Didn't think I'd be saying this but, you might have to start considering the aerodynamic performance of the print head. That's just a joke, I realize it's still a negligible factor even at much higher speeds. Impressive job controlling the play in the system BTW, I assume that must have been the most challenging part at such speeds.
Fast but low quality. Everyone knows you can increase speed and sacrifice quality. Anyways. The military is looking for people built their AI robot army... you could make some money that way by helping to kill off humanity in the next great, and last, war.
A cette vitesse d'exécution pour mouvement de table rapide quel serait la force linéaire utile et ou disponible sur toutes les directions du plan pour charge maximum "masse maximum de la tête" en newton ou kg force maxi disponible? ....et quel est le coup financier pour notion de budget?...merci pour votre attention et félicitation pour votre réalisation 🤙🦿
Impressed - and thank you for making the code available for others. One question: What kind of sensor do you use? Hall sensors or back emf wouldn't provide enough detail for a servo, or am I wrong?
+Jens Overby You are completely correct, I use optical encoders for position feedback. Specifically these ones: www.ebay.co.uk/itm/Encoder-6mm-Shaft-600P-R-Incremental-Rotary-AB-phase-for-Length-Measurement-/281475951875?hash=item4189462d03:g:qlMAAOSwMmBVyuzu they are mounted on the carbon fiber axle, on the opposite side of each of the motors.
I assume there is a huge torque on those axles which may cause torsion along it. This could lead imprecise positioning. Did you experience such thing? Btw I would build a 3D printer based on this motor system! If only I had the knowedge to do that...
Nice and quiet. But It is possible to see that angled lines not wery straight, it would be a problem for real application. So, what is the highest speed of test setup when it moves axes consistent, drawing straight lines?
Yeah, the cogging torque from the 12n14p style brush-less motors is quite small, and the resolution of the encoder is quite good, so it is able to run smoothly (:
What kind of "string" are you using to replace belts? Really nice job, this is an amazing project and I am looking forward to ordering some when they are ready!
Still see one drawback using BLDC, it's magnetic slip off. In this example and other people examples its reproducibility of same coordinates is poor due to weaker magnetic coupling in comparison to stepper motors. But where very high accuracy is not needed (like normal pick and place machine etc.) I guess it is super nice result. By the way the pcb and software looks kinda promising too. Thumbs up for the author.
Thank you!The controller uses active encoder feedback control to servo the motor very accurately. Therefore the low pole count in the brushless motor is not significant for precision, only the encoder resolution is. You can get a 10000 count/rev encoder for $35.
"DC" in DC motor reffers to how the motor is commutated, i.e. with brushes. "Servo" in servo motor refers to the fact that you have feedback signal. They are different concepts. You can have a DC Servo motor, or a DC motor without servo feedback, or a brushless motor with servo feedback (which is what ODrive is).
+RFS Sales Account The servo motor consists of this motor: www.hobbyking.com/hobbyking/store/__15199__HobbyKing_Donkey_ST3508_730kv_Brushless_Motor.html and this encoder: www.ebay.co.uk/itm/Encoder-600P-R-Incremental-Rotary-Encoder-AB-phase-encoder-6mm-Shaft-W-coupling-/301553008421?hash=item4635f5bf25:g:J~YAAOSwqu9U-I2m and the controller is this project: github.com/madcowswe/Servo
Ah thanks. So magnetic encoders aren't that precise? Well it makes sense because they are the cheapest and seemingly easiest to use (hence why I'm asking lol).
How does it interface with the computer? can I use Mach3 or 4 Old conventional cnc programs? I see that each driver can handle to motors or servos... do I need an extra interface board to connect two or more drivers together??
We do not have the facilities to interface Mach3 to the ODrive directly via the USB port. If you want to use Mach3 today, you will need an interface board. The USB port currently only interfaces to the Python driver, and could be extended functionality by us or the community.
![[TEST] Odrive Motor Control](http://i.ytimg.com/vi/dntYSxTFm3c/mqdefault.jpg)
Didn't think I'd be saying this but, you might have to start considering the aerodynamic performance of the print head. That's just a joke, I realize it's still a negligible factor even at much higher speeds. Impressive job controlling the play in the system BTW, I assume that must have been the most challenging part at such speeds.
RIP pen
Wow. I almost pooped myself at how fast these are. I can't wait to buy your boards when they're ready.
Missed abit
Fast but low quality. Everyone knows you can increase speed and sacrifice quality. Anyways. The military is looking for people built their AI robot army... you could make some money that way by helping to kill off humanity in the next great, and last, war.
A cette vitesse d'exécution pour mouvement de table rapide quel serait la force linéaire utile et ou disponible sur toutes les directions du plan pour charge maximum "masse maximum de la tête" en newton ou kg force maxi disponible? ....et quel est le coup financier pour notion de budget?...merci pour votre attention et félicitation pour votre réalisation 🤙🦿
dude, try to build a 3d print and see if it a customised high speed hotend would handle those fast movement witout making a mess.
Impressed - and thank you for making the code available for others.
One question: What kind of sensor do you use? Hall sensors or back emf wouldn't provide enough detail for a servo, or am I wrong?
+Jens Overby You are completely correct, I use optical encoders for position feedback. Specifically these ones: www.ebay.co.uk/itm/Encoder-6mm-Shaft-600P-R-Incremental-Rotary-AB-phase-for-Length-Measurement-/281475951875?hash=item4189462d03:g:qlMAAOSwMmBVyuzu
they are mounted on the carbon fiber axle, on the opposite side of each of the motors.
I assume there is a huge torque on those axles which may cause torsion along it. This could lead imprecise positioning. Did you experience such thing?
Btw I would build a 3D printer based on this motor system! If only I had the knowedge to do that...
@@Sekir80 if you want something and don't have the knowledge, find it, if someone else can do it so do you.
Wooowww so much POWEEEEEERRR ohh yeaahh
Nice and quiet. But It is possible to see that angled lines not wery straight, it would be a problem for real application. So, what is the highest speed of test setup when it moves axes consistent, drawing straight lines?
Awesome!
You should have carried a mic, and dropped it at the end of the video.
impressive!
Holy crap - thats is fast
Wow, so fast & quiet. Nice.
stick a hot end onto it plz lol
Holy SMOKES!!!!!! Don't get in the way of that thing!
This is what we need in 3d printing!
well, if you look at how wobbly those lines are :/
wow
it's great!!!
Hello from Belarus!
jesus awesome! HOW THE HELL is it tellign the position! did you put a rotary encoder somewhere? also you're using strings instead of pulleys cool!
Which Servo motor used?
It will rotate both directions?if we we give +/-10 or 24V
Jaw droppingly fast! gj
Nice job
How does it cope with slow speed. Does it run smooth without cogging ?
Yeah, the cogging torque from the 12n14p style brush-less motors is quite small, and the resolution of the encoder is quite good, so it is able to run smoothly (:
of course they are fast, the video is sped up. what is the actual travel speed?
Its amazing
Thank you, very nice project, great work. Will follow.
What kind of "string" are you using to replace belts? Really nice job, this is an amazing project and I am looking forward to ordering some when they are ready!
It is "spectra line". Thanks!
can you replace the wire with trapezuim threaded rods?
Yes, using lead screws is also possible.
Ther you have the fucking star. hahahhaa
Looks shitty, thanks, good bye.
holy crap 3d printing with these would be amazing!
You'd get spaghetti all over the floor and the walls!
yeah, this with 3D printing xD
Which open source license?
Very impresive! What are those servos? Link you post 2 years ago is broken... Can i contact you? I have more questions!
You can find out more here including contact details: odriverobotics.com
well done!
Still see one drawback using BLDC, it's magnetic slip off. In this example and other people examples its reproducibility of same coordinates is poor due to weaker magnetic coupling in comparison to stepper motors. But where very high accuracy is not needed (like normal pick and place machine etc.) I guess it is super nice result. By the way the pcb and software looks kinda promising too. Thumbs up for the author.
Thank you!The controller uses active encoder feedback control to servo the motor very accurately. Therefore the low pole count in the brushless motor is not significant for precision, only the encoder resolution is. You can get a 10000 count/rev encoder for $35.
Resolution is not the same as accuracy though.
Hi i wanted to make PID controller for actuator, which motor is best?
Dc motor or servo motor
"DC" in DC motor reffers to how the motor is commutated, i.e. with brushes. "Servo" in servo motor refers to the fact that you have feedback signal.
They are different concepts. You can have a DC Servo motor, or a DC motor without servo feedback, or a brushless motor with servo feedback (which is what ODrive is).
What servos did you use?
+RFS Sales Account
The servo motor consists of this motor: www.hobbyking.com/hobbyking/store/__15199__HobbyKing_Donkey_ST3508_730kv_Brushless_Motor.html
and this encoder: www.ebay.co.uk/itm/Encoder-600P-R-Incremental-Rotary-Encoder-AB-phase-encoder-6mm-Shaft-W-coupling-/301553008421?hash=item4635f5bf25:g:J~YAAOSwqu9U-I2m
and the controller is this project: github.com/madcowswe/Servo
+madcowswe Great! Thanks for the quick response.
Boom
Sonny.
Amazing
awesome
Wow!
WOW
Sick!
Wow.
Very nice work! Do you get much overshoot?
No, not with correct tuning (:
That's awesome :)
I've checked your website, why no magnetic hall effect sensors?
We will add that at some point, check here: github.com/madcowswe/ODrive/issues/70
Ah thanks. So magnetic encoders aren't that precise? Well it makes sense because they are the cheapest and seemingly easiest to use (hence why I'm asking lol).
Magnetic encoders and hall effect sensors are not the same thing.
Great video! What kind of controller board did you use to interface the odrive?
The motion was generated on the ODrive.
@@ODriveRobotics No, I mean in terms of microcontrollers
@@eltonrodrigues5885 the only microcontroller used is the one on the ODrive
@@ODriveRobotics Okay Thanks!
Fucking hell, that... I need that
What do you need it for? ;D
EVERYTHING!
I kinda wanna make a 3d printer like that, but Id probably need a nuclear extruder to keep up...
How does it interface with the computer? can I use Mach3 or 4 Old conventional cnc programs? I see that each driver can handle to motors or servos... do I need an extra interface board to connect two or more drivers together??
You can use step/direction signals with the ODrive. You don't need an extra board to use multiple ODrives together.
And what kind of programs I can use?? I have a few machines already using Mach3 .... ??
Anything that outputs step/dir is fine. For example Mach3 with the parallel port output.
So that means that I will have no other option but to use the Serial Parallel port "25 Pin"? I was hoping to be able to use a USB Port??
We do not have the facilities to interface Mach3 to the ODrive directly via the USB port. If you want to use Mach3 today, you will need an interface board.
The USB port currently only interfaces to the Python driver, and could be extended functionality by us or the community.
ROFL I started laughing so hard when it suddenly started to freak out!
Try clearpath servos all integrated cheaper and faster overall
What advantages does that controller has over the vescs?
It's more geared towards robotics.
ODrive Robotics can still use the vesc...