Yes; it has an excellent thrust/weight ratio and a decent (though not great) ISP. The combination makes it really, really good.
Want a 3-component SSTO (from Kerbin)? Probodyne OKTO2 + FL-T200 tank + Rockomax 48-7S. Done.
How about an orbital craft in 0.83 tons? OKTO2 + Oscar-B + LV-1 + TR-2V (upper stage) + FL-T100 + 48-7S (lower stage)
Maybe you want a SSTO that can take a decent payload to orbit? Jumbo-64 tank + 19x 48-7S engines (using the Cubic Octagonal Strut for placement). Will carry several tons to orbit with enough left over to land itself on its tail.
I set my script to look for tank+engine combinations that maximized imparted energy to payload per total unit mass: in other words, the most efficient upper stage. I also had the requirement of a TWR > 2… on Eve (so 3.4 on Kerbin). FL-T100 + 48-7S blew everything else away. My Eve return craft has 17 of those on the upper stage.
Thanks for the heads up on the mod. With it, I was able to land on Minmus without tipping over and get back to Kerbin in one piece. Unfortunately, I was so low on fuel I decided to fall straight from Minmus to return and was going so fast when my chutes opened fully that they ripped off. I lost all my instruments, but the probe body survived and I was able to claim the science for a safe return. (I choose to count getting into Kerbin’s atmosphere in one piece as getting back to Kerbin in one piece)
Heh. I must have lost about a third of my missions from parachutes ripping off! Eve missions especially are problematic. Congrats, though!
You might consider powered landings for future missions. The fuel requirements for a final burn are fairly modest if you are careful. You only need 200ish m/s of deltaV on Kerbin and (IIRC) Laythe. Even less on Eve.
Congrats on the mission
I have just restarted career mode as I am traveling and have to use my work laptop . Is trying to do a Mun landing without solar panels even doable . I’ve got a rocket with enough DV and an ass load of batterie, just need to weigh up if loosing Jebediah is worth the risk .
Quick question for the experts. How do I keep my central engine from disengaging from the central stack and staying on the launch pad when I fire off my rockets? Asparagus staging with the basic big narrow tanks. Everything works beautifully…except the central engine stays on the pad. So once I drop the 1st pair and then the 2nd I have this lovely full tank of fuel doing nothing.
Hard to say exactly, but I’ve found that resting a rocket directly on the engines will sometimes snap the engines off. I almost lost a mission from Vall when some engines snapped off on landing (fortunately I had some backup engines… low efficiency but they got me home).
This. For rocket engines, put the towers in the same stage as your engines. For jet engines, put them in a separate stage just afterwards, so the engines have time to rev up.
Yep, and right click on the jet engine to see how many kilonewtons it’s giving you before release the clamps. Do not rely on MechJeb for this; Mech Jeb gives you the theoretical thrust of jet engines, not the actual one; If you have a 225 Kn engine and throttle it 100%, MechJeb will tell you that it’s giving you 225Kn which is false. At launch, you’ll only be getting about 112 Kn at most. For example, if your ship weighs 15 tons, you will lift off with 225Kn of thrust but crash with 112Kn.
Then divide the number of Kn by the number of tons. If you end up with a number at 10 or more, you’re good to launch.
Been seeing how far I can optimize a manned Mun mission, with return to KSC. So far, I’ve got it to under 12 tons. However, I know I can do a lot better! I think 10 tons will be easy. 8 tons might be just doable.
Yes, I did use Mechjeb extensively :).
Jeb got quite excited at that 6 G burn just 180 meters above Kerbin…
I can do it (occasionally!), but I waste a lot of delta-V doing it by hand. As I said, I use Mechjeb pretty extensively. For getting into orbit, I can usually beat Mechjeb, but for landing I don’t see how anyone can come close, either in accuracy or efficiency. Watch my video starting at 24:15. It decelerates at literally the last second at 6 gees, and only experiences a tiny bit of float at the end. All of my descents had to start much higher; I wasted fuel basically just hovering. No way could I have pulled off that particular landing by hand–it ended up with only 38 m/s of delta-V remaining.
I’ve had some success with manual “suicide burns” on my landers, with reasonable precision and efficiency. Nowhere near MechJeb levels of course, but I can pick a smaller crater on the Mun and probably land within it. My approach is thus:
From a low orbit, set your sub-orbital trajectory, with the impact point just a bit past where you want to land. (e.g. on the rim of the larger craters if you want to land in the middle.)
Place a maneuver node slightly before the impact point, but slightly after your desired landing point, and set up a retrograde burn that completely cancels your velocity. The post-burn “orbit” will be a line pointing straight down.
Note the estimated burn time on your navball, and double check it with Engineer since the built-in burn estimator can be glitchy. Let’s say this is a 60 second burn.
Start the suicide burn when you’re 60 seconds from the node. Actually I think less than 60 seconds would be optimal, since you’re decelerating hard which means you approximately double the time it takes to reach the node, and the maneuver node should leave some margin for error. So if you’re feeling ballsy, start the suicide burn at 40 or 30 seconds.
I don’t know for sure how efficient this is vs. MechJeb, but I’d WAG that I’m wasting much less than 100 m/s ΔV during the final few seconds of descent. Basically this replicates the burns in your video at 10:12 and 11:00 with reasonable efficiency, though my final descent is probably still a little sloppy.
Thanks for the tips. I’ll have to try a few experiments to see how close I can come to Mechjeb. I really wish there were a better approach to throttle control; up/down buttons are just so finicky. What I really want is a joystick with an analog throttle knob.
Here’s my latest effort. 7 tons; again a “flag and footprints” mission to the Mun and landing back at KSC. Actually, “landing” may be generous (see 26:30). But Jeb walked away, so it’s all good! I think I know how to fix the Mechjeb control problems; I’m going to see tonight if I can improve my landing a bit.
I had a craft that could almost do it in 6.5 tons. However, I’ve found that landing a FL-T100 tank with a Kerbal on top is essentially impossible, at least with an OKTO2 for control. The near-empty tank is so lightweight (lighter than Jebediah) that it acts as a parachute, and orients the craft Jebediah-first. The OKTO2 doesn’t have a huge amount of torque and so there’s no way to flip the craft back around. I’ll try some more experiments, though; maybe there’s something else I can do.
For comparison, I just tried some manual landings from a 20 km circular Munar orbit, to compare with your MJ landings. Your lander used 770 m/s to land. On my best attempt, I landed with 600 m/s – I started the 48-second final burn at 27 seconds before impact. However that was with very shallow approach, probably coming in at something like 15°. At that angle, there’s no margin for error, and it’s hard to avoid mountains or crater walls on the way down. From a steeper landing, you can pick your landing site with more precision, and have more margin for error. Even then, my landings coming down at 60° used 700-740 m/s ΔV, less than MJ.
Anyhow, I’ve completed all the Science on my first career run through. My missions were:
Several orbital and sub-orbital flights.
Manned flybys and then landings on the Mun and Minmus.
Unmanned landers on the Mun, Duna, Eve, and Laythe.
And finally, a manned flyby of Moho. It was supposed to be a landing, but my Kerbin-Eve transfer was… suboptimal. Wasted a lot of fuel on the transfer and subsequent correction burns, and then with the bad trajectory Moho capture was expensive. Barely had enough for the return transfer.
Basically I was relying on Protractor for the transfer angles and just burning prograde when it told me to. Otherwise I didn’t pay any attention to the launch windows or optimal transfer trajectories. Playing around with this transfer calculator, it seems that a truly optimal transfer will only take 4200 m/s. But if you just take the first transfer opportunity, and you do the simpler mid-course plane change trajectory, 5800 m/s might be the best you can hope for. Sloppy manual flying, of course, can add a whole lot more to that.
Nice! That’s a huge difference. I wonder how well a straight down landing would do? It solves the problems with a shallow descent angle, and since the Mun has no atmosphere and a minimal rotation rate, I think it’s going to be close to the most efficient approach.
I also suspect that my Munar return is also distinctly suboptimal. I’ve been landing on the Mun’s retrograde side–really I should just burn straight up until I leave the SoI, instead of going into orbit first. Mechjeb has made me lazy!
Looks like I’ll have to install the protractor mod. My past manual attempts had me just eyeballing the phase angle.
How are you getting 4200 m/s for a Kerbin->Moho transfer? Best I can get on the calculator is around 4800 m/s. That’s with a 610 km initial obit, though that doesn’t seem to matter too much.
I have flags (and no stranded/dead Kerbals) on the Mun, Minmus, Eve, Duna, Laythe, Vaal, Dres, and Eeloo. Did a manned flyby of Moho and an unmanned one of Kerbol. Another goal of mine is to plant flags on everything but Jool proper and Kerbol.
The Joolian outer moons are going to be pretty tough, though I think I finally have a handle on getting to/from them. There’s a bit of luck involved in not hitting Laythe, though.
There’s a 4123 m/s ballistic transfer on day 341 of year 1. The next transfer that’s as good is day 368 (?) of year 3. Most of the local optimum for the more common transfer windows are much worse, as you see.
Your suggestion worked great! I’ve got my landings down to 643 m/s delta-V–not as good as yours, but a lot better than the 770 I had before. I still used Mechjeb for the final little bit, but in any case your technique works very well.
I was having trouble at first with the second maneuver node, when eventually I realized that you can just keep pulling retrograde until things stop moving. The loop will close, but go no further, and at that point you have the correct burn.