When to and when not to use two-stage landers

[Kelderek]

Nice thread necro!

I generally go with single stage landers because of how I choose to plan my interplanetary missions.  I use a transfer stage to bring a lander from Kerbin to another planet/moon, then use the lander to get down to the surface and then back to rendezvous with the transfer stage in orbit.  I refuel the lander and then send the lander back to kerbin, leaving behind the transfer stage in orbit.  The DV requirements work out nicely for many destinations and I find that my landers have better TWR than transfer stages so it makes burning to return to Kerbin a little quicker.

[AeroGav]

Fundamentally, there aren't many savings to be made separating the descent and ascent stages of your lander, Apollo style.

Ascent module needs -

1) Engine 

2) Pod

3) Fuel for ascent to orbit.

Descent module needs the above plus

3) Science instruments (negligible mass)

4) Landing legs (negligible mass)

5) fuel for landing 

Going full Apollo style means doubling up on the engines which are much heavier in game than IRL. What you save in not lifting the empty tank mass of your descent fuel back to orbit, you more than lose by having to land twice as many engines on the lunar surface.

You can get most of the benefits by laying out the lander stack like so -

Science Stuff
     I
Descent Fuel
     I
 Decoupler
     I
Inline cockpit
     I
 Ascent Fuel
     I 
  Engine

That way you're not dragging a Science Jr + empty tank back to lunar orbit.

The real savings are to be had by separating the fuel tank with your Kerbin return fuel before descending to the lunar surface, then docking with it after coming back up  and sucking it dry.   In the case of the Mun however, I still can't see it being worth your time.   

[magnemoe]

One design I use a lot on early Mun landers is core and four radial drop tanks, landing legs on drop tanks, the first set hold material lab, goo container and extra battery and solar panels, it will run dry during decent but you keep them, second set is used during accent, core take you to orbit and back.
Expensive science equipment in service bay below pod. 
I tend to use more drop tanks below the radials to take me from suborbital to Mun orbit. 

Later on I mostly use SSTO landers as I can reuse them, 

I would say Mun or stronger gravity, multi stage is more efficient, from Ike and down or for probe / command seat its pointless. 

Edited October 13, 2016 by magnemoe

[septemberWaves]

For low-gravity bodies it's usually most effective to use a single stage lander. Even if it's not going to be reused, a single stage lander means that you don't have to use multiple engines, which saves a lot on mass. Since you're excluding Eve and Tylo, the only places in the stock system where it's generally going to be better for a single-use lander to be two-stage are Duna and Laythe. The designs will be radically different though. For Duna, I've found that, despite its atmosphere, it's most effective to land propulsively using a dedicated descent module, and then use a separate ascent module to return to orbit. As for Laythe, it's generally best to have two ascent stages and land using parachutes, possibly with some assistance from the first stage (which probably should have some excess fuel).

[Aegolius13]

For single use landers, I sometimes go the 1.5 stage route and put extra fuel in a "hat" above the command pod.  That way you can avoid doubling up on engines but still shave some mass.  And if you have a docking port on top of the lander, you can use that as a decoupler for the hat so there's no extra cost or mass.

[Tyko]

I've tried drop tanks for single-use landers on Mun. I had a central core surrounded by 4 drop tanks to which are attached lander legs and anything else I don't want to bring home. While it works, the empty tanks weren't really that heavy and I'm not sure it was worth the extra complication in design - 5 small tanks instead of one big one, 4 decouplers, etc.