[Stock/DLC] Brikoleur's Guide to Helicopters

[Guest]

Level: Intermediate/Advanced
Craft used to illustrate this tutorial: BAK-52NS
Version history:

1.2 - Updated with a note on 1.7.3 built-in rotor and propeller blades
1.1 - Updated with better rotors, thanks to a tip from @Hotel26
1.0 - Original version
 

About this tutorial

This tutorial is a basic primer on stock helicopters made with parts from the Breaking Ground DLC. It does not discuss pre-Breaking Ground stock rotary motors, nor helicopters made with mod parts. I have limited experience with both and it would expand the scope of the tutorial rather too much. I also do not claim to being the inventor of any of the construction techniques or principles discussed here; a quite a bit I have discovered on my own, and a quite a bit I have picked up around the forums. If you feel you ought to be credited, please say so and I'll add you.

What's a helicopter?

A helicopter is an aircraft that flies by producing lift from one or more powered rotary wings, or rotors. If the rotor is not powered it is not a helicopter, it is an autogyro; they are also very cool but out of scope of this tutorial. And if the rotor is not used to produce lift but for some other purpose -- thrust, for example -- then it is not a helicopter either. Helicopters can have other forms of propulsion as well: real-life choppers with jet engines bolted on exist and work well. If it's necessary to make the distinction, they are known as compound helicopters.

This is a helicopter. It's the BAK-52NS. This variant uses hydraulically sprung and damped landing skids instead of wheels, making precision landings easy...ish.

How is it different from an airplane?

Airplanes fly by producing lift from airflow around wings. They need to be moving forward to do this and stay in the air. With helicopters, the spinning rotor moves the lifting surface through the air, producing lift. This allows them to hover. However, the big rotating propeller on top of the craft produces a whole set of complications, many of which are shared by kerbal helicopters and human ones; others however are specific to one or the other because kerbal physics aren't quite like real-life physics, and stock kerbals lack certain highly useful bits and pieces used to make human choppers more manageable. On the other hand, kerbals have some amazingly powerful components to build with.

Cyclic and Collective

Another obvious difference between a plane and a helicopter is how they're controlled. Planes are controlled by moving control surfaces -- rudder, ailerons, elevators, and canards -- which modify the lift produced by each lifting surface, applying forces to the plane and causing it to turn. Pull the stick back, and the control surfaces move to produce more lift near the nose and less lift near the tail, pitching the nose up; push it right, and port control surfaces move to produce more lift while starboard ones produce less, causing the plane to roll to the right. Since helicopters need to be controllable even when they're hovering, they work differently. The primary controls on a chopper are cyclic and collective. 

Cyclic means adjusting the pitch of the rotor blades differently depending where they are in the cycle of rotation. Imagine that your chopper sits in the middle of a clock face, nose pointing at 12 o'clock. Now, if you want to pitch up, you will want the blades to increase their pitch as they near the 12 o'clock position, and decrease their pitch as they near six o'clock, thereby producing more lift towards the front and less towards the back. You'll also want to adjust cyclic as you start going faster: if your rotor spins counterclockwise, the blades at three o'clock will have a faster airflow over them than the blades at 9 o'clock, because the airflow from your forward motion will get added to the airflow produced by the rotor's rotation. This means you'll want increased pitch around 9 o'clock and decreased pitch around 3 o'clock, or else your craft will roll to the left.

This makes helicopters rather hard to fly in real life as well as on Kerbin. What's more, kerbals have no direct control over cyclic: instead, when you adjust the pitch, yaw, or roll, the magic control surfaces try to figure out what you want them to do. This works acceptably with regular aircraft; with helicopters, not so much. So cyclic control on Kerbin is crude at best and you will need partial or total workarounds for this.

~ * ~

UPDATE: FooFighter has built a working swash plate with collective and cyclic control. If you want to make a realistic helicopter that is controlled without reaction wheels, now it's possible! 

• https://kerbalx.com/FooFighter/Swashplate

~ * ~

Collective is a much simpler proposition: it just means the average blade pitch on the rotor. Increase collective and the rotor produces more lift, causing you to gain altitude. Increase it more and your motor will run out of torque to spin the rotor: the RPM will drop and eventually the rotor won't be able to produce any more lift. You'll leap up and then drop down again. Increase it too much, and your rotor will stall, causing you to plummet rather precipitately. And conversely, decrease collective to descend and reduce the torque needed to spin the rotor, allowing it to rotate faster.

Collective gives really fine control over hover, and makes a helicopter extremely responsive in vertical motion, comparable in KSP only to a wildly overpowered rocket-powered VTOL. Thankfully, it is possible to make a really nice collective in kerbal helicopters.

Perhaps surprisingly, hover on a helicopter isn't actually controlled by throttle. The motor's job is just to keep the rotor spinning; collective and cyclic do the rest.

Torque effects

In addition to the asymmetrical aerodynamic effects described above, rotorcraft have one more issue to contend with: torque. Spinning up a rotor and, when flying, pushing against the air to produce lift requires torque. Because Sir Isaac Newton is no fun with his laws of motion, this torque will have to get transferred somewhere in an equal but opposing manner. If you don't want your helicopter to spin in the opposite direction of the rotor, you will have to find some way to balance out the torque produced by spinning the rotor.

Most real-life helicopters do this with a tail rotor: the helicopter has a pretty long tail which works like a lever arm, and at the tip of the tail is a propeller producing thrust in the opposite direction of the main rotor's torque. The pilot controls the pitch of the tail rotor using yaw controls, and will in fact be continuously adjusting it in different flight conditions (unless he has a computer to do it for him). 

Sadly, this does not work all that well in KSP. It is possible to make a smallish single-rotor/tail-rotor that is somewhat controllable, but it is hard, it won't be all that easy to fly, and it will very likely require a lot of reaction wheels to paper things over. That's why we're going to discuss a different type of helicopter here: one that flies with twin coaxial contra-rotating rotors. This solution neatly balances out the asymmetrical torque and aerodynamic effects, making for a stable, neutral basis for your craft.

By all means attempt to make a conventional main rotor/tail rotor helicopter. Just expect it to be quite hard!

This has real-life counterparts as well, notably the Soviet/Russian Kamov Ka-50 and its relatives, and the solution is used there for the same reason it works for kerbals. It makes the craft stabler and easier to fly.

By Dmitriy Pichugin - http://www.airliners.net/photo/Russia---Air/Kamov-Ka-50/0920728/L/, GFDL 1.2, https://commons.wikimedia.org/w/index.php?curid=5896037

The coaxial contra-rotating twin rotor powertrain

The simplest kerbal rotorcraft powertrain uses a similar solution as in the Ka-50. Kerbals have the advantage of having incredibly powerful, yet compact electric motors that can be placed anywhere, so that's what we're going to do. The powertrain only consists of two parts: at the top a motor (the standard or heavy electric rotor work well for most craft), and below it, a flat servo with its motor disengaged (with no motor at all). The rotor blades attach to the motor above, and the freewheeling servo (or the bottom half of the motor) below. When you spin up the motor, the torque will be evenly split between the two rotors, which will start spinning in opposite directions.

Note: this isn't the only way to make a contra-rotating powertrain; you can also use two electric motors surface-mounted to a base, then gizmoed into being coaxial; in this case, each motor will be spinning its own rotor. It has twice the power. For most purposes, the single-motor/freewheel solution is sufficient, however, and has the advantage of being simpler and stabler.

Collective

Since KSP 1.7.3, Breaking Ground includes propeller and rotor blades as parts. Clip them onto a motor, deploy them, and bind their authority limit to an axis group to control collective (e.g. up/down). Note that they come in clockwise and counterclockwise variants: if building a contra-rotating powertrain, be sure to use mirrored variants for each rotor so that the marking decals point the same way on each, and set the deploy direction on each of them so that adjusting collective up increases pitch on both of them.

When building your own rotors (see below), mount an elevon on a servo as pictured above, limit the servo's angle to some relatively sane values, and bind it to an axis group as above. 

Rotor design 

The built-in rotor and propeller blades differ greatly in performance from ones made from elevons. They are much more powerful in the lower atmosphere, producing a great deal more thrust/lift. However, their performance drops off much more abruptly and their service ceiling is much lower. A craft powered with a rotor made from elevons can reach 20 km on Kerbin and operate easily on Duna. Therefore, for such special off-world uses, hand-built rotors still have a niche. 

With rotors, light weight is everything, so use the lightest components available for the job.  Your rotor blades should be control surfaces -- FAT-455 for bigger rotors, elevons of various sizes for smaller ones. Here's the best way I know to make a rotor:

1. Place servos onto the motor or the freewheel in radial symmetry. Small ones work most of the time; for very big rotors you might want to use larger sizes.
2. Attach a control surface to the servo and rotate it to the correct orientation. 
3. Hold down the shift key and offset it outwards to your desired radius.
4. Set the angle restrictions on the servo. Values of about 12 to about 35 degrees depending on rotor size work for me.
5. If making a bigger rotor, add a second control surface and repeat step 3 for it.
6. Optional: add a strut connector from the servo to the nearest control surface. It won't do anything much but it will make it look better.
7. Copy the entire blade assembly onto your other power element and turn it upside down.
8. Assign servo angle on both sets of servos to up/down, reversing one of them.

Important: Disable yaw control on all the control surfaces on your rotor, leaving pitch and roll enabled.

Powering it

Rotorcraft require electricity to run the powertrain (and also operate collective). Small craft like the BAK-52NS "Kranefly" above could actually run just on a pair of the larger solar panels, or you could bring enough batteries to give you the endurance you want, but the all-around easiest solution is to use fuel cells as above: the golden tank contains enough fuel to fly the Kranefly for probably longer than you have patience, and it only needs a few cells to run. For the heavy rotors you pretty much have to use fuel cells; a pair of large fuel cell arrays is sufficient to power a single heavy electric motor.

Controlling it

You can set up whatever control scheme you like of course, but I have found the following to work for most things:

Action group 1	Toggle fuel cells and engage motor(s)
Main throttle[1]	Adjust engine torque (you'll want this at maximum most of the time)
Up/Down axis	Adjust collective (K increases pitch, I decreases pitch -- this places them at the same positions on your right hand as pitch on your left)

[1] Since 1.7.2, F/B in 1.7.0-1.7.1

Additionally, brake will apply brake on the motor driving the rotor. Because you have a freewheel between the rotors and the craft's body, this means you can stop the rotor very quickly by disengaging the motor (action group 1) and hitting the brakes -- both rotors will stop with the torque canceled out between them.

The magic of reaction wheels

Kerbals may not have cyclic but by the Kraken's tentacles they have reaction wheels. You can paper over minor misbehaviours in the craft by adding some reaction wheels... sometimes quite a lot really. Don't feel bad, it's a kerbal solution.

Tuning it

The powertrain described above is fairly docile and you can stick it on top of the centre of mass of pretty much any craft light enough for it to lift, and it will fly and hover. Getting it to fly well is a different kettle of fish altogether. If there is a science to tuning kerbal rotorcraft I haven't discovered it -- all of my tuning has been through trial and error. I suspect the unpredictability is due to the way KSP translates control inputs into control surface positions on the rotor, which is a bit on the flaky side:

• Change the number of rotor blades. I've had good results with rotors from 2 to 6 blades. More blades require more power but run smoother.
• Adjust blade length. Larger rotors are more efficient but less stable unless you feed them with more power.
• Move rotor forward/aft. Moving it forward and back changes the craft's tendency to pitch forward or back as you increase/decrease collective; it also changes its sensitivity to roll and yaw controls although I have no idea exactly why and how. Even tiny adjustments can make massive differences; less than a "click" of snap-to motion can completely change the handling characteristics of a chopper. I suspect this is due to the way the rotor blades respond to your control inputs.
• Move rotor up/down. Up tends to make the chopper more stable but less responsive to control inputs, down does the opposite. It's quite possible to make a really numb chopper that only goes up and down and barely even responds to pitch, roll, or yaw controls!
• Tilt rotor forward. It does something so it's worth a try.
• Adjust control authority. Less authority means less judder but less control; more does the opposite (and might cause blade stalls which is no fun at all).
• Adjust the craft's centre of mass. Generally speaking you will want a high centre of mass, close to the rotor: this is why the fuel tank is right below the powertrain in the BAK-52 above.
• Add or remove reaction wheels.

Tip: Tune with SAS off. You might find that your chopper flies rather pleasantly without it in fact!

Flying it

To fly a helicopter, spin up the rotors with collective at zero, engines at maximum torque. Then increase collective until it takes off. Pitch to accelerate, slow down, or fly backwards; roll to fly sideways, yaw to spin around. When you're moving forward at a decent pace airplane-like aerodynamics start to enter the picture which is fun and different.

Developing it further

The basic Ka-50 style craft plan is just one possibility among many. Once you've got the power train figured out, you can make bigger ones and smaller ones, choppers powered by more than one set of rotors in a variety of configurations, tilt rotors with heavy servos making for an Osprey-style VTOL craft, and so on. You can stick on a jet or two just below the rotor assembly to make it go faster -- making fast choppers is a completely different and much harder challenge than making fast planes, since the limiting factor is stability rather than thrust to weight ratio; you will need to design rather different rotors for choppers that go very fast. You can also attempt different solutions altogether, like with non-coaxial contra-rotating rotors, or even attempting a main rotor/tail rotor style craft. There's a lot of room for tuning in rotor design as well, and if you feel the stock electrics don't quite produce the oomph you want, research turboprops and start breaking records (ht: @Azimech).

You might have to get creative to find a practical use for helicopters in career missions but they are a lot of fun to build and, eventually, to fly. There are at least two helipads on the KSC just begging to be used, so go out and use them!

Edited August 2, 2019 by Guest

[Geschosskopf]

Very informative and quite well-written.  Thanks!

[Hotel26]

14 hours ago, Brikoleur said:

Helicopters might not have much practical use in career missions but they are a lot of fun to build and, eventually, to fly.

Yes, great tutorial, thoroughly well-researched and written!

I use helos for my Accident Investigation crews.  Including Air-Sea rescues.  I have a rule that nothing can be recovered from the surface of Kerbin (except at BKB or KSC), unless a manned flight within []-switch distance is in attendance.  For VIPs living in very remote eyries, helos are also essential.

For example, I run a regular Kingfisher service into Kamp David, 300 klicks to the north of KSC...

craft courtesy @FleshJeb screenshot without permission [too beautiful to withhold any longer]

 

Edited June 6, 2019 by Hotel26

[FleshJeb]

On 6/6/2019 at 4:46 AM, Hotel26 said:

craft courtesy @FleshJeb screenshot without permission [too beautiful to withhold any longer]

Is this your way of trying to get me to rig that thing as an actual helicopter?  (PM me about it)

Major obstacles:

1) I’m frightened of non-coax. (If I do that, it’s getting a Fenestron)

2) I want it to still do Mach 3. (I think I can enclose the rotor assembly in a fairing that will be ok)

 

[Hotel26]

5 hours ago, FleshJeb said:

I want it to still do Mach 3

I had thought about this and I don't think there's any way KSP can handle real rotor blades at that speed; so, no: I am perfectly happy the way it is!  It's beautiful.  And so much fun to fly on air-sea rescue...

I'm glad (or hope) you're not mad at me for posting that (on the spur of the moment)!  I like it just the way it is, particularly without the tail rotor!

[Klapaucius]

This is really great, and very well written. Thank you for taking the time to write this up.

[fulgur]

Thank you so much!

I would rep you but I re-read E:O0 today, so I am out of rep.

[Lapis]

this looks nice, i used to be good at stock motors. Breaking ground now lets me build them without.... complications.

[James Kerman]

Congratulations @Brikoleur, your comprehensive tutorial has been awarded thread of the month.

[Sppion1]

Hi @Brikoleur

I've some difficulties to make a working rotorcraft....

I've a simple one : main ship part, a servo (no motor, I'll call him "main servo"), and a rotor part as described above, with a rotor, 4 servo and 4 "elevon 1".

When the main servo is locked, the torque effect make the main ship part spinning, but at least it flies until the main part spin so much that the rotor spin less, and not produce enough lift. It then fall down, hit the ground, the main part stop spinning, and it flies again. When I unlock the main servo (which as 4 servo and 4 "elevon 1" attach, as the main one, but in the opposite direction), It won't lift off. And I don't know what to do. can it possibly be due to FAR ?

Edited July 21, 2019 by Sppion1

[Guest]

Going by what you describe and how your craft looks, it ought to fly just fine with the main servo unlocked. It's probably some fairly trivial thing that you're doing wrong, or it just possibly could be FAR although as far as I know FAR works just fine with rotors.

Some things to check:

1. With the main servo unlocked, when you increase torque on the motor, do the rotors start spinning in opposite directions as you intend?
   • If not, disassemble and reassemble the servo/motor/small servo assembly -- something's not attached where you think it is. For example, the blades attached to the servo might have clipped to its bottom part, which means they rotate with the body, not with the upper half of the servo. You can try attaching them to the bottom half of the motor also.
2. When you increase/decrease collective with no power on the rotor, do all the blades change their angle like you intend?
   • If some of the blades don't do so, re-assign the servo angle to the axis that controls it in the action group set -- there's a bug that causes some of them to be "lost" if you detach/reattach them. 
   • If they don't or they move in the wrong direction, re-assign them with the "invert" box checked on one set so they do.
3. Do you have pitch/yaw/roll disabled on the blades? 
   • If not, do so. That eliminates one source of potential problems. You can re-enable pitch and roll later once you've achieved lift-off.

I can't tell what mods you're using apart from FAR and I'm not super familiar with them, but it is also possible some of them might be causing problems. 

If I was a betting man, my money would be on (1) above -- you've inadvertently attached the blades to the bottom half of the servo, not the top half like you meant to. In any case, your basic design is sound and it ought to fly just fine.

[Sppion1]

Thanks for your reply !
1) Yes, the rotor part spin clockwise, the servo part couterclockwise.
2) I can't make secondary axis to work, so I do that manually at the moment. I put all the servos manually in the right angle (12° for the eight blade, which was enough when main servo was locked). You can actually see that they are turned in the picture above. (to get an idea of the angle). So yes, all "elevon 1" have a 12° AoA.

3) I disabled it, and it isn't changing anything :'(

FAR is the only one that may affect the way part interact with atmosphere. Other are one graphical, KER, KIS, KAS, navutilities, science alert.

didn't know I can attach them to the bottom half of the motor. tried it, unfortunately doesn't change anything....

[Sppion1]

@Brikoleur

I actually removed only FAR, and its kinda works. it flies a bit, but even with the main servo unlocked, and spinning, the main part is also spinning.... So the "rotation servo M-06 doesn't cut all torque effect.

I don't know why FAR got that problem, because with servo locked, it flied with FAR, even if the body was spinning.

So.... still many questions to solve.

Anyway, Thanks for your replies.

[Guest]

Do you have damping set to zero on the servo?

Do you have any reaction wheels on the craft?

[Sppion1]

Dampling ? interesting. And yeah forgot the reaction wheel.

[Sppion1]

Eventually, that's a lot better, even if there is still a bit of spînning, but it might be controllable, with more reaction wheel, or whatever.
About FAR, it seems to be known :

 

 

So thanks for all @Brikoleur

[jpinard]

@Brikoleur do you have to do those little girder pieces for helicopter blades?  Why not just attach them right to the motor?

[Guest]

You don’t, there’s a bit in the tutorial about it (Rotor Design.)

[jimmywa]

I have made stock rotors work in a standard helicopter configuration. Using reaction wheels for control and tail rotor blade angle bound to yaw and turning my joystick 90 degrees to counter the gyroscopic effect. collective is by blade angle. Main and tail rotors are both constant rpm. Correctly trimmed, it will happily fly strait and level with sas off and hands off and can be precisely landed. Range about 100km.  It can pick up a 400u fuel drum and place it for local ops.

[Guest]

Nice work. Do you have the craft file up on KerbalX for sharing?

[sperou]

are the builds shown on the images on steam

[Klapaucius]

3 hours ago, sperou said:

are the builds shown on the images on steam

Check out his page on KerbalX

 

https://kerbalx.com/Brikoleur

 

[surge]

As many have said, this is an incredibly good guide for helicopters. Let's say intermediate level

 

A few points:

1) Flex - real world (non-coaxial) helicopter rotor systems have ALOT of flex in the axial-horizontal direction. See an inert heavy lift soviet helicopter like the Mi-6. The need for this is partly because the rigid underslung frame tends to shake apart, but mostly for control. As history goes, the swashplate was invented many years before someone (Igor Sikorsky?) thought to allow the rotor blades to flex and forgive the rigid control system, thus making them inherently stable - and therefore flyable.

I notice in the photo of your rotor hub, you (by experiment or accident) have multiple "cuboid struts" before the rotor blades are attached. These are very floppy parts, and I suspect a flyable single rotor design using that wouldn't be hard to make.

 

2) SAS. Wouldn't really be a point if 1) above was taken into consideration. KSP does not make this easy, but no helicopter has ever had any form of "SAS" and the latest versions theoretically allow this.

If we put 1 and 2 together, a "standard" single rotor design should be possible and flyable. I've been messing around with it myself, (limited) time allowing. Reading your post, I was hoping you'd fully solved the problem but alas.

With the addition of the KAS-1000 controller it should be possible to modify the tail torque to match the main rotor torque such that the craft presents a linear and human-flyable input system just like a real helicopter.

 

3) Updates: The first few versions of breaking ground did not control the surfaces sensibly. They still treated the spinning blade as a fixed wing and pushing cyclic forward would increase the flap angle on *both* sides of the rotor.

For these versions is was impossible to make helicopters other than the original purely SAS controlled ones.

I implore you to look up the exact version and mention this, I think it's 1.11.0, but I may be wrong.

 

 

Edited April 3, 2023 by surge

[surge]

Addenum, types of rotor hubs:

 

Teeter-totter (fully-rigid):

Boeing Iroquis variants (UH-1, AH-1, Jetranger).  2 blades rigidly connected together with a small stabilising crossbar. The blades are free to move up and down like a see-saw as they rotate (thus the name), the crossbar absorbs the remaining vibration.

 

Semi-articulated:

OH-6 Cayuse? Possibly the american UH-60s still use this design.  This rotor head has independent single axis hinges on each blade allowing for more stability in windy and unstable  environments.

 

Fully-articulated:

Nearly all modern helicopters, cheap plastic drones from toy stores. Has 2 way hinges on each blade (essentially a ball joint, but ball joints are weak); the original vertical axis to assist with unstable environments, plus a rotational axis forward and backward "flex" allowing for high speeds. The retreating blade is allowed to "fling" itself forward some 3-10° preventing blade stall, while the advancing blade is damped by drag from moving forward & occasionally preventing supersonic flight.

 

You're welcome

 

Edited April 3, 2023 by surge