Is a one day, solar polar race possible?

[Tricky14]

Noticed some mass irregularities in wing parts that may indicate an issue with FAR or Tweakscale. Or both. Will have to ask around a bit before I continue this challenge. Similar craft with the same basic design end up weighing a almost ton extra without any reason.

[Pds314]

17 hours ago, Tricky14 said:

My first attempt (straight North past the halway-point route) has failed, but I got close enough to entice me to try a few more times.

• At the 3:00:00 marker I still had light, as anticipated.
• Around 3:05:00 the shadows started becoming a problem, but I could still quite easily dodge them by staying on the left side of hills and such.
• By 3:15 the sun was still up *but* higher distant hills began blocking it. By my best estimate the sun actually set roughtly 10 minutes later.
• I coasted on my remaining battery charge for something like 30 minutes. The challenge was lost at that point, *but* to my own surprise I only ended up 42.7 km away from my second to final marker, the narrow 'glacial passage', a land bridge that connects the ice sheet with the main continent. At my average cruising speed of about 200 kph that would've only taken me about 15 minutes to drive.

 

 

 

I got sooo close to the edge of it!

 

Things I can still optimize:

1. Route. I think -15 minutes is achievable here, but I'm in an optimistic mood and hoping for -20
2. Craft weight. An earlier, less aerodynamic rover weighed a full ton less than this current one does.
3. @Pds314 suggestion to take a westward approach is something I can only take advantage of past the halfway point, due to the ocean. Even if it doesn't gain me time, it will help me avoid those early shadows of the taller hills in the middle of that continent.
4. Possibly re-arrange the solar panels. I thought the vertical ones were well clear of the rover wheels, but I saw 'blocked by vehicle' maybe 15 minutes before I saw the 'blocked by celestial body' notification.
5. Spin. With the airplane-like aerodyniamics the rover is really stable at higher speeds, but at lower speeds (below 40 m/s) I do sometimes end up in an unrecoverable spin. While I don't lose too much time getting back on track, near the end the electricity cost becomes a bit too high. Traction contol and further wheel balancing will need to be looked into. Is there a downside to having too much traction control?

 

Hmm... Not sure how much efficiency you will lose by starting your westward movement at ~44 degrees north.

A quick eyeballing of the math indicates that the difference will add about 19 degrees of sun progression at 57 m/s. Ergo, the sun will set when you are at ~65 degrees north. That's 6 degrees = ~70 km short of the latitude of eternal sun for 57 m/s. However, increasing the rover speed helps on all fronts.

Edited October 8, 2018 by Pds314

[Tricky14]

I'm not sure how much I can still improve on that. The FAR aerodynamics are already as optimal as they can be, so the only thing holding it back at this point is the frontal profile of the rover. I unsuccessfully tried fitting a kerbal in a Tweakscaled (Tiny) sized service bay before. It might fit in horizontally, but then I'm pretty sure the Valentina will just clip through the length of the craft.

Before trying that though I need to figure out what is wrong with FAR / Tweakscale. Yesterday I noticed for the first time that I have been getting incorrect mass sums for the entire craft in the planner. My first run might have been entirely rigged.

 

Edit: I think I may just circumvent that issue for now by manually calculating the craft weight to verify the parts I use aren't broken & switch to different parts if they are. It's not ideal, but I'm having a sick day and I need something to do.

 

Edited October 8, 2018 by Tricky14

[Tricky14]

It seems that for now that the plane parts only bug out when they are connected to the root part. My current craft is almost half the weight of the last one, at 0.8 tons. The wings are really flimsy as a result, but they hold together fine because it literally is a leaf. Very little battery capacity, but it's light so it doesn't need much to get going (also it's nice to have an upper limit on how much I can coast by nightfall). Still climbs steep hills like a spider, without losing much speed. I think that will do a lot to make up for time.

 

The weirdest thing yet is that I can actually fit a Kerbal in there, horizontallly. And it's so self-stablizing that I sometimes forget to turn on SAS during a gliding jump.

High hopes for this one.

Edited October 8, 2018 by Tricky14

[Triop]

1 hour ago, Tricky14 said:

My current craft is almost half the weight of the last one, at 0.8 tons.

Cool design, why not also slap some solar panels on the wings ?

[It'snorocketscience]

On 9/30/2018 at 3:23 PM, 5thHorseman said:

Kerbin is 600km in radius so you need to go 2*pi*600/4 or about 942km, or let's say 1000 for easy math and to account for deviations.

You have 3 hours from sunup to sundown.

To travel 1000km in 3 hours, you need to go 333km/hr or 5.55km/m or 92.5m/s.

That's about 1.6 times your max speed.

I don't know how much more time you get at the poles due to the sun taking longer to set, but don't think it's that much? If it gave you an entire extra hour you'd need to go 250km/hr or 4.17km/m or about 70m/s, which is still not doable.

Also you'd have to drive a rover for 3 (or 4, or 5) hours non-stop without making a single mistake, or multiple quicksaves that eat into your average speed.

There is a kinda exploit-y way to surpass 92.5 m/s using rover wheels only. I'm reluctant to show this as this is kind of fun and i sorta-kinda hope that the only use for giant rover wheels doesn't get patched out, but anyways:

150 m/s+ can be achieved using the giant rover wheels. There are two ways: 

1. Disable symmetry and invert steering for all wheels on one side. When you try to turn, the craft will go straight (because you inverted steering on one side) and accelerate past 150 m/s.
2. Spam ADADAD to steer rapidly. It's hard to control and a bit slower but doesn't need any special setup.

How does this work? The giant rover wheels steer like tanks: they increase power on one side to steer.

Will you guys allow this? It can be done in game without editing files.. and it burns electricity like crazy... regular solar panels fly off at those speeds, and you'll need plenty of fuel for fuel cells OR tons of flat static solar panels to keep this running to the North pole. At these speeds, even with the wheels' insane 150 m/s crash tolerance, careful control and structural design will be crucial.

You can see this madness for yourself in a video by hazard-ish. I verified it myself and this hasn't been patched yet.

 

Edited October 9, 2018 by It'snorocketscience
edited for clarity

[Tricky14]

I'm aware. It kinda takes away the challenge though. Too gimpy.

On 10/9/2018 at 1:19 AM, Triop said:

Cool design, why not also slap some solar panels on the wings ?

I might, but I'm not quite sure if it needs more. Weighing only 0.8 tons, it needs very little to get going. I can presently accelerate for something like 40 seconds straight on a single full battery and the only time I really stand to lose a lot of velocity is during the mountainous parts of the middle section of the journey, during which sunlight is readily available. Then at the polar ice cap I can pretty much keep accelerating so long as there is *some* trickle of electricity. That only leaves the early parts of the race, the hills casting long shadows north of the KSC. A couple trial runs showed me it's possible to clear them without losing too much speed.

Edited October 10, 2018 by Tricky14

[Tricky14]

On 10/8/2018 at 11:26 PM, Tricky14 said:

 

So here is something interesting I don't quite yet know what to do with. If I invert the wheel base, I get two nice little platforms for solar panels at each side of the rover. Unfortunately it also makes the wheel base really quite rectangular and tips easily even with SAS enabled at 100%. Wider wing parts increase the profile of the craft significantly and reduce overall top speed.

The solution seems to be to just make the rover shorter and shorter until the wheelbase somewhat resembles a square again. But now the wheelbase is so tiny it still offers little to no extra stability, even with the main body of the craft lowered to just above the ground. I'd love to solve this problem right now and take advantage of the extra side panels, but for now I have to leave you with a screenshot of this horrible, mangled creation. It exists only to suffer.

 

"kill meeee"

Edited October 10, 2018 by Tricky14

[Tricky14]

Early / updated concept drawing. I will get there.

 

[Triop]

That concept drawing is a real piece of art.

[Tricky14]

Once you start rotating parts it's kinda hard to make the lot of them fit together. So, a failed attempt. This bird is also super floppy and a bit on the slow side. Only the mass and general shape were on target.

It got me thinking though, a car that's part boat is obviously an amphibian car. Is there a nice word for a car that's also actively a glider (not the folding-out type car-plane dealies)?

 

I really wish there were rover wheels (of all existing types) you can place directly below a craft, sort of like a shopping cart wheels. That'd solve so many placement issues.

Edited October 12, 2018 by Tricky14

[Tricky14]

Bit of a busy weekend, so didn't have too much time to play or do another trial run. 9-ish prototypes onward, it's kinda dawning on me that the problem is accelleration, or maintaining a constant speed. I can't make a faster rover, but I figured I could try and make KERBIN SMALLER *coughs*. So here a larger rover with more wheelie parts:

Actually accellerates about as fast as my lightest rovers do and reaches the fastest possible land speed on Kerbin with electric motors of about 57.2 m/s under FAR. I like how the wheels are spread to form sort of an imaginary circle around the center of mass, it must surely make it very stable to drive. It's something like twice as heavy as the lightest craft I can make and while my initial thought was 'well, I'm sure having the suspension doubled too will lessen the shock impact on the wheels, but ultimately there are too many situations at downhill speeds where only one or only two wheels make contact with the ground, causing damage.  It's probably safe-ish to use on the longer, less hilly route north, but I'm not interested in longer right now.

I need to think of something radically different to reduce mass, like an SAS unicycle or something. I'm not quite certain how I can make that easy to control or aerodynamically sound though. I will have to think more thoughts.

By the way, I could just Tweakscale the wheels/engines to something suitable too, but it seems like too much of a cheat. I'm sure not everyone approves of my current use of tweakscaled and FAR rebalanced wing parts either, but the truth of the matter is that there's just not enough Tiny and Small counterparts of most wing parts (heck of most parts). If I were in charge of Squad, I would have made sure there are versions of all part types available in all standard sizes. As it stands, Tweakscale is the closest I get to this even being a fair challenge.

Edited October 15, 2018 by Tricky14

[Tricky14]

FAR seems to act like there's a bit less air resistance since the most recent KSP update, so I'm holding off a bit until I figure out what's going on with that.

I'm not sure if the challenge is possible without FAR. It would certainly be possible to have a rover not use aerodynamics to stabilize itself, but I think even with ways to adjust for that, I'm still looking at a 10 m/s on average loss. While I'd still have Tweakscale, I'd lose the ability to further tweak the material properties of the wing parts, which is highly needed at the smallest scales.

I'll give it a week or two, considering the state FAR already is in. I'll decide what to do after that.

Edited October 17, 2018 by Tricky14

[Cat91]

Well, Kerbin got no axial tilt, so wherever you are the daylight always only last 3 hours.

[Tricky14]

I've recompiled Ferram Aerospace Research, but the air resistance is still off. Moreover, my center of lift seems bugged in the editor, any older rovers I made seem to require more power to SAS to prolong their gliding capabilities.

It's uninstalled now and I'm not sure where to take this challenge next. There's nothing special about driving a regular rover to the North pole, so I'd have to think of something new. A different option might be replacing FAR with Infernal Robotics. There are some pretty neat tricks you can do with that to further increase stability of rovers. It's not going to make up for the average loss of 10 m/s, but it should make crossing the mountainous regions an afterthought. That's kinda nice I guess.

Edited October 18, 2018 by Tricky14

[Tricky14]

Ok, so IR seems quite out of date...

What is happening?

[Tricky14]

I ran some test and the FAR-less loss of speed only ended up amounting to a couple m/s. Something pretty big has changed, that's for sure. Maybe less air resistance overall?

[MaverickSawyer]

17 minutes ago, Tricky14 said:

I ran some test and the FAR-less loss of speed only ended up amounting to a couple m/s. Something pretty big has changed, that's for sure. Maybe less air resistance overall?

There's been some reports of wonky aerodynamics in stock 1.5.0, yeah... but when was the last time you ran without FAR?

[Tricky14]

2 hours ago, Tricky14 said:

I ran some test and the FAR-less loss of speed only ended up amounting to a couple m/s. Something pretty big has changed, that's for sure. Maybe less air resistance overall?

Actually strike that. I tested a craft with the same frontal profile as my other rovers and the average speed dropped close to 10 m/s. Only if I build them thin like rockets do I not lose too much speed. I had just forgotten how vanilla works differently than FAR. A slidly wider profile in vanilla and the craft basically becomes a brick. It's really noticable too, none of the wing parts I use to extend the wheel base prevent the craft from becoming too brick-like.

2 hours ago, MaverickSawyer said:

There's been some reports of wonky aerodynamics in stock 1.5.0, yeah... but when was the last time you ran without FAR?

I don't remember a time when I played the game without it. The difference is night and day.

Well that's not true, I did reinstall before starting this challenge and build a couple of rovers. Vanilla KSP I just never saw a rover go much faster than 46-48 m/s. 57 brings me to the edge of the perma-day zone on Kerbin. 46 isn't going to cut it.

Edited October 18, 2018 by Tricky14

[MaverickSawyer]

8 hours ago, Tricky14 said:

Actually strike that. I tested a craft with the same frontal profile as my other rovers and the average speed dropped close to 10 m/s. Only if I build them thin like rockets do I not lose too much speed. I had just forgotten how vanilla works differently than FAR. A slidly wider profile in vanilla and the craft basically becomes a brick. It's really noticable too, none of the wing parts I use to extend the wheel base prevent the craft from becoming too brick-like.

I don't remember a time when I played the game without it. The difference is night and day.

Well that's not true, I did reinstall before starting this challenge and build a couple of rovers. Vanilla KSP I just never saw a rover go much faster than 46-48 m/s. 57 brings me to the edge of the perma-day zone on Kerbin. 46 isn't going to cut it.

So, you probably are still thinking of the ol "Soup Bowl" atmo model...  Yeah, that's been gone for some time.

With stock, anything radially attached (unless in a cargo bay) is draggy af. Not sure how FAR handles it, but having a lot of solar panels attached to the side would absolutely make it slow down.

And you're trying to keep it vanilla parts, just with FAR? Hmmm... I think you'll hit a wall beyond which you cannot further optimize the design, and you're still going to be just short of what you want. You may HAVE to use a battery jump start to make the run in time.

[Tricky14]

On 10/19/2018 at 7:24 AM, MaverickSawyer said:

So, you probably are still thinking of the ol "Soup Bowl" atmo model...  Yeah, that's been gone for some time.

With stock, anything radially attached (unless in a cargo bay) is draggy af. Not sure how FAR handles it, but having a lot of solar panels attached to the side would absolutely make it slow down.

And you're trying to keep it vanilla parts, just with FAR? Hmmm... I think you'll hit a wall beyond which you cannot further optimize the design, and you're still going to be just short of what you want. You may HAVE to use a battery jump start to make the run in time.

So attaching wing parts through a cargo bay should work? I might as well try that. If I can get that 10 m/s back I'll call it even.

Update: eh no, cargo bays did not solve the problem. Well you did write *in* a cargo bay...

 

Maybe I just need to make a cargo bay as big as Kerbin.

[Tricky14]

So since the thread is pretty much on hold until FAR is resurrected from the dead, I deemed it appropriate to just make a big old party boat and all the serious rover business be damned for the time being. Mentally add your favourite White Stripes song.

 

[Pds314]

You know, if you're willing to forego timewarp, you can probably avoid wheel-driven power altogether in favor of putting electric fans on your rover. Then, provided suspension can handle it, you can probably do pretty crazy speeds.

Edited October 23, 2018 by Pds314

[vyznev]

12 hours ago, Pds314 said:

You know, if you're willing to forego timewarp, you can probably avoid wheel-driven power altogether in favor of putting electric fans on your rover. Then, provided suspension can handle it, you can probably do pretty crazy speeds.

Yeah, but running stock electric props on solar power has its own set of challenges. There was some discussion of it in another thread a while ago, but I don't recall anyone actually managing to get it working efficiently:

 

[Pds314]

6 hours ago, vyznev said:

Yeah, but running stock electric props on solar power has its own set of challenges. There was some discussion of it in another thread a while ago, but I don't recall anyone actually managing to get it working efficiently:

 

Hold my ginger ale.

Edited October 23, 2018 by Pds314