Jump to content

KSP Loading... Preview: Breaking Ground - Propeller and Helicopter Blades


St4rdust

Recommended Posts

9 hours ago, Maxsimal said:

3.  The blades have been set up so you can deploy and use the authority limiter to adjust the pitch of the blade - which you can easily do with KAL.  So you don't have to set up a servo system to do variable pitch, and therefore  you can set up a propeller with just a rotor, two blades, and a nosecone and have a workable system.

So, I'm inferring that this will allow for functional collective and cyclic pitch in rotorcraft setup now?  E.G. The pitch of any given blade depends on both the input state and the blades current position within the rotation.  I.E.  If given a 'pitch forward' command, the blades at the 'rear' will increase pitch, while the blades at the 'front' will decrease pitch, thus effecting the 'pitch forward' command (while still allowing for an 'increase all blades pitch' for vertical movement)?

Trying to build a functional swashplate (without obscene part count) was driving me nuts, which was seemingly necessary to get any sort of actual collective control functioning.

Link to comment
Share on other sites

4 hours ago, Shadowmage said:

E.G. The pitch of any given blade depends on both the input state and the blades current position within the rotation.  I.E.  If given a 'pitch forward' command, the blades at the 'rear' will increase pitch, while the blades at the 'front' will decrease pitch, thus effecting the 'pitch forward' command (while still allowing for an 'increase all blades pitch' for vertical movement)?

Depending on how accurately KSP's physics models real life, we may be going down the rabbit hole of unexpected behavior.  When you factor in gyroscopic precession (55 secs in the actual info starts), if you want to pitch a helicopter forward, the blades increase pitch 90 degrees prior to crossing the tail of the helicopter, not at the tail.

Helicopter Lessons in 10 minutes or Less is a good Youtube channel to look through to understand the intricacies of helicopter aerodynamics and physics.  However, since I have no idea how these parts actually interact with the KSP physics simulation, I have no way of knowing whether this is a thing or not.

Edited by Raptor9
Link to comment
Share on other sites

21 hours ago, Maxsimal said:

1.  The helicopter blades in those images were tilted back in a fashion to get them to fit within the image frame.  Like someone took a picture from near the root-end, which is making them look leaf-shaped.   The in-game ones are much closer to what you'd see in a real helicopter.

But are the given sizes correct? are there 2.0m helicopter blades?  if so, what is the difference between the 2.0m helo blade and the 2.0 prop blade, just lift coefficients and offsets or what?

Quote

2.  The propellers, and to a lesser extent the helicopter blades, calculate their velocity/AOA from an artificially offset position.  This allows them to act like they're at a higher RPM than they actually are, to get around the PhysX RPM limit, with the consequent effect on AOA, drag, and lift.   They also use their own aerodynamic values with a lift peak closer to a realistic AOA value than standard KSP aerodynamics.   Pitching your blades is still 100% necessary, but the pitch amounts are more reasonable and realistic.

Ok, this is what I thought it might be. The problem in general with low RPM is low radial velocity. For a given RPM this is increased by increasing the radius. A CoL offset to increase radius increases radial velocity, good. I'm going to guess the CoM is at the base (or offset slightly inward), while the CoL is past the tip, so that centrifugal forces are minimized

Quote

3.  The blades have been set up so you can deploy and use the authority limiter to adjust the pitch of the blade - which you can easily do with KAL.  So you don't have to set up a servo system to do variable pitch, and therefore  you can set up a propeller with just a rotor, two blades, and a nosecone and have a workable system.

I had tried this before with control surfaces and varying deployment/authority limiter. It worked well on pre BG stock rotors:

Spoiler

gnqA2BE.png

But it works funky with BG rotors, at least on tilt-rotor craft:

Observe how this "deployed" blade changes its pitch as it moves forward:

everything looks fine initially:

emUNClB.png

But then the blade snaps to a new orientation (and moves back to normal, but I don't have video)

u2vJxJd.png

See how that blade pitch changed? compare the blade closest to the camera  (the one that has a visible yellow aero force marker farthest to the right) with the other blades, or the previous image.

I hope that behavior has been fixed.

Also you mention the KAL controller, can it just be controlled by axis groups?

 

Link to comment
Share on other sites

11 hours ago, Raptor9 said:

Depending on how accurately KSP's physics models real life, we may be going down the rabbit hole of unexpected behavior.  When you factor in gyroscopic precession (55 secs in the actual info starts), if you want to pitch a helicopter forward, the blades increase pitch 90 degrees prior to crossing the tail of the helicopter, not at the tail.

Indeed; precession was going to be my next inquiry :)  I was specifically posting rather crude approximation of the effect, as when you start getting into real rotorcraft aerodynamics, even most of the plane-jockeys start rolling their eyes at you.

Background:  I build and fly RC rotorcraft; singles, duals, coax, quads.  Everything from basic fixed-pitch, old-school CCPM with flybars, flybarless gyro-stabalized, and full autonomous sensor based navigation.  Have been wanting to be able to build these in KSP for a long time, esp. being able to play around with some of the crazier designs I've thought up (without spending the $ to build and crash a real physical model).

Of course, as you say, KSPs aero might not be up to simulating all of the effects, so the entire thing might be a moot point.

Link to comment
Share on other sites

As far as i'm concerned helicoptors work by magic, and you lot can't convince me otherwise. We don't need all these fancy gyroscopic words when you can just brute force it!

 

If you truly have faith, you will be lifted by the Krakens arms!

Link to comment
Share on other sites

Maybe this will help?

As far as I can tell, there's no word that these things will be congifured to act as control surfaces that respond to pitch/roll/yaw.

That said, KSP's physics engine should reproduce the effects of gyroscopic precession accurately. If they implemented code that allows the helo blades to have proper cyclic control, I would be amazed (and huge respect to them)

Link to comment
Share on other sites

@Raptor9 @Shadowmage Gyroscopic Precession is what pilots are trained to refer to this trait, but the engineering term is phase lag and it's a bit more complicated. Gyroscopic precession plays a part, but aerodynamics and rotor hub design also factor in. The blades don't act like a rigid gyroscope like the bicycle wheel in the above video. The blade have the freedom to move independently (flapping/lead/lag) to each other relative to the hub. What all this boils down to is the control inputs happen at some angle (advance angle) other than 90, so that the helicopter responds correctly to pitch and roll inputs. Every helicopter is different in this regard.

Below is a photo I took during a project at work. We were gathering data for a flight simulator. In the photo the rotor is aligned so that the fore-aft blades respond only to cyclic roll inputs and the lateral blades respond only to cyclic pitch inputs. You can see the blades are not perfectly aligned with the airframe, but rotated slightly, this is the advance angle to account for phase lag. If gyroscopic precession were the only effect the rotors would be square with the airframe.

Swashplate Indexed

 

Link to comment
Share on other sites

1 hour ago, Tonka Crash said:

Gyroscopic Precession is what pilots are trained to refer to this trait, but the engineering term is phase lag and it's a bit more complicated. Gyroscopic precession plays a part, but aerodynamics and rotor hub design also factor in. The blades don't act like a rigid gyroscope like the bicycle wheel in the above video. The blade have the freedom to move independently (flapping/lead/lag) to each other relative to the hub. What all this boils down to is the control inputs happen at some angle (advance angle) other than 90, so that the helicopter responds correctly to pitch and roll inputs. Every helicopter is different in this regard.

I'm quite familiar with phase lag (been flying helos professionally for over a decade) and the engineering techniques involved, like blade pitch horns, offset control servo inputs, etc, but it's not that complicated.  They're just design techniques built into the aircraft's control system to achieve the required input in advance of the desired effect.  However, just so there's no confusion here on the forum, blade flapping is used to counteract dissymetry of lift in directional flight, not affect phase lag.  And the lead/lag motion of a rotor blade (on some rotor systems, not all) is used to compensate for conservation of angular momentum as a result of the flapping motion.

Edited by Raptor9
Link to comment
Share on other sites

I tried 1.7.3 last night and noticed there is a lot to learn about the new propeller blades.

When the blades are mounted to a turboshaft or rotor [centered nearly over the CoM], pitch/yaw/roll can act as cyclic and collective.  The blade-pitch actuation is instantaneous, and is advanced by 90° of rotation of the blades around the shaft.  [Edit, but this is only by coincidence]
Pitch, yaw, and roll response are turned on by default on 'helicopter' blades, off by default on the 'propeller' blades. 

The pitch and yaw and roll axes, for controlling rotating blades, are linked to the orientation of the rotor (as opposed to the craft's control point) and effect pitch and yaw as seen looking along the rotor.
Pitch and yaw(!) work the cyclic, while roll works collective(!).   You need to try it with a prop rotating at 5 RPM so you can see.  
This seems strange for a helicopter, but might be the wisest choice for a game where we build anything we like. 

[Edit: Sadly, no, the propeller blades respond to pitch/yaw/roll only in the same way as normal aero surfaces.  The direction of actuation happens to work nicely for a single-rotor helicopter, including SAS control if we point the control orientation up. (Kerbalx link) For two or more props, the pitch/yaw/roll responses usually make two full cycles per rotation of the prop, so do nothing but shake the propeller apart.]

The adjustments of effective radius where forces act (the acknowledged bit of cheating the physics) might be a little too much.  [Applying the lift so close to the root hides the torques induced by airflow, such as the P-factor that makes a climbing single-engine prop turn left.] Particularly the drag of the blades has very little moment arm with which to slow the rotation of the blade.  Newbies complain the engines are under-powered, but I suspect they quickly figure out how to exploit the blades and engines and are then too busy playing to re-post.  Oldbies point out how unrealistically slick the blades are (link link link).

The blades encourage more craft where the rotor or turbo-shaft torque would naturally be linked to main throttle.  Maybe KSP should make that axis-group link the default for turbo-shafts (along with giving them a sound).

The rotor and turbo-shaft (and robotic actuators) drain EC or fuel in 1.7.3 according to their maximum-Torque × maximum-Rotation-rate while they are moving.  Thus, reducing the maximum-Torque down to the actual torque needed to keep the rotation, saves a lot of fuel/electricity.

The propeller blades do avoid the glitch we saw when the former aero surfaces crossed the CoM while deployed, so 'Deploy' and 'Deploy limit' work nicely for control of collective blade-pitch.  The new propeller blades use the same modules as the old surfaces, so it seems we should be able to figure out some change to configuration files so that the former surfaces fix the bug; then we can use any control surfaces we like on propellers.

Edited by OHara
Kerbalx link showing the one case where cyclic does work
Link to comment
Share on other sites

  • 2 weeks later...

I made a functional swashplate as soon as the DLC dropped. I can confirm that gyroscopic precession is not accurately modeled when the force is aerodynamic lift. It is accurately modeled when the force is thrust from an engine though. I haven't been able to use the new blades yet but I assume they'll behave the same as the blades I used in that helicopter.

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...