No throttle sent to ESC on channel 8

OK. So, here’s the deal. I downloaded the manual for that heli and it’s supposed to be leading edge control. You assembled the head wrong. The blade grips have to be rotated 180 degrees.

Let’s go over this, and this will be kinda long :grinning:
First and foremost, in today’s FBL systems where the delta angle is zero, leading and trailing edge control are exactly the same. In other words, there are no advantages or disadvantages.

When you have a rotor head system that has a non-zero delta angle, there are clear pros and cons. To start off, what’s delta angle?

Look down on your heli’s head with the grips at zero pitch. Draw a line between the two grip balls on a two-blade head and a line perpendicular to the blades and through the center of the head (like a virtual flybar). The angle between these two lines is a measure of the delta or delta-3 angle.

Think about a typical heli head. It’s what we refer to as semi-rigid because the feathering shaft and blade grips can actually teeter in the head block due to the compliance of the damper material. Helicopter rotor systems are designed to flap to compensate for the di-symmetry of lift in forward flight.

Simply put, the advancing (left blade in a CW rotating system) blade sees a higher-speed air flow due to the blade speed plus the heli forward speed while the retreating (right) blade sees a lower-speed air flow.

The result is higher lift on the left side of the rotor disk and a tendency for the heli to pitch up as a result (remember gyroscopic precession says that the resulting action on a rotating body occurs 90 degrees from the point of application of the force).

When the blades are allowed to flap, the advancing (left) blade sees more lift and flaps up, which reduces its relative angle of attack, thus reducing lift. The opposite occurs with the retreating (right) blade and the result is a more balanced lift distribution across the rotor disk.

Ok, back to delta angle…

When the blade flaps, the blade grip moves up or down with it, but the ball is constrained by the link that is rigidly attached to the swashplate. So, where that ball is located with respect to the teetering point or center of rotation of the feathering shaft dictates how the pitch of the blade varies during flapping.

If the ball is located at the teetering point, then the pitch does not vary with flapping. However, if the ball is shorter (i.e. non-zero delta angle), then when the blade flaps the pitch will increase or decrease depending on the flapping direction and whether it’s leading or trailing edge control.

Leading-edge control with non-zero delta angle will cause a decrease (increase) in blade pitch with an upward (downward) flap…this causes a natural restoring force that tends to reduce the lift on the blade. This is referred to as correcting or positive delta.

Trailing-edge control has the opposite effect…a non-zero delta angle will cause an increase (decrease) in blade pitch with an upward (downward) flap…this causes an increase in lift, which can cause an instability and lead to blade fluttering in extreme cases, but in general will only increase the stability of the heli. Of course, this effect can be partially reduced by stiffer dampers, but only to a certain extent. This is referred to as uncorrecting or negative delta.

In general, for quick handling you tend to want to run stiffer dampers to reduce the amount of flapping because this allows for a more direct transfer of disk forces to the heli airframe and results in a more responsive feel. Beginners would prefer more flapping (softer dampers) because it has a stabilizing effect.

Your three-blade head looks like a rigid design - no dampers. But because it’s DFC the phase angle will be WAY off. The blade will be late in the cycle. So don’t change the firmware - take a closer look at your manual and especially where you are building the head. Flip those blade grips over to make it leading edge control and then it will work.

1 Like

You made an honest mistake without realizing it, but not all heli heads are the same. Here’s one with leading edge control, but it has reverse collective - the swash moves down to increase pitch. Because the pitch links are driven by the swash followers. This is a zero-delta head and it can be assembled either leading or trailing edge control, which only changes the direction of the collective control, but not the cyclic control

100_0043

Here’s another head with swash follower drive on the plate. But it uses pitch links direct to the swashplate. This is also a zero-delta head and it can be either leading or trailing edge control. But with trailing edge control, everything is reversed - collective and cyclic. And ArduPilot does not actually care about this. Simply reversing all the servo directions will make it work with trailing edge control.

100_0045

The phasing of the head does not change with either of the above two types of heads because they are follower drive. Your Triabolo heli is DFC drive. So the pitch links both control pitch of the blades and drive the swashplate. Which side of the feathering shaft they are on will affect the head phasing. I believe you are probably about 90 degrees off on the head phasing with the control arms on the trailing edge. Which means forward cyclic (down elevator) will make the heli roll right. The control phasing will be 90 degrees late thru the whole cyclic range.

Which would make it really interesting to fly. I think a human pilot could get the hang of it with a little practice. The autopilot never will :face_with_raised_eyebrow:

@ChrisOlson when you wrote that ArduCopter doesn’t support DFC head with trailing control, did you meant just this configuration or any trailing edge control. I am little confused as I am building a Trex450 Pro clone with flybar head and it has trailing edge control, with the swash movement not reversed.

Flybar heads are trailing edge control because of the mixing arms. They have normal collective direction (swash up increases pitch) because of the mixing arm interaction.

What ArduPilot does not support at present is heads that use reversed collective (swash goes down to increase pitch). This is normally associated with trailing edge control, but it is not tied to it. It is totally dependent on the geometry and linkage in the head.

Copter will support it soon - I just wrote the code for it and working on support for H3-140 swashplates, so haven’t PR’d it to 3.6-dev yet.

https://github.com/ChristopherOlson/ardupilot/commit/ff057bb67d71ff2d3bcba136b4b7dffcc4623393

Thanks Chris for the explanation, now everything is clear :slight_smile:

So I was able to flip my blade grips, and set up everything. All servos go in the right direction and the collective pitch is also correct! Great learning experience at least. :slight_smile:

I’ve tuned the helicopter following http://ardupilot.org/copter/docs/traditional-helicopter-tuning.html for an initial hover flight. Now I just need to wait for the wind to die down.

Thanks for all of the help Chris.

Mike

Awesome! That’s a really cool heli. Just make sure the tail holds good before you start flopping it around to tune roll and pitch. Nothing worse than having a heli shaking like a leaf in the wind when you hit the limit on gains, with no tail control :wink:

Not a big deal with smaller helis. But with big ones I generally like to run 'em up on the ground with an adequate hold down of some sort (usually involving a lot a weight) to make sure everything mechanically works right under load before going airborne the first time. You can have blade tracking issues or similar that can be spotted and fixed before flying it.

A 800-class machine can easily produce 100lbs of thrust, so have to be a bit careful on the collective on a ground run so you don’t become airborne with your holddown. But it will give you a chance to make sure spoolup/spooldown works right, etc… I generally prefer to stay at least 100 feet from a big one on the first run in case something comes apart or it goes out of control for some reason.

Guys I know this is old topic! but I am very new to this and already got lost here with me ESC

I have T-rex 600 ESP with PIXHAWK and arducopter 3.6.9

I been chasing my tail for a month now!

I did everything what Chris in his videos ( thank you Chris for that help)

but the ESC wan’t trying to help me and get it to work!

I have 130A platinum OPTO V4
I calibrated alone with my FRSkY receiver and also used the wireless devise to setup the flight mode (linear throttle)
soft cutoff cutoff voltage 3.0
motor rotation CW , active freewheel ,

I did video to show your guys all my futaba 14sg ch setup and my mission planner

also I show the problem when I try to calibrate the ESC its not want to work!!

at the end of the video my camera stopped I was trying to say Chris says the pitch must be REVERSED but I can’t reversed otherwise all will change and will make a mess!
so now its moving with the direction of the sticks!

please Help ,
thank you
Ozzy

As explained in the wiki for RC setup, the pitch direction must normally be reversed in the RC in order for the slider to go the right way in the radio calibration screen.

http://ardupilot.org/copter/docs/common-radio-control-calibration.html

Any of the servos can be reversed for heli after doing that. You might find it easier to use QGroundControl to set up your helicopter than Mission Planner. QGroundControl has a heli setup page that makes reversing the servo directions, and setting them quite easy.

Thank you Chris for the videos they are very helpful !

I found that I was wrong! in the Pixhawk there is no throttle stick instead in mode one its only switch to arm and start the motor!!

now I am on the final tests \
today I took the bird outside for the first test and first the motor start the swash plate was shaking and I turned it off! I took the blades of and I spin it on the bench and first the motor start the swash start tilting to the right direction !!! and I notice also the both (pitch and aileron servo ) shaking

don’t know what to do next!

I fallow your steps on Qgroundcontrol and yes its easy ti use than the mission planner!

thank you
Ozzy

That sounds like a potential vibration problem upsetting your attitude solution.

Chris, please watch the new video I uploaded !

its look like there is something not quite well going on in the swash plate when I run different modes !

I think if I fly it as is will crash!

I tried to go back to 3.6.9 but when I install it in the pixhawk the radio setup page and the swash not responding, I tried to use the frame type _class as Heli but nothing moving

on the Qgroundcontrol telling me there are so many parameters missing!

I tried to uninstall the both Qgroundcontrola and the mission planner but no hope!

so I am staying now with the oldest 3.6.7 version as you see it in the video !

please help me with this
Ozzy

Your video doesn’t seem to be working, Ozzy.

What you are expecting from the Autopilot on the bench is impossible. Wait for Chris he will explain to you how it really works.
BTW I got the video going,Chris.

I was able to download the new firmware
3.6.9 and after 4 days of fighting he Pc and he Pixhawk

Every time I was installing it then Pixhawk missing a lot of awash plates parameters

But last night I was about to upgraded to 3.6.9
With no problems and all my radio setup is working and parameters in there !

I heard that the Bench test is not going to give the clue about the system and because it need GPS and real flight to operate

But is there a way to test the flight modes and the system before the first test ???

I don’t want to get a heart stroke :slight_smile:

Thank you
Ozzy

Ozzy, the only valid bench test is in unarmed state in Stabilize or Acro. The system expects different behavior in the other modes, and when armed, that can only be duplicated with a live helicopter ready-to-fly. For instance, in Loiter you do not have collective or cyclic control. The autopilot does. And you can do is make requests to the autopilot to move it or make it go up or down.

The only two manual modes are Stabilize or Acro.

Acro has strange behavior in Copter on a bench test. If you want to fly a greatly improved version of Acro you can load the ArduHeli 3.6.7 build from here->

The information on the new acro in the ArduHeli build can be found here and this has been a very, very popular enhancement for helicopters. There are some outdated links in that thread, but the latest version is in ArduHeli 3.6.7. There are several other features in the ArduHeli build as well, that Copter does not have. But you don’t need to worry about any of that stuff for right now, just to get your heli flying.

I have updated to 3.6.9

Is that the newest one ?

It took me 4 days to fix the missing parameters with that version

Thank you
Ozzy

Ozzy,
The 3.6.9 is the latest build for Copter not for TradHeli. You should follow the link Chris gave you for the latest stable Heli build, its version 3.6.7.
Use QGControl to upload it manually to your Heli.

There is a 3.6.9 build for ArduHeli that has the same heli features as the 3.6.7 with the addition of four-servo swashplate support, but it is based on Copter 3.6.9 instead of Copter 3.6.7.

The difference between the two is the IMU “hotfix” that was put into 3.6.9 for Cubes, and the new “watchdog” that reboots your controller in-flight if the main loop locks up. There is both ChibiOS and NuttX builds for both. And, of course, the ArduHeli 3.6.9 build has support for four-servo swashplates, so some of your params may be missing with 3.6.9 due to different handling of swashplates. There is a description in the release notes that explains what the new swashplate settings are, and what they do for the 3.6.9 build as this was not yet documented in the wiki and it was targeted for the Copter 3.7 release.