Lowest Delta-V SSTM

[herbal space program]

IME experience gravity losses trump drag. Getting sideways as soon as possible is the best play dV-wise, just have to be careful to avoid burning up.

Well, I can’t say I’ve made a perfect ascent yet to LKO with this ship, but so far the best I’ve done to get to Minmus going sideways is 3108dV to LKO, around 60 m/s shy of what PLAD reported with a very similar stack, plus around 850 to get a Mun-Minmus transfer, for a total of 3955. The best I’ve done so far boosting straight up at the Mun is 3794 to the same encounter. Although there are clearly fewer gravity losses to LKO going sideways, all the sideways momentum required to get to Kerbin orbit is kind of superfluous if you can harvest enough lateral motion from your Mun encounter to kiss the SOI of Minmus at a fairly low rerlative velocity. I dunno, maybe if you have some time you could mess around with it yourself. I think it’s an interesting boundary condition to investigate.

FWIW I also spent some time investigating various alternative configurations, using Mainsail, aerospike, etc. but nothing else matched the Mammoth-based stack that PLAD originally posted. I don’t think that performance is really beatable, so I think I’m done with this challenge. Congrats PLAD. AFAICT you nailed it to the wall!

Edited November 8, 2015 by herbal space program

[PLAD]

I think herbal space program hits the nail on the head above. If Kerbin had no atmosphere it would take an explosive burst of 3336m/s to get to 11000km altitude, where Mun can grab your ship and throw it to Minmus. This is the absolute minimum for getting to Minmus. You could either go straight up, or blast straight sideways into orbit and then do a second departure burn to Mun, in theory if you do it perfectly the two ways would cost the same. (In practice the orbit method would cost more because you have to go up to clear mountains in your orbit, and the trans-Mun burn would have to be done at higher than 0 meters altitude and would lose Oberth effect). But once you add the atmosphere it gets complicated. The orbit-first method generally reduces gravity losses as your sideways speed drops the effective downward acceleration (thanks to the planet surface 'dropping away'), but raises drag losses as you spend longer in the air. The straight-up method drops drag losses as you spend much less time in the air, but you are fighting full deceleration the whole way so gravity losses can be much worse. Just from empirical experiments I find the straight-up method becomes more efficient at very high TWR values, though the break-even point also depends heavily on the drag of your vessel. At high enough TWR straight-up inevitably wins out though, because sideways spends so much time in the air. Most ships have TWR's far too low to reach the break-even point because of total mass considerations so straight up is rarely the best way to go (not to mention the aiming accuracy problems), but sometimes...

Even for straight up there is a point where raising the TWR no longer helps. In my best launch I got from launch to the 11000km apoapsis for 3736 m/s expended. The fuel used was the equivalent of 3790m/s in a vacuum, so we see I lost 54m/s from burning fuel in the atmosphere with a lower iSP. Mechjeb says I lost 65m/s to drag, leaving 335m/s in gravity losses. Let's imagine the Mammoth had a TWR of a million so the burn would be an explosion on the launch pad. (Ooh, I just had images of many past launches flash before my eyes). Now the gravity losses would be 0, but the dV of the ship would drop from the current 4092 to 3833m/s because all of the fuel would be burned at ground level where the iSP is only 295s instead of 315. I'd have to add about 204 m/s of fuel to the ship to get the 4037m/s expended dV for the whole mission. And the drag loss would be ferocious. My tests explode when I try 3500m/s at ground level, but I bet they'd be more than the gain from no gravity losses.

As an aside, note that if you make a custom motor that is the same as the Mammoth except the air iSP is 315 (same as vacuum) as RIC suggests then you'd cut 54m/s from my score. Or even a bit more as you'd have higher acceleration at launch. So I contend custom motors can thoroughly beat my score. Who will try it first?

I hate to take credit for my ship design as it is so similar to what cubinator used in his example. It's hard to beat a Mammoth on the bottom of a conical shape for high TWR/low drag.

[cubinator]

Yes, it seems this challenge has reached it's ceiling. Sorry for accidentally making a near-perfect design on the first try. I'm sure it's possible to get a little lower delta-V, but not much.

Nobody made a turtle yet though...I might make that into a separate challenge.