Ore vs Fuel. Which is more Efficient?

[stephensmat]

I'm playing Science Mode, and about to build my second interplanetary craft. Thanks to my first one's flight to Duna, I have several new parts unlocked. Including the Mining/Converting parts.

What's more efficient to refuel on-mission?

1) Build a Lander that can mine and convert ore into more fuel, and launch it up.

2) Build a large converter into my 'Mothership' and fly the ore up via the Lander, convert it in space.

The first option makes my lander heavier, but makes the return flight cheaper, assuming I can carry enough fuel tanks.

The second option depends on getting enough ore up to my Mothership, or else I have to make multiple trips.

Which is more efficient? Because my Mothership option includes the Engineers, and lets me carry more radial tanks, but if the Lander is all I need to fuel up, then I can keep feeding ore from a small storage, into a larger fuel tank.

[king of nowhere]

the first option is more convenient, because in the second case your lander will take off with partially empty fuel tanks. with a converter on board, you can use your fuel tanks fully.

both options are perfectly viable, though. stock isru is so powerful, it kills the need for optimization

[Admiral Fluffy]

You could also do an apollo style mission, sending a lander down and leaving a ship in orbit.

For the proposed options, I would go with option one. You could also send the Refiner/Miner separately from the lander.

[18Watt]

I’ll throw out another option.  The mothership is capable of landing and refueling itself.   For CBs where landing the mothership is not practical, use a dedicated lander.  Tylo, Eve,  Duna, and Laythe are places where landing the mothership isn’t always practical.  Vall is marginal.   Everywhere else in stock KSP is fairly easy to land a large mothership on.

Not sure where you’re going, but if it’s say Duna- Perhaps land the mothership on Ike to refuel.  If taking the mothership into Duna’s atmosphere is too risky, just send a smaller vessel to the surface.

As @king of nowhere mentioned, you can do so much with ISRU that worrying about the ‘best’ or ‘most optimized’ way of doing things doesn’t make much sense.  In any case, what works ‘best’ on one body may not be the ‘best’ solution for another body.

[18Watt]

Just my personal preference, but if I’m sending a lander down (leaving the mothership in orbit..), then I prefer to just haul the ore up to the mothership.

My reasoning is the converter is heavy, and hauling that much mass to the surface and back each time is a little counter-productive.   
 

So if I need to fuel at a body I can’t land the mothership on, I prefer to send a lander with a drill and ore tanks, and haul the ore to the mothership.

But for me, the easiest way by far is to just land the  mothership.  Doing multiple fuel or ore runs to the surface becomes tedious and grindy, much easier to just refuel the whole ship in one shot.

In stock KSP there is always an easy moon nearby the ones that are difficult to land the mothership on.  Eve has Gilly.  Duna has Ike.  For Tylo and Laythe there’s Bop, Pol, and Vall.

[jimmymcgoochie]

Ore is very heavy. ISRU equipment is very heavy. Heavy stuff reduces delta-V.

Build an ore mining ISRU lander and put it on Ike: it's easier to get to Duna orbit from Ike's surface than Duna's due to the low gravity and Ike's low orbit of Duna and you don't need to deal with any aerodynamic effects or risk breaking solar panels etc. to get to Ike and back either. Once that's in place, use a fuel tanker to carry fuel made on Ike to any waiting vessels in orbit of either Ike or Duna, leaving the heavy mining and converting gear on the surface to keep on mining and converting while your other vessels can be lighter and simpler as they don't need their own ISRU stuff and the associated power/cooling gubbins. The alternative is to send the entire mothership to the surface, but that would just use more fuel and could be difficult to do depending on your design.

Once you have a mining rig in place, you can use it for all future missions. Put them on Gilly, Dres, Pol, Minmus and maybe even Vall and you'll never need to worry about fuel again- so long as you can get there, you're all set.

Spoiler

It's quite difficult to transfer fuel between vessels in stock KSP as you need to connect them together somehow- docking ports or klaws, or if you use mods KIS/KAS connectors and pipes- but if you're going to do a lot of mining then the mod Simple Logistics makes it all so much easier as it allows transfers between nearby landed vessels without connecting them up: you can plug each vessel into a surface network to share resources, or land a new one and request resources from the network and then leave again in mere seconds.

[Fraktal]

On 5/6/2022 at 1:35 PM, king of nowhere said:

the first option is more convenient, because in the second case your lander will take off with partially empty fuel tanks. with a converter on board, you can use your fuel tanks fully.

Why not do both, then? As in, put a converter on both the mothership and the lander. Lander lands, fuels up, loads up ore, takes off, docks with mothership, offloads ore. Mothership processes ore while the lander is now free to go about its own business in the meantime.

[king of nowhere]

35 minutes ago, Fraktal said:

Why not do both, then? As in, put a converter on both the mothership and the lander. Lander lands, fuels up, loads up ore, takes off, docks with mothership, offloads ore. Mothership processes ore while the lander is now free to go about its own business in the meantime.

because that requires 2 converters instead of 1. it's more mass.

actually, the most effective way would be to have a land base to take care of mining, and landing a ship to only carry fuel. that, however, requires more micromanaging.

ultimately, as i said, you can do everything with stock isru. it doesn't matter. you get infinite fuel, and you get it fast - because, when a transfer window comes one per year and requires half a year travel time, whether you fill your tanks in 5 days or 10 doesn't matter at all.

I soon stopped using stock isru because there was no more challenge to the game

[Fraktal]

18 hours ago, king of nowhere said:

because that requires 2 converters instead of 1. it's more mass.

Compared to the rest of your ship, it's a drop in the ocean.

I'm more concerned with running a converter on the mining rig slowing down the rate at which the ore tanks fill up.

[Hotel26]

19 hours ago, king of nowhere said:

actually, the most effective way would be to have a land base to take care of mining, and landing a ship to only carry fuel. that, however, requires more micromanaging.

My Gremlin lands with a mining rig but then separates into an ascending shuttle, carrying just the fuel.  Coming back for the next load, it does require a slightly tricky "top-dock".

If you can pinpoint the landing and then maintain a slow descent to dock, it can be very effective to Ctl From the landing ship's stern dock with SAS Radial In selected, giving intuitive RCS controls for nailing the landing.  (An idea from @Atkara)

                                                                                          

You do  say "mother ship", however, which does imply you are "travelling with your own infrastructure".  So, as has been pointed out, bringing back just ore, assuming your mining lander can make the round trip on one set of full tanks seems more efficient.  But if you have to make more than one trip, your yield may be low.

I personally, think it is always worth the investment to set up local infrastructure everywhere I go.  You know: "act like you mean to own the place".

Edited May 8, 2022 by Hotel26

[Duke MelTdoWn]

15 hours ago, Tony Swallow said:

One of the things I need to consider/investigate is the advantages/disadvantages of converting ore to fuel on a moon as compared to moving Ore to Orbit and doing it in orbit.  

I am reviving this thread , here are some of my observations.

1) You need a converter somewhere for sustained operations. I like to put it on the Miner because of the convenience already mentioned by others. Land with almost empty tanks, refuel Miner on the surface.

The drawback that you have to lug around the drills and converter can be mitigated by splitting the Miner into two stages, one with all the equipment needed on the surface (drills, converter, landing legs) and one with only the tanks to hold the ore or fuel meant for other ships.

Another benefit of split into two stages is that you can tune the TWR of the Miner stage for takeoff (landing should be no problem because you land almost empty) and that of the Fuel/Ore Ferry for orbital operations.

I implemented this two stage concept in the 'Behemoth' Mining System (https://kerbalx.com/Meltdown/Behemoth-Mining-System).
 

On 5/6/2022 at 10:42 PM, 18Watt said:

My reasoning is the converter is heavy, and hauling that much mass to the surface and back each time is a little counter-productive.   

On 5/6/2022 at 1:35 PM, king of nowhere said:

the first option is more convenient, because in the second case your lander will take off with partially empty fuel tanks. with a converter on board, you can use your fuel tanks fully.

First note that if we use a Convert-O-Tron 125 we must convert on the surface because of its poor conversion efficiency.

Lets look at the two variants assuming 100% efficiency: we want to go from orbit to the surface, gather ore or fuel, and return it to orbit. Then start this cycle again. We need to look at both landing and takeoff to get the full picture.

A) Haul around the converter. Refuel Miner on the surface.

B) Haul around extra fuel for takeoff. Refuel Miner in orbit.

Mass budget comparison

A Orbit before landing	B Orbit before landing	A Surface before takeoff	B Surface before takeoff
dry mass of vessel (incl. dry  payload tanks)	dry mass of vessel (incl. dry  payload tanks)	dry mass of vessel (incl. dry  payload tanks)	dry mass of vessel (incl. dry  payload tanks)
fuel for landing	fuel for landing	fuel for next landing	fuel for next landing (in the form of ore)
dry mass of fuel tanks for landing	dry mass of fuel tanks for landing	dry mass of fuel tanks for landing	dry mass of fuel tanks for landing
	dry mass ore tanks for next landing		dry mass ore tanks for next landing
	fuel for ascend	fuel for ascend	fuel for ascend
dry mass of fuel tanks for ascent	dry mass of fuel tanks for ascent	dry mass of fuel tanks for ascent	dry mass of fuel tanks for ascent
			fuel for next ascent (in the form of ore)
	dry mass ore tanks for next ascent		dry mass ore tanks for next ascent
converter mass		converter mass
		payload mass	payload mass

Note that fuel needed for ascent will be much larger than fuel needed for landing, because we want to lift additional payload to orbit.

-> Using no converter on the Miner only makes sense if mass of fuel for ascent is less than half of the mass of the converter. At 4.25 tons for the Convert-O-Tron 250 this break even point will be reached for any practical vessel.

2) The  question I want to investigate now is:  Is it better to ferry the fuel meant for other ships around in the form of raw ore or already converted?

Lets start by looking at the dry mass to fuel capacity ratio of some tanks. I assume 100% efficiency when converting ore to fuel (i.e. Convert-O-Tron 250 at optimal temperature). Mass stays constant during conversion. Conversion ratio: 1 ore = 2 LF [1] Tank data taken from [2].

Dry mass / LF ratio

Tank	LF capacity	Dry mass (t)	Wet mass (t)	Dry mass / unit of LF (kg)
Large Holding Tank	1500 ore = 3000 LF	2.0	17.0	0.666...
Small Holding Tank	300 ore = 600 LF	0.5	3.5	0.833...
Radial Holding Tank	75 ore = 150 LF	0.125	0.875	0.833...
MK3 LF Fuselage	5000	3.57	28.57	0.714
MK2 LF Fuselage	800	0.57	4.57	0.7125
Mk1 LF Fuselage	400	0.25	2.25	0.625
MK0 LF Fuselage	50	0.025	0.275	0.5

So if compactness or flexibility is of no concern, and if you use the right tanks, it is more efficient to ferry around the converted LF instead of the ore. If part count is of no concern, go for Mk0 LF tanks. Otherwise go for Mk1 LF tanks or Large Holding tanks.
___

Well ... in theory. Practically compactness is a concern. E.g. I would need to replace the 8 Large Holding Tanks on the Leviathan Mk.3 Ore Freighter with 60 MK1 LF tanks to achieve the same refueling capacity of 120 tons.  I am not sure I could fit all those tanks in such a way that they would not cause problems during landing (e.g. tendency to topple over because too tall).

This would save 1 ton of dry mass on a vessel that weights ~190 tons when fully loaded. Current dry mass of Mk.3 is 37.435 tons, so it might be worth it - I will investigate further
___

Quick prototyping shows I could build a new orbital LF Ferry that is compatible with the Leviathan Mk.3 Miner. This Miner stage holds the drills, converter and the strong engines (Darts) and OX needed for takeoff; also some excess LF for landing on Nervas.

Spoiler

On further thought, I decided to split the whole process into three stages instead of two:

• Orbital Fuel Fairy: will bring the fuel to the target ship: 4x Nerva, lots of Mk1 LF tanks, airbrakes for efficient return to Kerbin orbit, maybe small crew compartment to ferry tourists to the Mun and back.
• Mun descend/ascend stage: Landing legs + excess LF for landing; high TWR + OX tanks for takeoff and getting the payload back into Mun orbit; will be left in Mun orbit waiting for next Fairy
• Refueling Rover: This should be a Samaritan like Rover so that it can easily connect to the landed Fairy. Drills, converter, thermal, solar, fuel cells, science experiments etc. Maybe buffer some ore here while we wait for the next Fairy? All this heavy equipment can be left permanently on the surface.

Edited September 22, 2023 by Duke MelTdoWn
Breakeven point converter in orbit or on surface