Fixin and testing now.

 

And the engine cover cannon works nicely.  I might tone that 10000 ejection Force down a bit.  http://imgur.com/a/FNLXP

 

The main change that will make the engine cover function is this file; https://www.dropbox.com/s/ws1de1nme5zl0a7/NewModel.mu?dl=0

It's the model for the descent base, it overwrites the one in NAR_MEM/Parts/MEM-DescentBase/

The various tweaks will eventually be rolled into a mod update, probably once 1.2 is a real thing.  This is getting a bit piece meal though.  :-)

Edited September 21, 2016 by TiktaalikDreaming
Tested

[TiktaalikDreaming]

On airbrakes; I had a go at adding a generic animation where they would fold down as a kind of spread out ramp. And they told out nicely and my first guess at angle turned out pretty close to touching the ground.  But once the animation is done the airbrake module tales over again and sets the angle back. I might be able to parent the airbrake to an animated empty and get it working that way. I'll have to try after work.

[Mad Rocket Scientist]

Subbed!  I really enjoyed that book too.

On 9/14/2016 at 10:33 AM, Nittany Tiger said:

Therefore, trying to do a Voyage-style descent in the base MEM results in death and destruction. 

Just blame the polynomials.  

[TiktaalikDreaming]

13 hours ago, TiktaalikDreaming said:

On airbrakes; I had a go at adding a generic animation where they would fold down as a kind of spread out ramp. And they told out nicely and my first guess at angle turned out pretty close to touching the ground.  But once the animation is done the airbrake module tales over again and sets the angle back. I might be able to parent the airbrake to an animated empty and get it working that way. I'll have to try after work.

Mixing a generic animation module and the ModuleAeroSurface doesn't work so well.  With messing around, you can get things deployed at different combinations, but nothing works to start with without right click actions, and usually not until you've deployed, retracted, and deployed again.  So I'm thinking of just making the airbrakes deploy much further, as their actually fairly easy to limit to a smaller angle after attached.

[Nittany Tiger]

On 9/21/2016 at 7:05 PM, Mad Rocket Scientist said:

Subbed!  I really enjoyed that book too.

Just blame the polynomials.  

Curse those Taylor series!  I knew it.

Actually, with the upgraded descent engine and more fuel, I probably could do a Voyage-style descent now.  Right now, I'm going with airbrake-assisted descent.

Time for rendezvous testing.  Let's see if I can fly the Saturn VB to the propulsion module.  I put the new MEM in the Saturn VB and put a propulsion module in a 500km x 500km orbit at about 37 degrees inclination.

Since the famous four are either respawning or orbiting Mars, I took this chance to create the prime Ares crew.  One of the mods decided to make Phil Stone and Ralph Gershon Soviet, but I won't tell Natalie about that.

Using MechJeb's rendezvous planner to get some info, I launch...

...and the rocket fails.  SLA panels started to cluster and clip and it eventually pitched my rocket upward, so I aborted.  The rocket became a very expensive firework after that point.

Crew survived at least.  Good thing this was a simulation.

Turns out time warp was causing some crazy issues with rocket stability.  The rocket would wobble and flex like mad after time warping, which I had to do to get under the orbit of the propulsion module.  I also had to do some severe maneuvers during launch to get the correct azimuth the first time.  Either of those could have resulted in the kraken coming out to play.  Therefore, I added FASA launch towers to the rocket and it fixed the flexing.  Second launch was a lot more stable.

I'm sure you notice the missing fin on the MS-IC, which was destroyed during contact with the launch tower.  I'm going to fix that.  Luckily, it didn't affect the launch in any significant way.

My final orbit here was within 0.03 degrees of the target orbit, so great second attempt I guess.

The RCS I currently have on the command stack works well, though it does need tweaking and it gulps fuel, so I do need to add more RCS fuel to the Ares Mission Module.  The OMM works well and I feel for now it has enough delta V to get to the Ares propulsion stack.

Unfortunately, the game glitched on calculating close approaches, and when I set up a maneuver node, it miscalculated a close approach, causing me to miss the stack by quite a few kilometers after this burn.  Luckily, I quicksaved after orbital insertion, so I'll try again later.  Maybe the stock game maneuver node system doesn't do well with RSS sizes or it was a random bug.

Edited September 24, 2016 by Nittany Tiger

[Nittany Tiger]

More playtesting and more images.  Second time's the charm with rendezvous.  Not sure why the first rendezvous was off by so much (100km or more IIRC).  Second attempt got me within 1.5 km of the stack.

Docking was a tedious nightmare after this point.  Attitude control with Smart ASS ate up RCS until I had to cheat in some more.  The Ares Command Stack turned very slowly and was frequently victim to over-corrections by Smart ASS, which ate up more fuel and took up time watching the command module oscillate.  I plan on adding reaction wheel to help solve this along with stronger RCS ports and more fuel.  I also plan on trying to dock without Smart ASS so I can conserve RCS fuel.

Using the medium-sized docking ports meant docking had to be very precise with the large parts, but I don't plan on upsizing the ports since I need them all the same size for the transposing and docking.

I did manage to dock both craft together.  Since I was on the night side of Earth, I had to use the Docking Port Alignment Indicator, but it worked beautifully.  For the first time in this run (and ever for me not counting stock KSP runs), I have assembled the Ares.

Unfortunately, a new problem crept up after docking.  The RCS on the Ares command stack doesn't work.  Only the RCS on the propulsion stack works (the RCS on the APS modules of the MS-IVB).  I'm not sure why this is at the moment, but it means I have little to no attitude control of the Ares craft.  Given that the command stack alone was sluggish, I'll need to add more attitude control on the propulsion stack as it is, maybe in the form of reaction wheels.

In conclusion, rendezvous and docking work, but both craft need tweaks to help conserve RCS and produce faster and more responsive attitude control of the command stack and assembled ship.  RCS port placement needs to be tweaked as well so I don't get unwanted rotations when I try to translate and dock or burn up RCS trying to keep the craft aligned.

Edited September 26, 2016 by Nittany Tiger

[Nittany Tiger]

Minor update.  The RCS not working on docking was a temporary bug thank goodness.  Unfortunately, another bug with Action Groups Extended cropped up, and when I tried to test docking maneuvers, I accidentally staged the MEM shroud.  That ended the testing fast, but I got enough of an idea from the test of things I needed to fix.

So far, I've changed the RCS ports of the Ares Command Module (ACM) to 2 kN linear thrusters and added a reaction wheel set.

Another major issue was the docking port causing the Apollo CSM to get decoupled with a velocity of 2 m/s relative to the rest of the ship, which the RCS on the CSM had a lot of trouble cancelling.  This meant that the CSM would end up 100 m ahead of the Ares Mission Module (AMM) before I could cancel the relative velocity.

This was fixed by putting a procedural separator between the front AMM docking port and CSM.  You may also notice the APM ETs floating dangerously close to the ship.  That's an opposite issue with those radial decouplers as those are too weak.  Separating the ETs doesn't push the tanks out far enough.  Granted, I haven't decoupled empty tanks, so I might empty the ETs and test the separation before I change the decouplers, but they probably need to be strengthened nonetheless.

So these fixes along with solar panel realignment on the Ares Propulsion Module (APM) are all that needs to be done right now.  Nothing really major.  On that note, I leave with this pic of a test of the APM engines, giving an idea of what the ship will look like boosting its way to Venus whenever I attempt the full mission hopefully soon.

Edited October 1, 2016 by Nittany Tiger

[Nittany Tiger]

Another rendezvous and docking test.  I think I have hitting a target orbit down and rendezvous down.

Attitude control with the full stack is still very sluggish, but at least I didn't run out of RCS fuel trying to dock or going through the different configurations.

Took the MEM out for a spin.

So, yeah, configuration tests went well except for one Kraken strike that happened during a test of the MS-II

I still don't know what caused my ship to flex like that, but I was able to recover from it.  Still, enjoy the pretzel ship.

I've added more RCS ports to the AMM and MS-II stage in hopes that it will help with attitude control.  Reaction wheels work great for roll authority.  I think I'm pretty much done with testing here and there's not much left to test, so I'll be ready for the full mission attempt soon.

My next focus is the SRB thrust curves on the Saturn VB to keep the vehicle from hitting over 40 kPa of dynamic pressure on ascent.  My SRBs separate very cleanly now, so no more worries about them taking out my F-1As on ascent.

[Nittany Tiger]

Slow progress with SRB thrust curve tweaking.  I got what I hope is a decent thrust curve.  Made the SRBs throttle back to 60% through max Q before slowly going back to about 85%.

Also downloaded kOS to automate the throttling of the F-1A engines.  That took a day of learning kOS and debugging the script I wrote.  May end up having kOS automate first-stage operations or even the entire ascent sans steering unless I want to give that a shot.

I've also noticed that G forces at certain parts of the ascent get really high, such as before SRB tail-off (4+ Gs) or before MS-IC sep (also above 4 Gs even with center engine shutdown 25 seconds prior to sep).  I know Saturn V launches never got above 3.8Gs or so.  I'm wondering if I need to adjust something or check my vessel mass, or maybe just accept this high G-loading.

[Nittany Tiger]

Update!  I've done the final test!  Descent and ascent at Mars with the NAR MEM.  Thanks to practicing earlier, it was a snap.  Nailed the landing 277 km near Mangala the first time and ascent and rendezvous with the hyperedited ACM was a snap.  I just need to remember to burn about 160 - 170 degrees or so opposite of Mangala in a 300km orbit and I should get close enough to it when I land.

This is the last major test I needed to do, so I'm more or ready to attempt the mission.  Just have to do some minor additions, set my game to the appropriate date, and figure out how to use Flyby finder to set up the flight, and it should be ready to go.  I still want to do some minor things, mainly added a Venus probe and Mars rover, expand my kOS script, try to create an interior for the AMM, and set up the camera mod because I might make a long video of this mission.  Might also write up a mission plan, like something that looks somewhat professional detailing the rocket specifications and timing.

For the mission attempt, I might post pics here and/or stream it.

Edited November 4, 2016 by Nittany Tiger