What type of motor do I want to control a boat's rudder, a servo?
How do I set it up so that it does not swing beyond 35 degrees?
Which motor would be the most energy efficient assuming we have to adjust it from 60 times a minute to a few times per day depending on traffic? It will run on battery recharged by solar panels.
Since this will be the only way of controlling the craft, how do I know the motor isn't drifting and taking the craft off course? Is there a way to calibrate it remotely?
Please bear with me, I'm completely new to electronics.
I read this would do the job well but soldering that is out my league - brushed-dc-servo-drive
Servo makers usually quote a figure of "kg" which is actually kg.cm. The biggest I just saw on hobbyking after a quick look was 40kg ie 40 kg.cm which is 2 orders of magnitude different from yours.
Servo makers usually quote a figure of "kg" which is actually kg.cm. The biggest I just saw on hobbyking after a quick look was 40kg ie 40 kg.cm which is 2 orders of magnitude different from yours.
Sadly yes, I too found some but nothing close to even 600Kgcm - http://goo.gl/zusg9x
If this was in a lake we'd get by with a lot less, but ocean currents need more torque to navigate.
Would you only use a stepper motor for this application?
I don't. I manually steer, and I assure you that I am exerting nowhere near 6 kg-m of torque on the rudder. A very light touch on a 1 meter tiller bar is all it needs.
I have difficulty believing that your estimate is correct, especially for something moving at 1-3 knots.
Use Nm for torque and avoid all confusion. 6 kgf-m (kilogram-force metres) is about 60 Nm, which
sounds completely wrong.
Very rough rules of thumb:
0.1 Nm - finger and thumb twisting
1 Nm - wrist (screwdriver) twisting
10 Nm - steering wheel on a car
100 Nm - very tiring to control, will injure if not anticipated.
1000 Nm - snap you in half probably.
jremington:
I don't. I manually steer, and I assure you that I am exerting nowhere near 6 kg-m of torque on the rudder. A very light touch on a 1 meter tiller bar is all it needs.
I have difficulty believing that your estimate is correct, especially for something moving at 1-3 knots.
Thanks, obviously a 1m lever helps and I'm unsure if your "feel" of the rudder translates correctly to Kgm. Thank you for sharing your real life insight. I will reach out to the guy who did the torque calculation.
Looking at MarkT's rough chart what would be a good starting point you think?
Do you suppose 52 Kgcm / 5 Nm would do for our first prototype?
I used their Ultraflex method and a possibly oversized rudder of dimensions 0.6096m by 0.3048m and we get a rudder torque of T = 0.582461034817536 Kgm and it says to multiply this with 0.5 for sailboats which I guess we're sorta close to.
In any case, using the formula above the rudder torque should be between 2.9 Nm to 5.7 Nm which translates to 29.12 Kgcm - 58.25 Kgcm and motors for these loads are readily available.
For a rudder, does a stepper motor have advantages over an encoder DC servo?
How would I know if the motor loses track (unsure what this is called) over time/use and steers the boat slightly off course? Kinda like how a scope of a rifle may need adjustment to 'zero' it over time.
How can I ensure it does not swing the rudder more then 20-35 degrees?
jremington:
Significant disadvantages -- not recommended at all.
You should build a prototype for the basic craft before making all these design decisions. Most likely, everything will change when reality strikes.
Sage advice, we'll go with a cheap DC geared motor to start with - http://goo.gl/ezVLYY
I really appreciate all the great advice, please keep it coming
So 6Nm then? Gear motor with absolute encoder, or ABZ incremental encoder with the Z pulse hand-centred
on dead-straight? Sounds plausible. What speed of movement do you need (that will determine motor
power).
You've have to close the control loop with a PID and for the incremental encoder have a start-up routine to
swing the rudder gently to find the centre point (so you'd need an inner torque control loop and outer
position control loop?) Sample as fast as possible, many kHz, not once per second, you have to control torque
in the first instance. - basically implement a servo motor yourself. You'll need an output H-bridge with
current sensing.
You main technical challenge after learning how to build a servo control loop for a motor will be water-proofing
the gear motor and encoder - neither will tolerate moisture and live long, and if its salt water you require
a completely hermetic seal - the marine environment is extremely harsh to electronics, bearings and motors,
no salt water must get in whatsoever.
MarkT:
So 6Nm then? Gear motor with absolute encoder, or ABZ incremental encoder with the Z pulse hand-centred
on dead-straight? Sounds plausible. What speed of movement do you need (that will determine motor
power).
You've have to close the control loop with a PID and for the incremental encoder have a start-up routine to
swing the rudder gently to find the centre point (so you'd need an inner torque control loop and outer
position control loop?) Sample as fast as possible, many kHz, not once per second, you have to control torque
in the first instance. - basically implement a servo motor yourself. You'll need an output H-bridge with
current sensing.
You main technical challenge after learning how to build a servo control loop for a motor will be water-proofing
the gear motor and encoder - neither will tolerate moisture and live long, and if its salt water you require
a completely hermetic seal - the marine environment is extremely harsh to electronics, bearings and motors,
no salt water must get in whatsoever.
To be honest, I have no idea what you meant besides use an encoder DC servo and program it.
Where can I start reading on this?
I used their Ultraflex method and a possibly oversized rudder of dimensions 0.6096m by 0.3048m and we get a rudder torque of T = 0.582461034817536 Kgm and it says to multiply this with 0.5 for sailboats which I guess we're sorta close to.
You might also want to read up on how to use numbers in engineering.
