Interplanetary travel, windows, efficiency and more.

[LN400]

I have a few questions regarding interplanetary transfers.

1) Do you travel to other planets in career mode or do you do them in sandbox or science mode?

Reason for asking is, the ships I made for say, a one-way trip to Moho, was so expensive moneywise I find it hard to imagine this would be something I'd do on a regular basis in career mode.

If you do go to other planets in career mode then

1.1) Any tips on keeping the costs down? All tips, no matter how obvious or mundane they may seem, are welcome as long as they are constructive.

I do use a few mods but right now the only parts mods I use are MRS and SpaceY (waiting out on KW to see what will happen with KSP 1.1). Pricewise the parts are fair to stock parts I feel, comparable parts have pretty much the same pricetags to them.

I am aware that by missing the transfer windows can make huge differences in ship design and costs, which brings me to the next question

2) Not every window is the same. Take Moho as an example. The inclination makes for most efficient windows being few and far between (unless you have a tip, of course on how to deal with a 7* inclination costing in the neighbourhood of 2000 dv. Add to that the eccentricity of Moho's orbit and you get windows that vary greatly in terms of delta v, ship costs and travel time. So, the question:

On average, in your experience, would the dv "cost" be for a trip to Moho, or Eeloo for that matter, and what would you consider "good" in terms of dv on one hand and travel time on the other, and what price range are we looking at?

2.1) What would your never exceed limit on dv or time be, or costs, if you play career mode?

 

I do use the Kerbal Engineer mod for info, RT for commlink fun and automated control and I'll probably get KOS installed again at some point but I'm reluctant to use MechJeb or similar mods. I'd prefer to do the setup myself.

With that in mind, on to

3) How do you set up for interplanetary transfers? By setting up a node for an encounter not looking at dv or any sort of costs, or is there some good planning ahead in the way you do it? Or do you go all out free form jazz improvising everything knowing everything can be sorted out somehow?

 

[Plusck]

1) All in career. I've hardly touched sandbox since graduating from the demo (which I played to death at various stages over the last few years).

I don't find the cost prohibitive at all: I've currently finishing off a medium difficulty career, year 10, nearly everywhere visited by Kerbals and 28 million funds in the bank.

I was lucky for my first visit to Moho - if I remember correctly I sent a probe with an LV-N and four mk1 tanks (two detachable), with a sub-probe with just monoprop and two puffs, without having a very clear idea of how much dv it had. Got a good capture and landed on Moho easily. I've since experienced the harsher realities of Moho, so my Kerbal-crewed mission there (ongoing) is going via Eve and has a mining ship included.

Since then, I've installed KER so I know what my dv is, but I was fine without it until well after my first manned trips to Eve and Duna.

1.1) Only thing I can really say is not to try to do too much at once. Build a light-ish mining/ISRU ship that you're happy with, get used to landing it on the Mun, and you can explore the main three planets (Eve, Duna, Jool) using the same miner to get fuel from moons.

Also - don't try to bring much of anything back home: light pods, heat shields, kerbals and data only. Leave everything else in orbit or on the ground to be reused by the next crew.

Put a docking port on everything and assemble ad hoc stations, so that any spare fuel and/or electricity generation can be shared.

2) I use this https://alexmoon.github.io/ksp/  for approximate dates for transfer windows and as a rough guide to how much dv I'll need. I certainly don't try to plot the transfer precisely using any of the information that the site gives. A mid-course correction costing a few m/s is all it takes to arrive exactly where and when you want, once you're on a decent-ish transfer.

Again, Moho is a bit of an exception. I'm going via Eve to reduce both the Kerbin escape burn and the Moho capture burn, hopefully. Going by Eve also lets me improve my orbital plane (again, just a cost of a few m/s during the mid-course correction) by choosing exactly where I enter Eve's SOI. It won't get it perfect but it doesn't need to be as long as I cross Moho's orbit just once at a point where I can choose my arrival time without huge expense. I'd have to come back on the exact cost but I think I budgeted for about 5000 m/s dv per ship. I am sending three ships and they cost between 190k-330k each (iirc).

2.1) Nothing is really "never exceed" except for cost and size, because I don't need it. I've played around with the bigger rockets but until now, I have never needed anything bigger than the Skipper for 99% of my builds.

 

3) See 2) above. Basically get an approximate date and dv cost, send up rocket, plot a node with approximately that dv cost and drag it around to get a perfectly prograde escape from Kerbin SOI and some sort of encounter*. Then try reducing the dv cost until no enounter is possible, and use that last encounter as the good one. Halfway out (or at the AN or DN, if it is nearish halfway) plot a course correction node to arrive where and when I want to in the target system (i.e., to get a Tylo flyby for dv-free Jool capture, Ike for a cheaper Duna capture, or Eve to set up arrival at Gilly).

*edit: this might need a second node placed later for plane-change purposes, of course, but again I will tend to correct this once I'm actually out of Kerbin's SOI.

 

Edited February 16, 2016 by Plusck

[LN400]

First, heaps thanks for a great post! For one, using Eve to get to Moho hasn't occured to me but that is one up for testing!

Second, I forgot to mention I also use ScanSat and KIS/KAS.

Currently, my planned missions for each planet (or moons) are, in order:

1) Establish communication either via satelite or orbital command station.

2) Surface scanning and mapping for picking mining/landing sites . One satelite planned with all scanners and enough dv for altitude changes, for each planet.

3) Establish mining operations and orbital fuel stations around each planet or appropriate moon.

4) Manned expeditions for science.

Only the 4th will have a return ticket.

 

The mining operation I have planned involves building modules that are connected on site using KIS/KAS. Tests I have done look promising. I am trying to avoid sending an entire mining/refining operation as one build as I haven't found them very likely to survive the initial ascent. The price tag is pretty enormous though, around 1.8 million for a complete rig not counting the necessary rocket/lander. I can easily cut the costs a great deal by using solar panels but right now I have opted for the reactors (and whoah are these drills and converters power hungry!) to have a 24/7 capability.

Good idea too about docking ports. I tend to think of each mission isolated from the others in terms of reuseability. That could need some change.

It also seems to me I have quite a lot to learn about efficient builds as my first Moho scouting probe used a 3m cluster under a huge tank as its first stage lifter. These things are not cheap

[Plusck]

5 hours ago, Plusck said:

[For Moho] ... I'd have to come back on the exact cost but I think I budgeted for about 5000 m/s dv per ship. I am sending three ships and they cost between 190k-330k each (iirc).

I should correct that:

Budget was about 5-6k dv per ship after leaving Kerbin SOI. One of the three was a touch on the low side and I'm a bit worried it won't quite capture at Moho, without using up some of the fuel that I intended to land with. That doesn't matter too much because I will be able to dock with one of the other ships and syphon off fuel. Pics at the bottom of the post...

And I vastly exaggerated the cost. They cost 177k for the miner, 110k for the crew ship and 120k for the lab.

 

4 hours ago, LN400 said:

... The price tag is pretty enormous though, around 1.8 million for a complete rig not counting the necessary rocket/lander. I can easily cut the costs a great deal by using solar panels but right now I have opted for the reactors (and whoah are these drills and converters power hungry!) to have a 24/7 capability.

Woah! OK, I don't think I've ever put a ship up with even a quarter of that budget.

My mining rig is not 24/7 capable. In any event, the game doesn't let you mine and convert at the same time when ships are on rails, so I have biggish batteries to soak up energy while I'm doing other stuff, and then it only takes an hour or so of warping with the ship in focus to get about 2000 ore and convert half to fuel+oxidizer. Once back in orbit, that's more than enough potential fuel and/or oxdizer and/or monoprop to top up tanks, get home or move to the next bit of the system I'm in.

This is the ship I used to send my mining rig to Moho. The miner itself generally has more ore capacity and smaller engines, but Moho is bigger than the Mun/Ike/Gilly/Bop/Poll where I've refuelled before. Total cost of the ship (including a prospecting lander and NBS): 177,000 funds.

Spoiler

This is it approaching Eve for the gravity assist to Moho. Worried about fuel:

Spoiler

And these are the other ships I sent:

Spoiler

And their status approaching Eve, much more comfortable:

Spoiler

Since I was a bit worried about fuel before leaving, I sent this (hugely expensive by my standards) ship to go prospecting around Eve. Six mini landers to scan various biomes, and three orbital NBS ships to avoid wasting time once I get around to looking for an exact spot. I gave it a Klaw so that it could then go chasing down and helping out stragglers to Moho.

Spoiler

And finally, what I mean about attaching docking ports to everything and building ad hoc stations. This design didn't work as expected because of the ship tree structure and fuel availability, which I didn't really understand when I built this. And I don't think I'll be using its reconfigured setup as a return vessel either. But anyway...

Spoiler

So yes, my designs tend to be clunky but they're relatively functional. I've just got a couple of hugely difficult contracts to complete, then I'll start thinking big, sleek and pretty...

[WhiteKnuckle]

16 hours ago, LN400 said:

I am trying to avoid sending an entire mining/refining operation as one build as I haven't found them very likely to survive the initial ascent. The price tag is pretty enormous though, around 1.8 million for a complete rig not counting the necessary rocket/lander. I can easily cut the costs a great deal by using solar panels but right now I have opted for the reactors (and whoah are these drills and converters power hungry!) to have a 24/7 capability.

Good idea too about docking ports. I tend to think of each mission isolated from the others in terms of reuseability. That could need some

Disclaimer: I don't play career so I have no idea how much things cost.

When it comes to ISRU, the main question is where are you converting ore to fuel?

If you're converting on the surface, you can be as inefficient as you want, all it'll cost you is a few more seconds at timewarp. This means you can use fuel cells to supplement an RTG when your power demands are high. You can also use the small converter, but it takes 10x as much ore to make a unit of fuel, and that means significantly more time, even at timewarp. Frankly, the more I use them, the more I just suck it up and use the big converter for everything.

So for the surface: A few big batteries, an RTG to trickle-charge them when you're not mining, and as many fuel cells as you need to supply your operation going full tilt. 

However, if you're refining ore in orbit you need to be efficient, otherwise you'll be making more trips to the surface to fill up. This means using the large converter, and not using fuel cells to generate your power. That leaves solar, RTG, and engine power. Batteries are lightweight and (I assume) cheap, so packing a ton of batteries is the easiest/cheapest way to convert fuel. Just let one or two RTGs trickle charge your battery banks while your miner is down on the surface doing it's thing, then use that stored power to convert some or all of your ore. On your last fill up before leaving for another planet, you can use the electricity your engines produce during the transfer burn to power the converter too.

Also, I just checked some of my screenshots, I have no idea what you're building that costs 1.8 million.

This thing can SSTO from Kerbin, and would just need an interplanetary stage to get around:

 

This is a more basic miner that would fine for any smaller body:

Now both of those were earlier versions of a "go anywhere" lander, I don't have a VAB shot of my latest/best setup, but the whole ship, interplanetary stage and all can't cost more than a million total.

[juanml82]

51 minutes ago, WhiteKnuckle said:

Disclaimer: I don't play career so I have no idea how much things cost.

When it comes to ISRU, the main question is where are you converting ore to fuel?

If you're converting on the surface, you can be as inefficient as you want, all it'll cost you is a few more seconds at timewarp. This means you can use fuel cells to supplement an RTG when your power demands are high. You can also use the small converter, but it takes 10x as much ore to make a unit of fuel, and that means significantly more time, even at timewarp. Frankly, the more I use them, the more I just suck it up and use the big converter for everything.

So for the surface: A few big batteries, an RTG to trickle-charge them when you're not mining, and as many fuel cells as you need to supply your operation going full tilt. 

However, if you're refining ore in orbit you need to be efficient, otherwise you'll be making more trips to the surface to fill up. This means using the large converter, and not using fuel cells to generate your power. That leaves solar, RTG, and engine power. Batteries are lightweight and (I assume) cheap, so packing a ton of batteries is the easiest/cheapest way to convert fuel. Just let one or two RTGs trickle charge your battery banks while your miner is down on the surface doing it's thing, then use that stored power to convert some or all of your ore. On your last fill up before leaving for another planet, you can use the electricity your engines produce during the transfer burn to power the converter too.

Also, I just checked some of my screenshots, I have no idea what you're building that costs 1.8 million.

This thing can SSTO from Kerbin, and would just need an interplanetary stage to get around:

 

This is a more basic miner that would fine for any smaller body:

Now both of those were earlier versions of a "go anywhere" lander, I don't have a VAB shot of my latest/best setup, but the whole ship, interplanetary stage and all can't cost more than a million total.

He's using nuclear reactors, probably from Near Future, which are very expensive. Yesterday I was thinking about adding a large amount of rtgs for future mining rigs. If my numbers are right, it takes 20 rtgs to provide energy for a single rig, so part count will be through the roof... but I don't have to worry about batteries, fuel cells, solar panels, or heavy and expensive nuclear reactors.

 

Anyway, money usually isn't an issue in career after you upgrade all the buildings. I typically combine base/station/tourist/rescue contracts and also use them to train kerbonauts, and recover the ships at Kerbin. That can give you a lot of money.

 

You can, of course and over time, reuse most of the stuff you put in orbit. So, you send a manned mission to Duna. You return. You leave the transfer stage in orbit, undock the lander (or leave the lander as well, if it's still usable) and recover the lander, or send an SSTO to pick the kerbals. For a later mission, you have an interplanetary tug and maybe a lander already in orbit. Having fuel stations in LKO filled with mining rigs in Minmus or the Mun can also help save money (if that's really needed), because you you can send you bigger ships dry to LKO and fuel them there, which allows you to use smaller lifters.

 

Or you can recover the lifters

[LN400]

Thanks for the replies, guys!

A minor bummer the rigs won't drill and convert while on rails but then it becomes even more important to me to have a solid fuel production rate.

About the mining operation I had in mind:

The whole operation is designed to refuel a small fleet of ships enroute to any other planets (Moho-Eeloo, for instance). As such, it is desirable to produce fuel enough to fill as a minimum a full size 3.75m L+O tank, a 3.75m L tank and a large 1.5m RSC tank for each ship. This will most likely give me enough surplus to refuel the odd lone traveler as well. This will either take a lot of time or a lot of power.

I have abandoned the all-in-one rig for modules, each module a rover connected to the others via the KAS/KIS winch/socket. It gives me full flexibility in positioning each module. One engineer + rover will go with each package for connecting the modules.

Drills are large exclusively, 2 per drilling module, each one connected to a 2x1500 ore tank module, each tank module is connected to a large converter module, each converter is connected to a 2.5m jumbo tank, and each such rig is powered by 4-5 power supply modules, each driven by 4 reactors (145,000 a piece).

In addition, a rover driven by the engineer can hook up to increase production.

At least one of the drilling modules have the narrowband radar and that surface analyses gadget (whassisname), if the ore % varies a great deal in a small-ish area then each drilling rig will be fitted with ore detector equipment.

These modules are going to low gravity moons where shuttles will bring tanks up into low orbit where the fleets will refuel.