Having issues coming back from Duna with my rocket design (cannot get more than 10k dV)

[SalehRam]

Hello all,

I have been playing around in sandbox to design a rocket package that can do a 2 way trip to Duna, but I got some issues with the design:

1. Could not get past 10km/s dV,
   
2. Fuel does not cover the whole way back from Duna, and it runs out on the burn to make the orbit around Kerbin right after reentry to its SOI
   

I have attached 2 designs I have been trying to tweak, I would really appreciate some input on how to improve the dV and what is the correct ascend burn profile out of Duna, as I feel the issue is there...

Gemini IV (https://drive.google.com/file/d/0B4vxG7-rvmD2Uk02aVluUlcwd3c/view?usp=sharing)

Gemini III (https://drive.google.com/file/d/0B4vxG7-rvmD2VnI1YWZOZFZTMkk/view?usp=sharing)

I am using 2 mods on this design, RemoteTech, and TAC Life Support, so please keep that in mind while checking the craft files.

Also, below are screenshots for the 2 crafts with the KER window open to show the stats.

The middle engine is a Rhino.

Edited October 5, 2015 by SalehRam

[Alshain]

I'm a little confused about which parts are going to Duna. Are the boosters with the aerospikes for the return trip? What about Stage 5? This seems a bit large for Duna to me, but admittedly I haven't been there since 0.90. Let me check and see if I have any Duna screenshots.

But off the top of my head, you could add another of those aerospike boosters and asparagus stage them. Also, design your lander to detach from the drive section if you can (sort of Apollo style) and redock in orbit. While that doesn't really make as much sense for the Mun and Minmus, Fuel for the return trip from Duna is heavy, as are those nuclear engines. No point in dragging that all the way down to the surface only to lift it back up again.

EDIT: Ok, here is one of my Duna missions, I couldn't find the full lifter, but that should be simple enough (if I recall, it didn't even use the big parts, just the 2.5m)

This was in a previous version, but it shouldn't have changed that much, except now that Engineers can reset experiments, I wouldn't take that many copies, or the science lab. The lab stayed in orbit with the nukes and the big tanks (I can't remember why there was a docking port on the bottom of that), the second image shows the lander only. Those engines got a nerf too, so the exact design won't work but it should give you some ideas how to manage your fuel by taking only what you need to the surface on the lander.

Edited October 4, 2015 by Alshain

[SpeedDaemon]

For a relatively large lander like that, you might be better off getting the lander to LKO with one launch, then getting a more efficient transfer stage into LKO with a second launch (and maybe more fuel with a third). If you have to dock to the transfer stage anyway, it also means you can use it to get back home, which means you don't need to haul that engine and fuel (and long-term life support equipment, since you're doing that) down to the surface of Duna and back up.

[FancyMouse]

(your DL link got reversed)

Duna round trip shouldn't need 10k dV. Add some chutes to a Mun-landing capable craft and you'll be able to come back from Duna. RT shouldn't matter if it's manned mission. TAC-LS matters in another way, it only adds linearly scaling mass (scaling to your travel time and # of crew) and doesn't affect dV by itself.

For craft optimization:

1. ^^ what above says, employ a landing module design. Lander only needs to carry the food needed for land+do science+lift off, which should be minimal. Also landing-module can also just have at most 2000m/s dV for itself, because the returning fuel is in orbit.

2. Ditch the mk1-2 lander. It's damn heavy. I use two lander cans for my lander module (two advantages - I can store multiple copies of science like surface samples and science bay; and minimize crew to reduce food I need to carry). Then you should be able to ditch the aerospikes and use a single LV-909 for the landing-module engine. Also ditch the heat shield. You're not returning from Jool nor entering Eve.

3. Poodle should be enough for transfering to Duna. No need for nukes. Lower-stage optimization follows.

By those optimization, my 2-kerbal Duna lander is 180t (comparing to your 720+t for 3 kerbals), and I didn't even optimize the lower stages (which I could have). KER shows just 6900m/s dV, but I know it's incorrect because of docking ports and it doesn't know when I'll undock the landing module. I estimate the actual dV usage is around 7750m/s

And yeah I'm also confused by further looking at your rocket - it sounds that KR-2L itself should be able to carry you to Duna, so then what do nukes do? And if you're unable to return from Duna with 10km/s, could you share how you use your dV budget? It sounds you're not flying right somewhere.

- - - Updated - - -

"and it runs out on the burn to make the orbit around Kerbin right after reentry to its SOI"

um ok I probably know your problem - you didn't aerocapture? Both Duna and Kerbin have atmosphere for you to brake and capture. You don't need to waste your fuel to do that. And don't be afraid of heats - entry speed to Duna is even slower than Kerbin low-orbital speed, so that is harmless. Entry speed to Kerbin from Duna is comparable to a Minmus return and should not be faster than 3.5km/s, which shouldn't be too much a problem with a right attitude (but you should test). My experience tells me I need to seriously consider heating in craft design only when entry speed is above 4km/s.

[KerikBalm]

You seem to have LV-Ns feeding from LFO tanks.... have you removed the oxidizer?

(haven't loaded the craft file, I don't have the mods)

Anyway... Duna is a good place to employ an "Apollo style" orbital rendevous at the target between a dedicated lander, and a return vehicle.

Its a few hundred m/s more dV to get back to kerbin from Duna orbit than it is to get back to Kerbin from Mun/Minmus orbit.

And then its hundreds more m/s just to get to orbit again*... so you've got to carry more fuel, farther (farther in the sense of more dV).

*Although, on duna, the difference is not so different from Mun, since the atmosphere will give you a nearly "free" landing, that costs ~600 m/s on Mun.

But really... outside of kerbin's SOI, if you land anywhere, you should be using a dedicated lander and orbital rendevous (docking)

" except now that Engineers can reset experiments"

Engineers repack chutes, repair wheels, repair struts, and make the drills mine much faster.

Scientists are the ones that can reset the experiments (goo pod and mat bay)

Edited October 5, 2015 by KerikBalm

[Empiro]

There's a bunch of things that can be improved. Individually, each won't make a huge difference, but but them together, and you'll be able to accomplish your mission (and more) with a rocket half the size.

1) You don't need 10k delta-V for a Duna trip. I don't know where you're spending it, but 10k is more than enough for a Duna/Ike landing and return.

2) You can save hauling extra mass from Duna surface into orbit if you leave the fuel for Kerbin return in orbit around Duna and dock with it.

3) There's probably something up with your nuke engine stage. Maybe you're adding oxidizer to those takes, but it should have way more than 2km/s delta-V. In general, if your stage doesn't have 4000+ m/s delta-V, you're probably better off using something besides nuke engines.

4) Way too much "stuff" in your upper stage. It weighs almost 22 tons empty! You don't need so many parachutes, an SAS unit, or whatever else you have. Keep it lean in the upper stage, and everything else can be smaller.

[Alshain]

" except now that Engineers can reset experiments"

Engineers repack chutes, repair wheels, repair struts, and make the drills mine much faster.

Scientists are the ones that can reset the experiments (goo pod and mat bay)

right, thats what I meant

[KerikBalm]

dV for Duna, conservative:

3,800 m/s to get to orbit (I think it can be done for as little as 3.2k .... 3.4k is a good general figure though

+

1,200 to get to Duna (can be done with <1050)

+

200 m/s to land (can be done for much less, just direct aerocapture and then pulse the engines before touchdown with chutes)

+

1,800 to get to Duna orbit (can be done, with rendevous for 1,400)

+

700 to return to Kerbin (can be done for ~580)

=

7.7 km/sec

.... and that is very conservative, and a direct ascent, where you don't need to worry about aerobraking enough to capture, but not enough that your entire craft lands on Duna - if you do orbital rendevous and just send down a lander, then you need to be a bit more precise with your aerobraking so that the stuff for the return stays in orbit.

In the case of orbital rendevous its a little harder to calculate the dV, you can remove 1,800 from that, but calculate the dV for each stage except the return with the mass of the lander, and calculate the dv for the return stage without the lander.

That gets you to 5.9 km/s for the main craft for the entire mission, with good margins at each stage.

A narrower margin would have 5.2 km/s for the main craft.

But its much easier to loft things to LKO, and if you've depleted a bit of the fuel in the next stage, its easy to remedy at that point with a refueling mission...

Then your craft can leave LKO with only 1.8km/sec for the main craft, and ~1.4 km/sec for the lander.... (though this means from LKO - the lander, including its time as part of the main craft, has 2.45 km/sec)

Which is far far less than 10km/sec...

[SalehRam]

Thanks a lot for your valuable input guys...

Here is a brief for the way I am flying this rocket

Stage 9, 6, 5:

Take off from Kerbin, go into LKO, and perform transfer burn. By the end of the burn, stage 5 (Rhino engine) is completely used all the fuel.

Stage 4 (Nukes):

In Duna SOI, burn retrograde to get an aero-capture trajectory on Duna surface. By the end of this stage, some little fuel remains, which I use for slowing down in Duna atmosphere.

Stage 3:

Take off from Duna, go in LDO, and perform transfer burn, then in Kerbin SOI try to perform burn for aero-capture trajectory. Fuel runs out in the middle of the burn for aero-capture in Kerbin SOI

---

The only thing I am doing, which I doubt is the problem, is I am establishing an orbit after take off rather than keep burning to escape the SOI, I'm not sure how that efficient or if I am doing it right, but I time my take off from Kerbin with the best transfer window using the alarm clock mod, and same for the take off from Duna, I time it with the best transfer window to kerbin as well...

The rocket itself while taking off from Kerbin is in good controlling status, so I can do an ascend at 60-50 degrees while low in atmosphere, and I do the same ascend profile on Duna at 60 degrees...

I'm a little confused about which parts are going to Duna. Are the boosters with the aerospikes for the return trip? What about Stage 5?

Yes, the stage with the aerospikes is the lander stage, 3 engines. Stage 5 is the Rhino engine when it keeps running alone after separating the 6 surrounding tanks off it.

For the fuel tanks in the LV-Ns stage, there is no oxidizer, it is just liquid fuel...

2 parts lander package is something I would consider now, but I believe it will need lots of working in VAB and some testing as well...

But in general I guess my problem is in the lower stages below the lander part, where all my fuel is getting consumed so quick...

The reason I did this design, because in 0.90 and before I used to go to Duna with something like this but smaller and it used to work, but now I am not sure what is the major change that is causing the old designs to be such inefficient for me like this... It could be me forgetting how I used to fly them or something else...

What I could do, is I start with small design and scale it slowly to see where is efficiency drop happening.

Edited October 5, 2015 by SalehRam

[KerikBalm]

Many of the Isp's were nerfed in 1.0

The aero changes meant less drag losses to get to LKO, so it was sort of even in terms of difficulty in getting to LKO (maybe a bit easier now).

The result is that it is however harder to go to places beyond LKO.

And for those of us that used air breathing launchers, everything was made more difficult, which is good.

It was way way too easy to make a single stage to laythe and back vehicle before.

I still wonder what you're doing that 10km dV is not enough.

7.7 should be plenty...

Do you aerobrake or not?

Do you pitch over more as you get higher, or do you hold 60 degrees past 20/30 km?

How much dV do you have left after attaining LKO?

It should be over 6km/s (if you started with just over 10km/s)... and you should only need 3.9km to do the rest of the mission, without any orbital rendevous..

[FancyMouse]

Stage 4 (Nukes):

In Duna SOI, burn retrograde to get an aero-capture trajectory on Duna surface. By the end of this stage, some little fuel remains, which I use for slowing down in Duna atmosphere.

then in Kerbin SOI try to perform burn for aero-capture trajectory.

These are wrong. You don't burn inside SoI to get aerocapture. You burn in the middle of the transfer (and nothing pro/retro, only a combination of other 4 directions). This should not need more than even 30m/s for good orbits like Duna. If you burn inside SoI, it gets larger. It should still stay within 100~200m/s if you burn radial/anti-radial, but you burn retro - that's not good.

The only thing I am doing, which I doubt is the problem, is I am establishing an orbit after take off rather than keep burning to escape the SOI

No, what you do is correct.

Edited October 5, 2015 by FancyMouse
clarify myself

[SalehRam]

These are wrong. You don't burn inside SoI to get aerocapture. You burn in the middle of the transfer (and nothing pro/retro, only a combination of other 4 directions). This should not need more than even 30m/s for good orbits like Duna. If you burn inside SoI, it gets larger. It should still stay within 100~200m/s if you burn radial/anti-radial, but you burn retro - that's not good.

This I have never tried before, thanks for sharing this.

I'll test it at home and see how it goes...

And for KerikBalm, Yes I do try to perform aero-brake when landing on Duna, but as mentioned above, it seems that I was doing it wrong thus burning lots of fuel.

I will check the above method today and I feel it will go very fine.

[SalehRam]

Much love to you guys

I managed to return to Kerbin so easy after making sure I get an aero-brake trajectory in mid-transfer, turned out the only thing I never thought was wrong is in fact completely wrong...

I will re-work on the rocket, making it smaller as now I can properly do two way trips without worrying much about dV as I used to before...

Appreciate all your comments and notes on this!

[OhioBob]

and what is the correct ascend burn profile out of Duna

I generally burn straight up until I a reach a vertical velocity of 50 m/s, then pitch over about 10-15 degrees (to a pitch of +75-80^o). I then decrease pitch as needed to gradually lower my flight path angle. I want my flight path angle to be about +45^o when I reach an altitude of 7 km, and about +20-25^o when I reach an altitude of 20 km. Engine cutoff typically occurs at an altitude of about 20 km when the apoapsis altitude reaches 55-60 km. (You can burn longer if you want a higher orbit.) I then finish off orbit insertion with a burn at apoapsis. This profile should put you in orbit for under 1400 m/s.

Edited October 5, 2015 by OhioBob

[KerikBalm]

In fact, its even cheaper if you make the course change even earlier*, but the problem is often precision, when you first get your encounter, you sould fine tune it as soon as possible, to get a low perapsis (or even a perapsis that directly intersects)... but at that distance, the line is quite jumpy - you can try using the thrust limiter on your engine, or puffs of RCS, or a tiny ion engine to try and get more precision...

Then midway through, refine it again to get a PE between 10-30km, and then when you cross the SOI into duna, you can fine tune that to the hieght you need to aerocapture.

*in the case of plane changes sometimes you want to wait until close to the An/Dn.

You can't really change your inclination if you are equidistant from the An & Dn, and Longitude of ascending node and argument or perapsis are a bit more comlicated too.

But duna's inclination is only 0.2 degrees, so in this case, you can basically ignore it.

The earlier you can refine it, the better... but I often wait until half way anyway, because the manuevers are so small, its like doing an 8 m/s manuever instead of a 4 m/s maneuver

[OhioBob]

If I need to perform a mid-course correction, I usually like to do it when my spacecraft is 90 degrees from planet intercept. At this location we get the most effective use of normal/anti-normal velocity changes if needed. A MCC at the 90^o location and another just inside SOI usually works pretty well for me.