Thread Tools
Oct 18, 2019, 03:33 PM
Entropy is happening!
Jim.Thompson's Avatar
Thread OP
Discussion

Inflight CG adjuster feedback system.


I am planning on making an inflight CG adjuster for a big plank glider that I have built.
The concept is to use a weight on a threaded rod which will be rotated by a modified servo for full continuous rotation. I have already modified a servo, I did it some years ago, but never finished the project.
The servo rotates either direction when the signal is above or below mid-range.

Question:


Is there a way I can set up a position feedback system that will provide me with indication in OpenTx/Taranis X9D?
Perhaps a multi turn potentiometer could be employed (some how).
Without some kind of feedback system, there is a danger of just stalling the servo.
The simplest way to avoid this would be to have a microswitch at each end to open the circuit. However, then the device could not be driven back in the other direction.
Ideas and suggestions will be most welcome.

Jim.
Sign up now
to remove ads between posts
Oct 18, 2019, 04:27 PM
AndyKunz's Avatar
Your microswitch can be used to open a contact on one side of the feedback "pot" connections. That will tell it that it's reached the limit (and it might back up a little too).

I would go sail winch style myself and not worry about feedback so much. Measure it when you land.

Andy
Oct 18, 2019, 09:44 PM
Entropy is happening!
Jim.Thompson's Avatar
Thread OP
Quote:
Originally Posted by AndyKunz
................................... That will tell it that it's reached the limit (and it might back up a little too)...........
How will it "tell me" Andy? Just because it has stopped driving?

Quote:
I would go sail winch style myself and not worry about feedback so much. .................
Andy
How do they differ in operation compared to a servo that has been converted to continuous operation?
Thanks for the reply.

Jim.
Oct 19, 2019, 01:09 AM
Registered User
Modding a servo for continous operation is calling for trouble.
It acts as an ESC, but needs feedback, and this feedback is mechanically fixed, which will make it creep on minimal changes in temperature or voltage.
It's ok for LEDs if you're cheap and can live with flickering lights (or have a two position channel at hand), but not for anything that can kill people.

The way to go is either with a real ESC and the switches on the end, but each switch shorts a diode when not pushed.
Both switches are in series and in series to the motor.
This way when reaching either end the switch will open, allowing the diode to work, which will cause the motor to be polarized so it can only retract from the switch not push it further.

Or just make it a real servo, there are several ways to do that.
- have a multiturn pot on the spindle (maybe even with gears)
- use a slider pot on the weight
- use a winch servo that goes like 3 turns and not spindle but pulley the weight
- build something with an arduino that drives the weight with a stepper
and i bet there is more.

Each way can be used either as a standard servo approach just like any control surface servo, or can be adjusted to give you telemetry.
Oct 19, 2019, 03:58 PM
Registered User
GeoffS's Avatar
I'd also vote for a stepper and arduino.

If you're really worried about stalling the stepper, there are some drivers out there that can detect a stall amd report out to the controller.

https://www.filastruder.com/products...r-motor-driver
Oct 19, 2019, 10:32 PM
Scott
Pylonracr's Avatar
Jim, why not use a sail winch servo and a cable to pull the weight along a shaft in pull-pull. Kind of like a window regulator on a car. Wind the cable around the servo drum with the weight centered at servo neutral. A couple revolutions each way should move the weight a fair distance depending on the diameter of the drum. On a slider channel the center of the slider will always be the center of the weight travel.

Scott
Oct 19, 2019, 11:39 PM
Entropy is happening!
Jim.Thompson's Avatar
Thread OP

Sail winch servo.


Quote:
Originally Posted by Pylonracr
Jim, why not use a sail winch servo and a cable ......................
Scott
I am not familiar with sail winch servos, but I sort of get the idea. They would naturally reach travel limits like an ordinary servo. Correct?
It is beginning to sound like that would be the simplest idea.
Thanks Scott.
Oct 20, 2019, 02:10 AM
https://hmvc.eu/
this is something I want to do for a long time and I have already bought a winch servo for that. Because I also need it on a very light aircraft it is not yet implemented.

I have a rocket glider (Me-163) and its rocket is at the rear. After the fuel is burned the CoG is moved forward and I wish to be able to compensate that.

some trolly using that and the servo winch itself as the ballast weight was my idea of getting it done.

But the additional weight itself is too much. It then becomes easier to only increase the weight of the whole plane. That way the weight movement caused by burning the fuel becomes relatively less and stops becoming a problem.

Now the plane is extremely tail heavy at the start and nose heavy during its flight. I have put too much effort into making it as light as possible. it is not even painted

I think you want it in a relatively big glider, so it would work for you (winch servo with pull pull). Why would the servo stall if you correctly calibrated it?

Look forward to see pictures of what you end up with
Last edited by hisroyaldudeness; Oct 20, 2019 at 02:17 AM.
Oct 20, 2019, 05:03 AM
Oxford Panic
AndyOne's Avatar
Jim,

If I was doing this I would mod a servo for continuous rotation and remove the feedback pot, assuming the mechanics allowed for it. Then using a linear pot connected to the weight this could be wired into the pot connections on the servo PCB to allow full analogue position sensing QED.

A.
Oct 20, 2019, 05:23 AM
Entropy is happening!
Jim.Thompson's Avatar
Thread OP
Quote:
Originally Posted by hisroyaldudeness
.................................
I think you want it in a relatively big glider, so it would work for you (winch servo with pull pull)...................
Yes.
I want to use this in a 4 mtr span plank. Very low wing loading............so a little extra weight is not a penalty.
It is because it is a plank that I figure it will be very useful for the first few flights. Might save the model from damage during the trimming/sorting out flights.
Quote:
Why would the servo stall if you correctly calibrated it?
It will not stall if a proper sail winch servo is used, as has been already suggested above, and it is calibrated as you mention
That concern is now more or less redundant.

Quote:
Originally Posted by AndyOne
Jim,
If I was doing this I would mod a servo for continuous rotation and remove the feedback pot, assuming the mechanics allowed for it. Then using a linear pot connected to the weight this could be wired into the pot connections on the servo PCB to allow full analogue position sensing QED.
A.
I am not at all sure what you mean here Andy. Thanks for the reply.
However, at this stage, the far simpler application using a sail winch servo, like Scott suggested above, is what I favour.

Jim.
Oct 20, 2019, 10:13 AM
Scott
Pylonracr's Avatar
Jim, a little research is in order. I have never used a sail winch servo but I have seen them. It is my understanding that they operate like a standard servo, except they travel 1.5 to 3 revolutions from center instead of 90 degrees depending on the servo. A few winds of cable in both directions on the drum so you don't run out in either direction should work. Maybe drill a hole in your weight and run it along a carbon fiber rod or dowel. I would think that the Kevlar string we use for rudder pull pull setups should work.

Keep us posted.

Scott
Oct 20, 2019, 01:42 PM
Entropy is happening!
Jim.Thompson's Avatar
Thread OP
Quote:
Originally Posted by Pylonracr
Jim, a little research is in order. I have never used a sail winch servo but I have seen them. It is my understanding that they operate like a standard servo, except they travel 1.5 to 3 revolutions from center instead of 90 degrees depending on the servo. ...........
That is my understanding too Scott. I will ask over on the sailboat forums, just to confirm.

Quote:
A few winds of cable in both directions on the drum so you don't run out in either direction should work. Maybe drill a hole in your weight and run it along a carbon fiber rod or dowel. I would think that the Kevlar string we use for rudder pull pull setups should work............
Yes.
I have some 14 kg breaking strain Dyneema fishing braid, that will do the job.

Jim.
Oct 20, 2019, 02:18 PM
Dave the Rave
dmccormick001's Avatar
Another approach to this would be to keep the threaded rod idea, and use a small brushed motor to turn it. You could use an H-Bridge to drive the motor, that would give you bi-directional rotation, and a PIC to read the input from the receiver and know which way you want the motor to turn. And a simple current-sensing resistor to tell the program when the rod has reached the end of the run in either direction. Basically it would be a brushed-motor (electric) retract, but instead of moving a gear mechanism it would just slide a weight back and forth. No cables, no servos, easy to rig up.
Oct 20, 2019, 10:17 PM
Scott
Pylonracr's Avatar
Jim,
I will give you my address, you can send me my consultation fee..

Scott
Oct 21, 2019, 07:38 AM
AndyKunz's Avatar
Quote:
Originally Posted by Jim.Thompson
How will it "tell me" Andy? Just because it has stopped driving?
Yes. But it will make an oscillating noise, too.

By opening one end of the of the feedback pot, the signal to the motor controller now looks like the pot has exceeded its limit. The controller will reverse, causing the switch to close. That will cause it to go in the commanded direction again, opening the switch.

If you release the stick, it will stop, because the controller will see the input as "hit the commanded point."

Quote:
How do they differ in operation compared to a servo that has been converted to continuous operation?
I have only had one winch servo. It had a ton of torque, and made 10 rotations. I forget if it was 10 each way from center, or 10 from end-to-end.

I don't know if other winch servos are like that. Same idea, though - it will only let you go so far, then stop.

Andy


Quick Reply
Message:

Thread Tools

Similar Threads
Category Thread Thread Starter Forum Replies Last Post
Discussion Does Dx6i save inflight trim adjustments kurtm Radios 5 May 14, 2013 06:20 AM
Discussion tx ant. polarity, inflight adjustment device. JustinONE FPV Talk 4 Nov 26, 2010 01:52 PM
Would an inflight variable CG help in DLG? MattyB Hand Launch 6 Aug 24, 2005 12:57 PM
elevator adjustment or cg adjustment? coolboarder Electric Plane Talk 5 May 30, 2005 07:58 AM