Elcano III: Destination KSC

[damerell]

After success on Mun and Minmus in spite of difficulties with the rover, Svetlana and her engineering team have been hard at work on a Mk III design intended to be useful on a wide range of worlds. As a preliminary test, the B-squad of Podpont, Jeanette, and Litzi are going to take the first Mk III on a short trip to KSC. Well, a long trip. Somewhat unwisely, we are once again pursuing the Holy Grail of a rover that won't roll rather than one with nothing to knock off.

From the front, we can see most of the main features. The vehicle is amphibious; while the tracks for land travel should be practically indestructible, there are three independent systems for sea travel (in the hope that at least one keeps working). KF Screws under the fore and rear cockpits should also protect against accidents involving wedge-shaped pieces of land coming up between the tracks. On the rear is mounted a Karbonite scoop and turbojet - while the Karbonite density in Kerbin's atmosphere is very low, the long land traverses may enable us to fill the tank. Finally, on the forward tower there are two electric propellors; they guzzle charge and can only move us slowly across water, but the rover's huge batteries, solar array, and US fuel cell mean charge is readily available.

Unfortunately the US parts, when mounted on the main level of the craft, proved very prone to being destroyed by sea travel. I think this is down to their low crash resistance coupled with the way that they dip in and out of the water. I decided to make a virtue of necessity and mount them in the forward tower, which until then served only to keep the propellors clear of the tracks and as a mount for a forward look-down camera.

The US hexcore holds the standard assortment of hydrogen and oxygen tanks, a fuel cell, an electrolyser, a water purifier, and a KIS storage box for winches intended to make it possible to right the craft.

I'd prefer to use a rover-style cab like the ones in Mobile Frame System or ASET's ERS, but they all turn out either to have no RPM support or hatches on the sides which would be obstructed by the tracks. The Mk I Inline overlooking Mk I works well enough, though.

From the rear we can see the third cockpit, along with a 4-kerbal inflatable habitat for relaxation.

The middle section houses two reaction wheels to help MJ keep us on an even keel, along with life support supplies and a kOS processor. There is plenty of room here or in the forward tower to add more LS supplies for a longer mission, a secondary Karbonite tank, etc.

Edited November 12, 2015 by damerell

[damerell]

This post contains no screenshots. I think I might diverge from my Elcano I and II procedure and post non-IVA screenies even while driving IVA, because at night the IVA view is not very interesting.

The good part is that the design is fundamentally sound; the mole tracks have tremendous power and coupled with the way the screws bite on uneven terrain, the Mk III may be limited to about 18 m/s but it will do that up just about anything in Kerbin gravity. The Mk IV could easily be 40 tonnes without difficulty. Also, the new version of the kOS control which measures the vessel's capacity at startup (in case bits got knocked off) works well.

I added a Karbelectric generator (and kOS code to run it when Karbonite is full and EC isn't) but the Karbonite harvesting rate from air filtering is terrible. For the next design the habitat should go on the top deck so a water filter can be mounted on the rear, with some kind of land drill and sensing equipment. I could run a Karbonite harvest-and-refuel-rover flight, but I've got enough sea-crossing engines anyway.

I started out West and plugged away steadily into the mountains. The other unwelcome discovery (well, I kind of knew it already) is that Kerbin is really quite a lot bigger than Mun. That said, I think the sea parts of the voyage will be quite practically accomplished by plugging the heading and speed into MechJeb, turning on the propellors, and sitting down with a good book.

I've got Scansat satellites up, but it'll take some time to build a map - the lower altitude sat should have had a low-res altimetry sensor as well as a multispectral sensor. (I suppose I could launch a third one...)

Edited November 13, 2015 by damerell

[damerell]

I did send up a third Scansat probe with a low-resolution altimetry sensor. This took long enough for the Sun to rise over the rover, so here is a screenshot of a useful-looking mountain pass ahead:



I hope to get back onto flatter terrain soon.

Edited November 30, 2015 by damerell

[Overland]

Never been to the mun myself but kerbin has quite alot of uneven terrain from the 200km ive seen driving trains..

Your craft looks like a tog II of sorts..a capable and powerful vehicle.. I wish you good fortune and a safe journey :)

Admirable programming too..i envy your skills greatly... Best i can do is lock wheelthrottle to 1. When my foot gets tired on the deadman pedal..
Any chance of a guide? Edited November 17, 2015 by Overland

[damerell]

[quote name='Overland']Never been to the mun myself but kerbin has quite alot of uneven terrain from the 200km ive seen driving trains..

You're the train driver! And you're driving the trains again! I saw your early posts ages ago but didn't realise you were back in, aha, circulation.

I wish I'd seen the Kerbonov cockpit you use; that's definitely going on the Mk IV. A three-kerbal inline pod with good visibility will be a godsend.
 

Quote

Admirable programming too..i envy your skills greatly... Best i can do is lock wheelthrottle to 1. When my foot gets tired on the deadman pedal..

 

Any chance of a guide?

 

 

When I'm back at my games machine I will gladly post the kOS script I use. It's not very sophisticated, actually.

 

 

Conversely I wish I'd hit on your idea of having the regulator and reverser be separate controls. I drive with pedals and a yoke (and a separate throttle) so I have plenty of controls to map. I wonder if I regard mid-mission cockpit control upgrades as acceptable? [1]

 

 

[1] Yes, I do.

 

 

Edited to add: http://www.chiark.greenend.org.uk/~damerell/games/ksp/resources.txt is the kOS script. It depends on parts being tagged in the editor and is specific to the set of resource converters on the rover (although it could very easily be adapted to other vessels and at least all the resource capacities aren't hardcoded... anymore).

 

 

 

Edited November 30, 2015 by damerell
Irretrievably mangled by forum upgrade

[damerell]

I'm going back to the drawing board; we're only two hours out from KSC and I'm already half-planning the Mk IV. Already from driving the Mk III it seems clear that a wider wheelbase and more roll authority is desirable, along with the previously noted issues... and most importantly, it is vital that the pilot can look down as well as forwards, whether by camera or cockpit.

I'll continue in this thread; the IV will basically be a slightly wider and uglier III.

[damerell]

So here we have the Mk IV, a new and improved (ie, stupidly huge) design.


The Karibou cab provides superb visibility, particularly in "are you driving off a cliff" situations. Each track strut has a reaction wheel in as well as a large one behind the fore and aft cockpits; it should be very hard to roll the vehicle on a slope. Nothing above the main deck is critical, although losing the Universal Storage parts would severely impair life support endurance and electricity generation. The NFT mission support girders have empty monoprop tanks for this mission, but the rover could carry a huge supply for landing on very low gravity worlds. Six KF screws provide the main water motive power, as well as protecting the bottom of the rover from bottoming-out. I also gave in and tweakscaled the tracks to 1.2x size, which improves ground clearance.



From the rear we can see the auxiliary cockpit - one-kerbal, but also offering good visibility - and the inflatable shelter. It only holds two kerbals, but that means even if the Karibou cockpit is unusable the mission can continue. Solar generation is, regrettably, reduced by the area needed to inflate the shelter. Twin Karbonite drills will let the mission harvest Karbonite once the scanning satellites find some hotspots (or just by being really patient), which could be used either to drive the stern propellor or the Karbelectric generator. Karbonite is stored in crates above the main deck just in front of the propellor, but also in the adapter behind the auxiliary cockpit.

It's not really visible, but an RTG mounted in the middle of the vessel gives a reserve electricity supply not dependent on solar, ISRU, or the US fuel cell.



This shows the excellent visibility from the driver's position in the Karibou cab. As is usual, a high-mounted camera provides an extra viewpoint.



I set off near sunset, and had travelled 4 degrees West when I took this, approaching a likely-looking mountain pass.

Edited November 30, 2015 by damerell

[damerell]

Top tip for other circumnavigators: ensure you are not going the wrong way. I've been on a bit of a scenic tour of the Grasslands. Then I went to see an anomaly in the mountains. I was going to get out and take a screenshot on EVA, but it turns out to be on a horrendously steep slope; it's all the rover can do to hold position.



I reached the somewhat belated conclusion that since roving in the mountains in the dark is a nuisance, I could shut down for the night, inflate the shelter, and put down the Karbonite drills to see how much I could dig up.



While I was doing this I went and read the Scansat documentation, sigh, and realised I needed a portable Karbonite sample kit - ideally on the rover, but to be used somewhere in the same biome. This seemed like a good time to build an unkerballed biome helicopter to visit the Shores, Grasslands, Highlands, and Mountains, which are all conveniently accessible from KSP, and take samples there. (I could probably manage Desert as well, but my scanning coverage isn't complete enough to really know where I'm going). Sun has risen over the rover in the meantime, so it's probably time to start driving again.

Edited November 30, 2015 by damerell

[damerell]

In the words of the great Jeff Lait, a long time between updates can only mean one thing - a long time between updates. I have rewritten the RPM IVA for the Karibou rover cab to make it more useful, and I've increased the power consumption on the KF tracks - it was striking me as absurd just how much shove one gets for a unit of ElectricCharge.
I got to the West side of the deserts and entered the water. You could see me doing this here except that like all interesting things it happened in the dark.

Another top tip for circumnavigators is that long water crossings will not be the most interesting thing you do in KSP. Beyond throttling up the electric propellors as the Sun rose and then shutting them down at sunset, I had nothing to do but read a book as MechJeb held the ship on course. After a few hours, however, land ho! (As a few small pixels on the horizon). And what's with the biome in this screenshot?

Of course, since the sun is setting in that screenshot, by the time I came ashore it was dark again. Back in the deserts, for which I still haven't done a biome scan for Karbonite, but at least they'll probably be nice and flat for night driving.

 

[damerell]

I've been circumavigating a bit more. I set off into the desert in the hope it would mean some easy flat driving - and it did. A welcome development after the difficulties of the mountains.

I reached the anomaly I'd been aiming for. Why do I always reach these in the dark?

It didn't make any more sense close up, either. I headed North to try and work around the shoulder of the mountains I have completely failed to capture in these screenshots...

...and at dawn I am still steering slightly North to try and get back to the equator, since there are no known anomalies near our course. The desert seems much greener now, but I can see mountains ahead. With any luck sunrise will let me recharge the batteries and electrolyse some water back into hydrogen.

[damerell]

Finally driving during the day! You get some good shadows at sunrise, although the rover is a long way off not needing to run the fuel cell; when the sun gets more overhead the solar panels and the RTG can make headway on the power drain.

I've driven through this sort of beige terrain. Sometimes it's deserts, sometimes it's mountains, now it's grasslands. I don't see much difference myself - the mountains had some slightly bumpy bits, but nothing like the mountains west of KSC.

The mission continues.

[damerell]

This western continent is not the most interesting place on Kerbin. This is basically the first terrain feature I've seen that wasn't gently rolling Grasslands, and from the Scansat coverage I infer I'm pretty well to the Western shore. This update's mostly here so you know I haven't given up.

[damerell]

Seriously, why does everything interesting happen in the dark? I drove over most of a mountain range at night; here I am veering north to avoid the last (huge-looking) peaks. But you can't see much, because it's dark.

Here I am having just crossed the Kerbinternational Date Line. It's still dark.

And here I am having reached the shore and plotted in the course for a very lengthy sea leg over the large eastern ocean. Guess what? Now the scenery will be entirely flat and blue, the sun's come up. At least I can run the electric propellors.

 

[damerell]

This has been an odd leg. I updated my kOS script to run the electric propellers only if electric charge was between 95% and 100% (with a proportionate throttle setting, to find an equilibrium point), layed the course into MechJeb, and went to bed. When I woke up I adjusted the course slightly and went to work.

Finally I find myself ashore again. Sure, much KSP success is in pre-mission preparation, but this 800km leg is a particularly extreme example. Because the craft was right, I've barely touched the controls in circa 20 realtime hours of the mission.

[damerell]

And I'm back with another exciting update. Well, in all candour, an update.
My first thought was that I would drive around the north side of this enormous gulf. I gradually came to realise both that I'd already got diverted onto a peninsula and that sea travel is, in fact, enormously easier than land travel - slower, but given the mission's hefty provisioning, this is hardly an issue.

For once it is daylight. Land ho! Having realised a mariner's life is easiest, I plan to cross a thin neck of land here and sail across the vast lake beyond.

Perhaps (since it is light) the terrain will not be too challenging. And it wasn't, the entire expanse of the neck was easy going.

And in the next lake; lay in the course, check occasionally. I've nearly got out of this one, but it looks much the same on the other side.

[damerell]

Something I forgot to write about last update; in the last on-land screenshot, I am about to realise it's much easier to steer the rover without MechJeb trying to hold you to the at-sea heading with the reaction wheels. :-/

Land ho! Well, land bonk, since I arrived on land in the dark (surprise) and didn't notice. I turn off the tracks at sea, because otherwise they burn electric charge to no good end, but that also means the rover just gently nuzzles against the beach until I deign to pay attention.

Not much to see in this next one, but this is the start of the last (albeit a long) land leg in the route. When this is done, it's a straight sail back to the KSC.

 

[damerell]

Not much to add except that something has stopped kOS from working (IRL I suspect I updated mods with gay abandon and it just can't drive the US parts correctly now).

For the time being, the mission continues; if electricity becomes an issue, we shall try to control the fuel cell via Kronal Vessel Viewer. If that fails Podpont will get out and give the kOS core a good kick. (I'll fix the script in real time).

[damerell]

Here we are rolling away with no kOS:

I liked this shot of the stars ahead over the mountains. I was less pleased to realise that since my ScanSAT coverage was incomplete, I had driven into the mountains.

Given the electric charge issues, I was pleased to see Kerbol had risen.

I have driven up a number of apparent crests to get here, and so I was delighted to see that for once I was clearly on the spine of the mountains, the range stretching away to my left.

Here I am going down what appears to be a glacial valley next to the mountains from the previous shot. This leg has generally included the best looking terrain I've seen yet in KSP. I could well believe I am driving the Karibou on Earth.

Further on, I can see icecaps ahead. We shall steer clear of those.

Although the kerbal's eye view does mean that sometimes one is just here.

 

Edited March 4, 2016 by damerell
finalising

[damerell]

I think what that last comment was meant to mean is that one's screenies can't always be interesting and I happened to shut down somewhere boring.

[damerell]

I changed course to the South. The Waypoint display is leading me on a great circle course but of course I intend to drive up the Equator, more or less.

This Eastern continent is full of good scenery. Sometimes you just come over a ridge and whoah:

Cor!

Blimey!

However, the old vexation that in the dark some surfaces can't be illuminated at all has finally bit me. I have no idea what happened here, besides suddenly being upside down. The loss of an electric propellor before a sea crossing is going to be pretty annoying, but not critical.

Does anyone know of a fix for that? For my part, I think MechJeb's Stability Control needs a bit of tweaking.

[damerell]

I altered course again to shorten my route to the sea - sea travel is just much easier.

Here is the first glimpse of the sea at last - the final bit of land was frustratingly bigger than expected. I must have crested a dozen ridges half-expecting to see water only to be greeted with the sight of the next ridge.

Here's the good ship Mark Four setting off from the coast. Really feeling the lack of one of the propellors with solars going full whack and the fuel cell in reserve - charge to spend and nothing to spend it on.

Here's a cockpit view; I'm steering 275 to try and clear the headland to the West.

 

[damerell]

I awoke in the middle of the night owing to a biological necessity to discover not only had I cleared the first headland but I was at about the right time to steer for the second. Hooray!

Then the alarm woke me to discover... well, good timing, but not so much on the course calculation.

Not long now...
 

 

[damerell]

I must remember the JSI cameras are better in low-light conditions than the ordinary kerbal's eyesight. I can see land here on the external camera which has only been visible out the windscreen because it occludes the stars. You can see from the BTDT track on ScanSAT that I drove over the headland in the previous entry rather than sail around it.

Oddly, I've never been to the abandoned airstrip... well, now I have.

Land ho (again)! The mainland could have been spotted earlier; however, Litzi was asleep and Podpont and Jeanette were attempting to operate the Koffee machine.

As a sting in the tail, as soon as I touched land, I started to get a huge phantom steering input which I'm quite at a loss to explain. I had to limp over to the KSC using MechJeb's Rover Autopilot and yaw control; hence no fancy arrangements, but get the rover to the West end of the runway and shut down. Success!

 

[Geschosskopf]

4 hours ago, damerell said:

As a sting in the tail, as soon as I touched land, I started to get a huge phantom steering input which I'm quite at a loss to explain. I had to limp over to the KSC using MechJeb's Rover Autopilot and yaw control; hence no fancy arrangements, but get the rover to the West end of the runway and shut down. Success!

Congrats!  That was quite a machine and I'm glad it did wig out on you earlier in the trip

 

[damerell]

9 hours ago, Geschosskopf said:

Congrats!  That was quite a machine and I'm glad it did wig out on you earlier in the trip

"Didn't", maybe?
Thanks. I'm glad someone read right to the end (although ultimately I did it for me, sort of thing...)