I fail at Reusable

[MrOsterman]

So a while back in my Career mode I launched a 600k ship called the Tereshkova and sent it off to Duna with a lander in it's tow spot.  I managed to use my refueling operations there (which I had launched a while ago) to send it back and decided that while fully functional, I didn't want to keep sending stuff out that required me to spend a few hours of my time docking, landing and launching a mining ship to refuel it for a return trip.  I had planned for the Tereshkova to be a permanent space ship that would travel between systems and whatnot.  

Then I commissioned and launched the KSP Athena which is capable of doing a landing directly on the surface of a moon or planet.  It's outfitted with "belly engines" (Thuds) to allow it to do a direct descent along with rows of Chutes for landing on worlds with atmosphere.  It has its own pair of mining drills, ISRU convertors, Fuel Cells, and a pair of MSL's (though only one can function at a time it looks good for the symmetry).  The problem is that the Belly Engines still aren't balanced right and I've not yet tried to tack into the power of a nuclear engine (I'm nervous about the required cooling for them).  So I'm finding myself musing on the idea of building a new Athena II to send up that will feature Nerv engines, (in place of my current LF/Ox engines) and make some other general design changes (such as correctly setting up my fuel lines, adding a runabout shuttle, etc.).

Which also means, yet again, I'm taking a "I don't need to keep designing I can just use what's out there" and instead I'm planning to discard what's out there for something better....

 

[WhiteKnuckle]

1 hour ago, MrOsterman said:

...and I've not yet tried to tack into the power of a nuclear engine (I'm nervous about the required cooling for them).  

Your ISRU gear is going to need more cooling than a nuke or two will, so it shouldn't matter much.

[RizzoTheRat]

You could try tweaking the maximum thrust on the each engine of your existing ship to get it to balance properly.  I had a similar problem with a drilling rig I landed on Minimus, in that I hadn't got the lift engines correctly spaces around the CoM so it always pitched up.  A bit of practice thrusting in orbit led me to reducing the max thrust of the front engines, which I believe I believe then scales linearly with the throttle, and ended up with a small enough asymmetry that the reaction wheels could handle it and keep the ship straight.

[FullMetalMachinist]

54 minutes ago, RizzoTheRat said:

You could try tweaking the maximum thrust on the each engine of your existing ship to get it to balance properly.  I had a similar problem with a drilling rig I landed on Minimus, in that I hadn't got the lift engines correctly spaces around the CoM so it always pitched up.  A bit of practice thrusting in orbit led me to reducing the max thrust of the front engines, which I believe I believe then scales linearly with the throttle, and ended up with a small enough asymmetry that the reaction wheels could handle it and keep the ship straight.

Or instead of doing it manually (the hard way) you could install the Throttle Controlled Avionics mod.

[MrOsterman]

1 hour ago, RizzoTheRat said:

You could try tweaking the maximum thrust on the each engine of your existing ship to get it to balance properly.  I had a similar problem with a drilling rig I landed on Minimus, in that I hadn't got the lift engines correctly spaces around the CoM so it always pitched up.  A bit of practice thrusting in orbit led me to reducing the max thrust of the front engines, which I believe I believe then scales linearly with the throttle, and ended up with a small enough asymmetry that the reaction wheels could handle it and keep the ship straight.

So far for me landing, if I keep the throttle under 50% I can (generally) let my RCS thrusters and the two reaction wheels generate sufficent torque to keep me mostly level with the ground on the way down.  I did run into a small problem coming down on Duna in that I ran out of fuel (not good I know) and that created the situation where I hit harder than I would have liked.  I'm also not that well balanced front to back when empty compared to when fully loaded.  I twiddled a little with the max thrust on my front engines and I was mostly level when empty but when I lifted off I had a real serious pitch up, probably because all of my fuel was towards the back.

So on the Nervs, if I just attach one or two medium non-deployable radiators per engine I should be okay?  I feel like the in game tool tips aren't very tippy.

Oh.. and I'm fairly proud that I've managed to run only with Chatterer as a mod.

[Aethon]

 

This is from an older version (I don't have a current screenie) but it may give you some ideas.  It's called a system shuttle, here flinging a payload to Moho.  It's overbuilt to deal with contingencies, and returns to Kerbin for a 45 km aerobrake.  The current version has some small retractable radiators landing legs and Procedural parts LF only tanks.  Sorry, kinda in a hurry.

Kerbal engineer has a readout in the editor that shows thrust torque.  It will tell you when everything is centered and torque is zero.  I zoom waaaay in on the offset tool and shift different parts back and forth until KER says zero thrust torque.  Not necessary for the above ship, but may help you with your current problem.

Hope this helps.

Ok I've got a little more time now, so I'll flesh out my above post.

 

Here's my current System shuttle chucking a 5 kerb contract sat at Minmus.  The escape pods will detach and dock with the station upon arrival (I needed to fit them in the fairing).  It actually doesn't need any radiators. and will burn the main tanks dry without incinerating the solar panels, though it's very close to that temp when the fuel runs out.  Two small, retractable radiators might be a good idea (the non deployable ones are always sucking power and there is no stock way to shut them down).  Note the Ion engines on the bottom.

 

Here it is docked to Kerbin SciSkylab ready for the next mission.

 

 

You might have better luck splitting your ISRU into it's own dedicated ship.  Here's my large Kerbin system ISRU, and the fuel tanker on top is based on the reusable nuke tug concept of the system shuttle.  The ISRU roves on the surface with the small LFO engines and brings fuel back to the tanker, which takes it to the station in orbit.

 

 Here's a smaller interplanetary ISRU that will make Moho orbit (needs a refuel for landing) based on the same concept.  It has two small retractable radiators but relies on three of the four ISRUs' radiators for the loooong Moho burn.    The other side has a small ISRU and a single radial ore tank.  It's not perfect, but perfect is the enemy of good enough.  I plan to have one of these at each planet.

It's important to keep the Main LF tank centered so that as fuel drains, the center of mass doesn't shift around.  The LFO tank is locked since it only feeds the surface operation engines which aren't used in space. I slide various parts around with the offset tool until KER tells me there is zero thrust torque.  

I respect your attempt to be mod free but IMO, KER is indispensable, not to mention the fact that there are no stock LF tanks that work well with nuclear engines

None of these ships would be possible without the torque readout you can see in the Kerbal Engineer Redux window in the upper right.

Good luck.

Edited January 14, 2016 by Aethon

[RizzoTheRat]

What mod are you using for those LF tanks?  I agree the standard tanks aren't great for use with nukes, but can still work well

I've not really had much trouble with Nukes overheating except on very long burns, in which case you can just throttle them back and run them for a bit longer.

This used the landers engine from launch, and once in kerbin orbit went to Minimus and then the Mun without needing to throttle back.

 

[Aethon]

Procedural parts.

[MrOsterman]

Hmm... I played around last night with trying to build for the Nuke engine but I bumped into the problem of the engines only using Liquid fuel and the size tanks I think I'm going need not matching up for interplanetary travel.  I tend to over pack on fuel which I know is bad engineering but I also don't trust myself to not need to make extra corrections when I get to where I'm going.  Ideally I should be landing with my tanks emptying, but I always like a little wiggle room.

Maybe I need to get out of my design ascetic.  The current ship I'm flying has the MK3 main body fuselage, then 2.25m along the sides (for the ISRU, Ore tanks and Science labs, and then another "layer" of 2.25 tanks along side that.  I also realize that two ISRU's and two sci labs is redundant but my first space ship wasn't designed for landing (it had a lander attached) and I tried doing the "this is the processing side, this is the science side" and the lack of symmetry made it very unwieldy in space.

[problemecium]

Try starting simpler. Your super-duper-all-in-one-spaceplane-lander-base-refinery doesn't need to be reusable. Since it's expensive, you'll likely be motivated to use it again where convenient, but as far as reusable infrastructure, try something smaller first, like this:

In this case, it's a mini-lander with a set of small experiments, which is good for shuttling back and forth between the Mun and a station in Munar orbit. Reusability is a major boon there, as it can go on repeated Science collection runs to different biomes. Since the design is fairly simple, you're unlikely to make a new one so spectacularly better that it outmodes the existing one, and if it does, bringing back the old hardware for money back is easy.
For more ideas, here's some reusable things I commonly employ in my space infrastructure:
- Space tug (little MonoPropellant-fueled probe with one or two docking ports for moving stuff like station modules around)
- Mini-lander
- Crew shuttle spaceplane
- Orbital Science probe (for those "Science data from space near X" or "X scans from space above Y location"; parks in a low polar orbit and hangs around to scan whenever it passes over a mission site)
- Orbital crew ferry (for moving Kerbals between stations)

[MrOsterman]

Yeah the problem I'm dealing with might be partially attributed to Mission Creep.  Here's what I have as "must haves" on my ship:

I need an ISRU to get fuel for my ship so I can get around independently.  

I need at least 2 drills to fill the ISRU.

I need room to fit Tourists/ Training Kerbals (to get them from 1-3 stars).

I need the whole ship to land; I can't do the land, mine, lift any more.  I did that parked over Ike with a space station/ refueling Depot and I just can't do it any more.  It's too boring to me as a player.  It was the closest I got to simply hacking my safe files to show that I had done the work of landing and ascending with the ore to avoid doing that same flight 3-6 times to top all my tanks off.

I wanted to try to do this as an MK1 or an MK2 1.25 meter build but packing the required fuel with the ISRU is just proving to be more and more of a bear.  The Athena is still out there as my first full ship lander with all the "we're out here for the long haul" ship but I know I can do better than it.  It's got a LOT of parts and really struggled to land on Duna safely (broke half my landing gear doing it, though my engineer was able to effect repairs).

So now I'm building on the 2.25m scale or the MK3 scale and that's making one BIG ship requiring BIG engines.  My last effort at a nuke engine powered ship had 6 Nukes on it and it would have taken 9 minutes of solid burn to get me into an Munar intercept.  I don't want to think how long the burn would be to get me to Duna....

 

[WhiteKnuckle]

As someone who's been working on a few big ISRU projects recently, I can tell you, you're asking a lot out of a single ship. I'll go through your requirements one by one:

"I need at least 2 drills to fill the ISRU"

Small drills right? Don't tell me you're trying to lug two of the big ones around. More drills are a luxury, use as few as you can stand.

"I need room to fit Tourists/ Training Kerbals (to get them from 1-3 stars). "

How many? Use the lightest means of carrying them you can.

"I need the whole ship to land; I can't do the land, mine, lift any more.  I did that parked over Ike with a space station/ refueling Depot and I just can't do it any more.  It's too boring to me as a player.  It was the closest I got to simply hacking my safe files to show that I had done the work of landing and ascending with the ore to avoid doing that same flight 3-6 times to top all my tanks off. "

How much ore were you bringing up with each load? From my testing on the Mun I was able to bring up something like 2700 units of ore, and wasn't even close to maxing out my design. Converting all that to LF and feeding it to a nuke gets you quite a bit of DV.

Really it's a question of how you want to spend your time. Refining in orbit means you can use the large ISRU, have tons of solar panels, batteries, fuel cells, storage, the whole nine yards. You can set it and forget it, all the time is spent ferrying ore from the surface to orbit.

Refining on the ground (along with your other requirements) means you're going to have to cut every gram of mass you can, that means using the small ISRU (which only turns 10% of the ore into fuel, meaning you have to drill 10x as much of the stuff) It also (I believe) isn't efficient enough to support Fuel Cells, so you're stuck with solar/RTG so you'll be running out of juice a lot and be stuck timewarping till dawn. Really it's just a lot of babysitting, personally I'd rather be flying things back and forth than sitting on the ground, but I can see the other side.

 

"I wanted to try to do this as an MK1 or an MK2 1.25 meter build but packing the required fuel with the ISRU is just proving to be more and more of a bear.  The Athena is still out there as my first full ship lander with all the "we're out here for the long haul" ship but I know I can do better than it.  It's got a LOT of parts and really struggled to land on Duna safely (broke half my landing gear doing it, though my engineer was able to effect repairs). "

Two options, one: Leave your transfer stage in orbit, even if it's just the nuke, a few fuel tanks and a probe core. If you did that you might just barely be able to pack everything into a 1.25m package.

Option two: Wings

This is from .90, but as you can see, a single nuke is more than enough to make Duna orbit, and this baby was able to be refilled on the surface and go all the way back to Kerbin.

I'm currently designing something similar with a Mk2 cargo bay and all the ISRU gear in it, but I'm waiting for a transfer window to test it out. I can post a pic/craftfile if you're interested.

 

"So now I'm building on the 2.25m scale or the MK3 scale and that's making one BIG ship requiring BIG engines.  My last effort at a nuke engine powered ship had 6 Nukes on it and it would have taken 9 minutes of solid burn to get me into an Munar intercept.  I don't want to think how long the burn would be to get me to Duna.... "

Yup, nukes are slow. Water is also wet. It's the price you pay for efficiency. My Big ISRU ship took an 8 minute burn to get from LKO to the Mun, so yours sounds about right.

[MrOsterman]

Hmm... Imgur is blocked here so I'll look at the design tonight and yes I've been lugging the big drills around with me so maybe that's part of the problem.  In theory if I can convert the ore fast enough to fuel the drill with a fuel cell, and fill the tanks with "gas" I could use the small drills and the smaller ISRU to do the job.  That would save a LOT on weight as well as allow for smaller designs.  I could, conceivably, do it all with MK2 parts.

My practice so far has been with an ore lifter that was netting about 700 or so units of LF/O each round trip (or 1400 units total).  If I was only working with a Nuke engine (or two) I could improve that which might be a new focus.  I tend to run pretty convservative on my designs that way.  I often plan for more more fuel than I need for any given project as I just don't want to deal with rescue missions.  :/

For "seating", a typical mission carries 4 tourists and then I add in 1-2 kerbals for training purposes.  I don't usually bring scientists any more (I'm done with the tech tree) but I do need that engineer to "run" the drill and ISRU.  Plus, given that a rough landing can be murder on the landing legs, it helps to have someone float out and fix them.  Well that and repack the chutes as I'm happy to use those for a Duna landing where I can, though I guess I could run the analysis to see if trying to do a powered landing would save on mass over slowing to safe speeds for the chutes and then using them to eat up the rest of my landing velocity.

My current project is intended to be a mission to the Mun, then to Duna then ONWARDS and has to seat 5 tourists plus crew so I'm looking at accommodations for a minimum of 7.  Add in a kerbalnaut in training and that's seats for 8.

Maybe I need to go back to the Athena design and see if I can tweek it.  It's a beast of a ship, no doubt, but even doing a HORRIBLY inefficient burn to Duna (I was going from solar orbit for the Duna burn) it still made it to Duna, and took off again, and did so hauling an MK3 flight deck and the MK3 passenger compartment.  Oh and the fuel convertor, two large drills, two science labs (for symmetry and balance).  It's big, but it worked....

 

Edited January 20, 2016 by MrOsterman

[RizzoTheRat]

Symmetry/balance is a pain in the poophole at times due to the different weights/lengths of various items.  It would be nice if there were useful modules of similar weight and dimensions to the ISRU and drill to balance craft out properly.  I tend to have the ISRU on the centreline and the drill as close as possible, but I'd love to be able to have a battery pack that was the same weight as the drill to act as a counterbalance.  Batteries seem surprisingly light to me.

[WhiteKnuckle]

1 hour ago, MrOsterman said:

Hmm... Imgur is blocked here so I'll look at the design tonight and yes I've been lugging the big drills around with me so maybe that's part of the problem.  In theory if I can convert the ore fast enough to fuel the drill with a fuel cell, and fill the tanks with "gas" I could use the small drills and the smaller ISRU to do the job.  That would save a LOT on weight as well as allow for smaller designs.  I could, conceivably, do it all with MK2 parts.

My practice so far has been with an ore lifter that was netting about 700 or so units of LF/O each round trip (or 1400 units total).  If I was only working with a Nuke engine (or two) I could improve that which might be a new focus.  I tend to run pretty convservative on my designs that way.  I often plan for more more fuel than I need for any given project as I just don't want to deal with rescue missions.  :/

For "seating", a typical mission carries 4 tourists and then I add in 1-2 kerbals for training purposes.  I don't usually bring scientists any more (I'm done with the tech tree) but I do need that engineer to "run" the drill and ISRU.  Plus, given that a rough landing can be murder on the landing legs, it helps to have someone float out and fix them.  Well that and repack the chutes as I'm happy to use those for a Duna landing where I can, though I guess I could run the analysis to see if trying to do a powered landing would save on mass over slowing to safe speeds for the chutes and then using them to eat up the rest of my landing velocity.

My current project is intended to be a mission to the Mun, then to Duna then ONWARDS and has to seat 5 tourists plus crew so I'm looking at accommodations for a minimum of 7.  Add in a kerbalnaut in training and that's seats for 8.

Maybe I need to go back to the Athena design and see if I can tweek it.  It's a beast of a ship, no doubt, but even doing a HORRIBLY inefficient burn to Duna (I was going from solar orbit for the Duna burn) it still made it to Duna, and took off again, and did so hauling an MK3 flight deck and the MK3 passenger compartment.  Oh and the fuel convertor, two large drills, two science labs (for symmetry and balance).  It's big, but it worked....

 

I'm pretty sure that since you need to mine 10-times the ore to get the same amount of fuel out of the small converter, there is no way you can break even using fuel cells to power it. You always burn more than you make. Even using the large converter you need to be on a good ore concentration and have an engineer with you.

If you find lifting ore to orbit tedious, I'd start by designing a lifter that can bring a ton of ore in a single trip. When you say you were able to bring up 1400 units, is that including what it took to get your miner to and from the surface? (ie is it all fuel that can be used for the interplanetary stage, or did some of it have to go back into the miner for it's next run?)

Aerospikes for the lander are key. They're efficient enough to not blow your dv budget, and their excellent TWR gives you the power to get a lot of ore off the surface.

 

8 Kerbals is a lot. I have no advice for you here... my design started with seating for two, I realized quickly that was decadent and foolhardy. Now it has a single Mk1 lander can, and he's lucky he's not riding a lawn chair.

 

[MrOsterman]

I can look tonight.  The first station parked around Ike is still there as are the two abandoned mining vessels I used to use for it.  I left them in orbit figuring that maybe they'd be useful later as some kind of fuel ferry or as a life boat or something.  But for a while I was working two landers/ miners with the station holding the ISRU unit and I want to say it was on the order of 5 round trips to generate and store 4,000 units of fuel.  Now that could also be me not be top of efficiency for landing and lifting ore.  I could possibly have lifted more ore on a given trip and brought it up.  My miners are based loosely on my munar lander which was designed around stability and ease of flight more than capacity.

And 8 is a lot but it's unavoidable given mission parameters.  I could engineer it down or I could engineer add on passenger pods if I go that route which maybe I should, but passenger pods are still mass to move.  Though... going down to one ore converter and one drill will save on some mass... then if I make room along the center line to add "Mission pods" I could still have a two seater as my main Space ship... maybe.. though now I'm worried about atmospheric entry with a part attached strictly with docking ports, even if it's a triple connection.

Maybe I'm not failing as bad as I think.  I just wish there was a better way to "update" a ship already out there rather than building new and launching.  The Athena is a good ship... she just needs updates.

 

[fourfa]

5 hours ago, WhiteKnuckle said:

I'm pretty sure that since you need to mine 10-times the ore to get the same amount of fuel out of the small converter, there is no way you can break even using fuel cells to power it. You always burn more than you make. Even using the large converter you need to be on a good ore concentration and have an engineer with you.

I made a very detailed spreadsheet a while back to investigate this.  It's a rather more complex picture.  I'll try to be as clear as possible.  All the following assumes steady-state mining and conversion on the surface, not mine and lift.  I also only looked at 100% fuel-cell power - I was looking at destinations like Eve and moons of Jool where solar is essentially useless, and KSP's fuel cells are so efficient it hardly makes a substantial difference anyway.

With the large ISRU, large drill, 100% fuel cell power, and a zero star engineer, mining is profitable on about 0.3% ore concentration or better.  With no engineer at all, you need about 1.5% to turn a profit.  

With the small ISRU (which does require 10X ore for the same fuel output relative to the large ISRU), large drill, 100% fuel cell power, and a zero star engineer, you need around 3% concentration to turn a profit.  With no engineer at all, you can just barely turn a profit on 14% (this is so rare to find on the surface that it's basically impossible).

If you can use a 3-star engineer (the max you can get in the Kerbin SOI, and easily obtained by the time you're looking at ISRU ships), it's basically impossible not to turn a profit with the large ISRU.  The small ISRU will profit down to about 1% - for this reason alone, the small drills (which do not work on less than 2.5%) are a major liability.

Since large and small drills extract the same amount of ore per electric charge burned, the math comes out the same with either.  You simply need five small drills to equal one large drill.  And part count being what it is in this game, plus the 2.5% problem mentioned above, it's really, really worth finding a way to use at least a single large drill over any number of small drills.

Edited January 20, 2016 by fourfa

[WhiteKnuckle]

Thanks for the detailed info @fourfa, so basically a manned (engineer) miner, you can get away with using the small converter, but an unmanned op is going to require the large one if you plan on using fuel cells to power it.

 

Good to know 

[Laie]

On 1/14/2016 at 3:51 PM, MrOsterman said:

Which also means, yet again, I'm taking a "I don't need to keep designing I can just use what's out there" and instead I'm planning to discard what's out there for something better....

I see nothing wrong with that.

As for myself, the main game is Vehicle Assembly -- the whole simulator thing only serves to prove my designs (or occasionally to fail them, but nevermind...).

My last career was full of could-be reusable designs that were nontheless discarded for something even better; and the one time I devised a go-anywhere ISRU lander, it became boring once it had proved its mettle landing on Duna and Tylo.

What I'm trying to get at: there's no "right" way to play KSP. If you don't enjoy flying the same craft over and over again, then don't do it. And don't apologize either, because you don't need to.

[WhiteKnuckle]

So here's a few pics of my landers. First with nothing on it, then with a large ore tank, that would be brought to just about anywhere (problem worlds like Tylo, Duna, Laythe, you can add an extra fuel tank of any size on top with modular Sr. docking ports)

Uses Aerospikes, meets up with Nuke stage in orbit, carries no RCS, has extra docking ports to add... well anything you want, science pod, crew, extra fuel, snacks. Should have a fuel cell knowing what I do now.

Uses large converter cause aint nobody got time to mine 10x ore...

 

...Unless you're trying to get the whole mess off Kerbin SSTO style (ie anywhere but Eve)

 

 

 

This one is nothing but fuel and engines. Heavy lift package at the top would be swapped for big ore tanks on most bodies. Makes a 100x100 orbit and meets up with it's interplanetary stage. Wound up being too part heavy, hasn't left LKO yet.

Finally: My lightweight Duna Spaceplane.

 

Fairing at the front hides an ore scanner, otherwise it's pretty self explanatory. Haven't tested on Duna yet, use inspiration at your own risk.