Wow!
Impressive!
My bad! Had the wrong library, try this one, its super stripped down:
Iām running it on a teensy LC
@dbynoe - Thanks for thatā¦ Iām getting all 10 channels through the I-BUS interface. Works like a charm and actually makes my DMX code easier to writeā¦ Looks like you can also re-assign the FS-i6 controls to the I-BUS channels from the menu of the FS-i6 (though Iāll do it in my software)
1 Like
I re-coded the rcDMXer to use the i-BUS library (so it gets the servo data via a serial port from the RC receiver) instead of using the PPM code and mounted it in a boxā¦
I used a Teensy 3.1 but you can run this code on pretty much any Arduino controller
I added a XLR3 and RJ45 jacks for the DMX outā¦
Affixed the RC receiver and antennas with double face sticky tape ant hot glueā¦
Added a power ledā¦
I now have 10 seperate channels of DMX controlled by the 10 different controls on the FS-i6
4 channels for the right and left joystick X & Y channels (each channel ranges from 0 to 255)
2 channels of potentiometer (each channel ranges from 0 to 255)
3 channels of 2 position switches (each channel is either 0 or 255)
1 channel of 3 position switch ( channel is either 0, 128 or 255)
In the code I can assign any DMX channel to any RC control (hence the panel mounted USB cable). I may want to add a OLED screen and rotary encoder to allow configuring channel assignments without needed to change the code but will probably not ever get around to itā¦
1 Like
Here is my code for rcDMXer
Itās extremely simple and could be cleaned up with for/next loops but it worksā¦
I tried to include all relevant information in the commentsā¦
rcDMXer_1.1.zip (2.1 KB)
This is for I-BUS but I do have other code for PPM if anyone is interestedā¦
2 Likes
If you are wondering about animatronic control and why I am using DMX with Eddies various actuators, here is some decent information:
And here is some general information on animatronic control I have gathered:
And some animatronic projects information:
1 Like
After a bit of bit wrangling I managed to get the dual motor code workingā¦
This allows for two seperate PID instances so each motors response can be tailoredā¦
Also mounted the position sensor for the second motors axisā¦
Now I have 3 of the 5 motors under control:
Next is Eddies head. It has 2 motors for rotation and tilt. Along with 2 servos for the mouth and horns. First I have to rip them all apart, clean and grease them, then put it all back together. Then hook it up to the second motor driver board.
Started working on the head mechanism:
Got it into pieces:
Bearing were gunked up with hardened greaseā¦
Soaked everything in solventā¦
Cleaned out the bearingsā¦
Refilled them with new greaseā¦
Want to check the lubrication in the gearbox but discover the gear on the motor shaft was glued on with epoxy:
This was done on some of the other motor gears but they were large enough to re-tap and put in a proper set screw. Not sure about this oneā¦
To complicate things the gear isnāt mounted square on the shaft so it wobbles quite a bit:
Got to figure out options for pulling the gear off, heat may be the easiest but the gearbox is plastic (though the gears all seem to be metal)
2 Likes
I tried testing the motor in place and even though the gear wobbles I think I will leave it as I suspect trying to straighten the gear will cause more more troubles than it is worthā¦
This larger gear it meshes with is for the head swivel and it will only move about 45 degrees back and forth at the most
2 Likes
for pulling that gear off maybe you just need to shield it from heat? like some ceramic tile or something as a shield
I decided to leave it as it isā¦
I opened up the gearbox and turns out most of the gears are plastic so figured applying heat or too much force would probably wreck the whole thingā¦
Got the next motor cleaned up, re-installed.
Rigged up a position sensor (I love these AS5600 sensors) and got the controller board configuredā¦
This motor rotates the head
I know have 4 of the motors configured and controlled by the PID motor controllers and each is controlled by a seperate DMX channelā¦
Next the head tilt mechanism and motorā¦
That will be the last motor that needs to be controlledā¦
The mouth and horn mechanisms are controlled by servos so I wonāt need to build controllers for themā¦
1 Like
And a note on the āMurder Botā issueā¦ Eddieās robotic mechanisms do pack a wallop and can hurt. Especially if they are suddenly moving to a new position. I have installed an emergency stop switch and have a seperate power switch for the motor power. After a few whacks I have tried to develop some protocols to follow in my compact shop.
I try to do this before energizing the motor power:
- Make sure the DMX controller is on and connected to Eddieās controller
- Make sure the DMX controls are set to known position that matches Eddies existing axis positions
- Make sure nothing is within Eddies reach
Only then do I power up the motorsā¦
I have made the mistake (many times) of powering up the motors with the DMX position data being way off the current axis positions and having everything move at once to go to itās desired position. Have the PID does slow down things but it can still cause pain if you are not prepared for itā¦
I may also put in a time delay to ensure the controllers have time to boot up before the motors can move.
There is also an issue where if the servo controller doesnāt see a DMX signal when it starts up it decides that all channels get a 0 input and this can make the axis suddenly move to the 0 value position. I may rig up a circuit that āinterceptsā the DMX input, monitors it and if there is no valid DMX signal it fakes something. Animatronics typically default to a center value but in my case I may make the pan/rotation axis default to mid position but the up/down axis default to the up position in the assumption this would be the safestā¦
2 Likes
After a bit more thought Iām gonna rig up a controller to control the motor power in the following ways:
- Give some amount of delay from power up before motor power is available. This will make sure all motor controllers have started up before motors can move.
- My DMX servo controller has 8 servo outputs (7 are in use so far) but it also has 8 PWM LED outputs. I plan to use some of the LED outputs for eye and mouth LEDs but wonāt need them all. I will have the controller monitor one of the PWM LED outputs and only enable motor power if the DMX channel is set for 255.
2 Likes
With initial testing of a few motors moving at the same time all was good except I would seem to get a crash of one of the motor controllersā¦ I suspected it was EMI interference from the motors but I had already done pretty much all I could to fix that (next step was possibly using shielded cable for the motor leads). But it only happened when I moved things at max speed and that was never going to happen in normal operationā¦
But with the additional motors it was happening more often. Again only when moving multiple axis at full speed (so not an issue during normal planned operation) but was still really bugging me.
Once I really started digging I realized that the 12V 12.5A power supply was actually dropping itās output to almost zero for just a short time when it got hit with the load of multiple motors moving quicklyā¦ It was so quick (much less than a second) that some of the controllers didnāt even notice (they have larger filter caps).
I tried a few other power supplies and ended up finding a 12V 29A supply that has no such issues. Only problem is that itās bigger than the old one and I have to re-jig all the power cabling and possibly move the controller mounted on the baseā¦
Here are all the 4 controlled motors moving at once (as fast as they can) in the full range of their motion:
3 Likes
For startup safety, maybe copy what RC transmitters do: If, on powerup, any INPUT axis or switch is not in its zero/default position (i.e. throttle zero, rudder, elevator and ailerons centred), refuse to send any OUTPUT signals until the controls are zeroed.
And, to go a step further, have a separate arm switch that canāt be toggled unless the INPUT controls are zeroed, and which, when actuated, implicitly zeroes all the OUTPUTS, so that nothing will automatically move from its current position once armed.
And a dead-person switch on an elastic lanyard is probably a good ideaā¦
1 Like
I have finally got all 5 of the DC motors setup with encoders and controllers:
Here are some pics of the electronics:
12V power supply, 5V power supply (steps the 12V down for the controllers and the servos) and the DMX servo controller (converts the DMX signal to the RC servo PWM):
The RC PWM to DC motor/encoder controller (single motor version) for the base rotation, its motor driver and the relay to control the motor power (right now just hooked up to the Emergency Power Off button):
The RC PWM to DC motor/encoder controllers (two of the dual motor versions) with motor drivers:
And here is the final motor for the head tilt and itās encoder:
The DC motor controllers will hopefully have been the hardest part. Next I have to put the head and mouth back together with its servo driven jaw and same with the horns. The existing mouth servo is toast so I am using a new one. Itās smaller so have to figure out some new mounting.
Also have to properly mount the horns as the mechanism that moves them was just glued to the headā¦
Then I have to figure out how to put all the body parts back together and mebbe a paint jobā¦
2 Likes
Had to do a little dragon dentistry today and rebuild the mechanism for Eddies mouthā¦ The original frame was cracked (it was made of plastic) and the original servo was toastā¦ Made a new frame out of a piece of aluminum (lots of drilling, cutting with jig saw, filing to make smooth, more drilling and tapping) and then installed a new servoā¦
Eddie can talk once againā¦ Or whatever dragons doā¦
1 Like
Made a bit more progress. Got the mouth mounted and connected to the servo controller:
In theory this was gonna be the easy part as the servo connects directly to the servo controller (compared to the previous complexity of having to make the additional controllers for the DC motors). However I did need to limit the range of the servo to avoid it jamming up. The servo controller supports this but itās configuration software was written in 2011. I thought it was having issues in compatibility mode (under Win10) but found the same issue running in under XPā¦ There actually may be a bug in either the controller firmware or the configuration software when you try to set the limits on one of the servo outputs (changes work on all the other outputs it seems).
Anyways next I have to get the head back together and get the horns workingā¦
2 Likes
Fun fact: Vimeo was a spinoff of CollegeHumor. The name was a cheeky reaction to Youtube (āMeā instead of āYouā, plus āVideoā)