Jump to content

how to stabilise an Eve propeller lifter


fommil

Recommended Posts

Hi all,

I'm designing an Eve return vehicle and I'm pretty sure I want my first stage to be an electric fan. I've got 4x radially mounted counter-propellors which each consist of a large motor (10% torque, didn't need any more), 81 inclination (optimised for lift, not speed), with 4 large turbine blades. It is able to lift double my expected payload mass at Kerbin sealevel so I expect to get to 15k-ish on Eve before having to dump it for boosters (currently testing that theory). 15k "for free" (1t of electric batteries, no flimsy solar panels) is pretty good going, I can get to orbit (cheating with the debug console) with 5.5k dv, instead of the oft-cited 8k.

My question is: what is the best way to stabilise a low speed propellor lifter? Mine does not behave as I would intuitively expect, so the physics is clearly kerbal's wonky aero/lift model. What I would expect in real life is that I could put the propellors at the top of my lifter, like an amazon delivery drone. That should be very stable because the centre of mass is very far beneath the lift. However what actually happens is that it gets really unstable very quickly and I need to put 6x control surfaces at the bottom (even though I'm going super slow and they shouldn't control anything!), as if it were an unstable rocket. I also find that I get the most stability if I put the propellors near the center of mass like I was placing wings (which is super weird, in real life that'd be pretty silly). I'd really appreciate any explanations of how the propellors are modelled, if anybody is able to go into that kind of detail.

I can add pictures later if there is interest.

Edited by fommil
Link to comment
Share on other sites

Real helicopters are unstable.  If any gust tips one off of vertical, it starts to drift sideways and then swings back and forth in oscillations that grow.  'Dynamically unstable' is the term.

Quad copter drones use differential pitch to get differential thrust on the four rotors.  This can be done in KSP but is not very convenient.   

KSP version 1.9 and later let you enable pitch, yaw, and roll, control on the propeller blades, which makes them change their angle of attack as they rotate, giving more lift on one side of the disk than the other (like the cyclic control on a helicopter).  This is easier to use in KSP because pitch yaw and roll are controlled by SAS.  With four small rotors, the differential lift won't have as much lever arm as differential thrust, but it might still be plenty to stabilize your lifter.

Link to comment
Share on other sites

3 minutes ago, OHara said:

KSP version 1.9 and later let you enable pitch, yaw, and roll, control on the propeller blades, which makes them change their angle of attack as they rotate

thanks OHara. That's something I tried already but it didn't make much difference. However, I only tried it once... maybe I should do some more experiments with the propellors mounted on different locations.

It's also possible that the counter-rotation setup I have is confusing the SAS. I am fairly sure I'm massively overspecced anyway with 8 spinny things. I can try with 4 (2 clockwise, 2 counterclockwise to avoid torque roll) just like a real quadcoptor.

Link to comment
Share on other sites

When I build helos I usually build them with a single pair of counter-rotating rotors on a single axis. That type of helo is very stable. Would that work for your Eve ascent vehicle?

Example helo using this setup:

 

Link to comment
Share on other sites

i never trusted the ksp controls of helicopter, so I always brute-forced the problem by putting lots of reaction wheels and set the SAS to point upwards (one of the two radial directions, i never remember which).

it has the advantage that it doesn't require active control on my part, so i can just skip the long, boring ascent by doing something else in background.

The heavy reaction wheels are discarded in the same stage as with the propeller blades, of course. the mass added is negligible compared to the rest of the rocket

Link to comment
Share on other sites

9 hours ago, QF9E said:

When I build helos I usually build them with a single pair of counter-rotating rotors on a single axis.

Then getting to Eve then becomes the problem :confused:

I've seen people put these inside cargo bays, but that's a step too far in terms of "exploiting Kerbal mechanics" for me.

6 hours ago, king of nowhere said:

i never trusted the ksp controls of helicopter, so I always brute-forced the problem by putting lots of reaction wheels

Heh, that'd be quite a lot of reaction wheels, but an option for sure!

Link to comment
Share on other sites

On 8/7/2022 at 9:28 AM, fommil said:

Then getting to Eve then becomes the problem :confused:

Not really:

I sent this thing to Eve inside a payload fairing as part of a challenge:

 

Edited by QF9E
Link to comment
Share on other sites

On 8/6/2022 at 10:41 PM, fommil said:

Mine does not behave as I would intuitively expect, so the physics is clearly kerbal's wonky aero/lift model. What I would expect in real life is that I could put the propellors at the top of my lifter, like an amazon delivery drone. That should be very stable because the centre of mass is very far beneath the lift.

https://en.wikipedia.org/wiki/Rocket#Pendulum_rocket_fallacy

https://handwiki.org/wiki/Astronomy:Pendulum_rocket_fallacy

It is behaving as it should. You'll have to rely on either differential throttling of the rotors (can be done via a kal controller, but its not ideal and has quite some response lag),

If you enable pitch, yaw, and roll, control on the propeller blades, you can make it work somewhat, but once you increase blade pitch to their optimum AoA, any change (increase or decrease) will decrease lift, so as you approach your max altitude, this won't work either

So lastly, there's the aerodynamic stabilization with good ol' fins/wings. This can work if you climb rapidly, but as you slow to a near hover at maximum altitude, this doesn't work so well either.

That's why my eve ascent vehicle designs were planes flying nearly horizontally

Spoiler

Ljsw0Om.png

Yj2d0in.png

VEFPye5.png

uI6hAz6.png

rm7z9dz.png

tDwYFcp.png

SOtl99x.png

qm0CJfF.png

a0jIPU8.png

qnQEv6U.png

Fully reusable, but re-entry for the orbiter and the lifter is sketchy, then they need to dock on the ground, and be refuelled (flying under electric poeer to a mining station)

 

Link to comment
Share on other sites

15 hours ago, KerikBalm said:

Pendulum_rocket_fallacy

oh wow, that explains it exactly! My intuition is for the lifter to be on a free moving gimbal, but of course it's a rigid body :confused:

Reaction Wheels and Fins it is (at the bottom, which is a shame from a modularity point of view). Control surfaces don't seem to be needed, so maximum lift.

I now have something that can get from the surface to 16.6k on fans, and from there to orbit with 3 Kerbals, most of the time. Margins are tight.

Now all I have to do is figure out how to slow this thing down and land it. The 8 propellors at the top have more of an impact on CoM than I had originally invisioned. Lots of parachutes to pull the top up, I reckon.

Edited by fommil
Link to comment
Share on other sites

  • 1 month later...

Here's the mission log :-)

https://imgur.com/a/VdOPUay

I needed to send the rescue because I (stupidly!) resized the motors before the flight but I think I must have done my testing on empty tanks.

@KerikBalm would using https://spacedock.info/mod/838/Flexible Docking between the propellors (which would be mounted at the top) and the payload help things out and break out of the fallacy?

XZ4akco.png

Edited by fommil
Link to comment
Share on other sites

On 8/8/2022 at 5:11 PM, KerikBalm said:

https://en.wikipedia.org/wiki/Rocket#Pendulum_rocket_fallacy

https://handwiki.org/wiki/Astronomy:Pendulum_rocket_fallacy

It is behaving as it should. You'll have to rely on either differential throttling of the rotors (can be done via a kal controller, but its not ideal and has quite some response lag),

If you enable pitch, yaw, and roll, control on the propeller blades, you can make it work somewhat, but once you increase blade pitch to their optimum AoA, any change (increase or decrease) will decrease lift, so as you approach your max altitude, this won't work either

So lastly, there's the aerodynamic stabilization with good ol' fins/wings. This can work if you climb rapidly, but as you slow to a near hover at maximum altitude, this doesn't work so well either.

That's why my eve ascent vehicle designs were planes flying nearly horizontally

  Reveal hidden contents

Ljsw0Om.png

Yj2d0in.png

VEFPye5.png

uI6hAz6.png

rm7z9dz.png

tDwYFcp.png

SOtl99x.png

qm0CJfF.png

a0jIPU8.png

qnQEv6U.png

Fully reusable, but re-entry for the orbiter and the lifter is sketchy, then they need to dock on the ground, and be refuelled (flying under electric poeer to a mining station)

 

The show off plane as in the need to lift 18 ton of Eve is limited, doing it fully reusable is an incredible achievement however. Remember the Eve SSTO was seen as impossible. 

Link to comment
Share on other sites

16 minutes ago, magnemoe said:

The show off plane as in the need to lift 18 ton of Eve is limited. 

Indeed, it was mainly for the challenge, and secondarily to supply fuel to an eve station - withh the idea of moving Gilly from a moon of Eve to an orbit around the sun similar to that of Dress - otherwise, getting fuel from Gilly is much better.

Also, for a while I was really on a modular surface base binge, with the idea of having standard base modules and a standardized cargobay (the surface-orbit shuttle would be different, but use the same cargobay).

I just couldn't make a design work for Eve, my standard cargobay+40 tons of payload required a plane with far too many parts.

I had to lower the payload mass requirement, and cut the payload bay dimensions (I think that saved 4 tons+ drag).

In a way, by making those concessions, Eve beat me

 

Link to comment
Share on other sites

2 hours ago, fommil said:

Eve beats all of us, eventually.

I mean... I think I could do it, but the partcount increase would make my computer scream in agony, and I would too because of the framerate.

OTOH, I don't need a 40 ton to orbit capability. Most of the modules don't have all that much mass, except for the fuel tank module (rockomax 64) intended for fuel delivery. I've thought I could do a dedicated tanker orbiter (integral fuel tanks, no cargobay to stick an additional fuel tank into). Its what I was doing for my reusable craft to orbit of Kerbin in 3, 4, and 6.25x rescales (1 orbiter for various payloads, 1 orbiter dedicated to just ferrying fuel up, same 1st stage carrier plane).

Dropping the long cargobay+ramp saves 10 tons, granted, that only gets me to 28 tons to orbit (instead of 18), and I would still need to increase payload by another 12 tons to reach the goal of 40 tons. I doubt that could be made up in aerodynamics alone (everything could be inline, instead of having side tanks+engines to accomodate the ramp)

On 9/12/2022 at 4:32 PM, fommil said:
 

@KerikBalm would using https://spacedock.info/mod/838/Flexible Docking between the propellors (which would be mounted at the top) and the payload help things out and break out of the fallacy?

I don't think so, the propellor disks would always have to point up for it to be stable. That seems like it would require active deflection of the joints, not just "flexibility".

Link to comment
Share on other sites

This thread is quite old. Please consider starting a new thread rather than reviving this one.

Join the conversation

You can post now and register later. If you have an account, sign in now to post with your account.
Note: Your post will require moderator approval before it will be visible.

Guest
Reply to this topic...

×   Pasted as rich text.   Paste as plain text instead

  Only 75 emoji are allowed.

×   Your link has been automatically embedded.   Display as a link instead

×   Your previous content has been restored.   Clear editor

×   You cannot paste images directly. Upload or insert images from URL.

×
×
  • Create New...