Clearpath Servo Motor

General Forum Stuff...Introduce yourself

Clearpath Servo Motor

Postby evil2002usna » Mon Jul 20, 2015 4:42 am

I am currently in the process of building and Openbuilds OX CNC. While in the process of selecting the electronics for my build I came across a video on youtube from Neo7CNC about the Clearpath Servo Motors.https://youtu.be/Z2kAcz-rG5s It sounds like he is only using a C10 Breakout Board combined with these servos. I was wondering if anyone here as any experience with these particularly the wire connections. I have already bought the breakout board. I will probably end up buying at least one initially to try it out on either the Y or Z axis. Here is a link to the motors. https://www.teknic.com/products/clearpath-brushless-dc-servo-motors/
evil2002usna
 
Posts: 49
Joined: Tue Sep 10, 2013 1:53 am

Re: Clearpath Servo Motor

Postby evil2002usna » Mon Jul 27, 2015 4:04 am

Well it has been a couple of days and no replies yet, maybe I can wet your guys appetite with a build picture for my new CNC. On top is the Ox CNC. I am in the process of building a mobile base for the CNC Router to be housed on top of. I still need to build the shelves, the doors, and the tabletop for the cabinet and then stain it, but I think it is looking good so far. I will post some more pictures as the project progresses. I still need to get the motors, cables, and power supplies and then integrate a computer into the base somehow. This will undoubtedly take me longer to build than my Ord Bot did.
Attachments
IMG_20150726_223210366.jpg
evil2002usna
 
Posts: 49
Joined: Tue Sep 10, 2013 1:53 am

Re: Clearpath Servo Motor

Postby rsm5178 » Mon Jul 27, 2015 4:47 pm

I've been hoping someone would comment on this...

To me, the motors seem really neat. Kind of hope that you use them so we can hear/see how they perform. Super quiet, repeatable, and strong.

On the other hand, what do you gain from these motors over the steppers on the final machine? Is the OX stiff enough to benefit from them? I haven't built my OX yet, but I can only assume the mechanical limitations of the machine itself would outweigh the benefits of these motors. Then there is the high cost too.

I'm running the G540 with Nema 23's and I think they are fine for the performance I need in my home built machine.

-Rich
rsm5178
 
Posts: 119
Joined: Tue Jul 23, 2013 9:57 pm
Location: 19425

Re: Clearpath Servo Motor

Postby terjeio » Tue Jul 28, 2015 6:32 am

My machine is based on a Neo7CNC design - some of the design choices he has made is, IMHO, not well tought out. If you view his videos you will find he has modified his main mill several times... I think servo motors will only benefit a belt driven design as long as the encoders are put directly on the axes, this then to compensate for any inherent backlash. If the encoders are placed on the motors then I believe servos are of no use over steppers unless the design is based on ballscrews. What is your reasoning behind the use of servos? Faster rapids? Better accurancy? What will the mill be used for?

Disclaimer: I am no expert, my opinion is based on what I have found out when searching for information prior to building my mill.
terjeio
 
Posts: 24
Joined: Mon Oct 27, 2014 11:22 am

Re: Clearpath Servo Motor

Postby r691175002 » Tue Jul 28, 2015 2:46 pm

Thanks for bringing them up, they are actually really neat. The marketing is pretty much bang on: they hit an amazing price/performance ratio that actually beats most Chinese systems in the Nema 23+ size range. I have been planning a build that uses leadshine easy-servos but will probably switch to the clearpath motors instead.

Unfortunately I doubt many hobbyist builds can make use of their increased performance. Unless your build is rigid enough to hold <0.005" under full load I wouldn't recommend a servo.

There are a huge number of differences between stepper motors and servo motors, but the most important ones are difficult to explain. Here is a quick overview:

  • Surprisingly, stepper/servo motors are electrically identical and completely interchangable - there are only so many ways you can wind a brushless motor. Stepper motors often have higher pole counts, but the real difference is purely the presence of an encoder+control system.

  • Servo motors often achieve a higher maximum speed but lower torque.

  • Servo motors can achieve enormously higher resolution. Realistically you cannot expect accurate motion below half-step resolution out of a stepper - there is just too much sticktion. Servo motors can get to within a few counts even at 10,000+ PPR.

  • The biggest difference is rigidity. Magnets are really springy and if you put a large lever on a stepper you will see just how much play exists at the step level. High frequency control systems on a servo can be exceptionally rigid (similar to direct metal on metal contact).

  • This difference is also reflected in motion accuracy/reasonance. When you "step" a stepper motor you are only rotating the magnetic field - the shaft itself will lag and can over/undershoot. A servo can move exactly as commanded, especially in control systems that integrate lookahead.

  • For an example of how messy stepper motors can be look at this chart of position over time:
    Image

In simplest terms, in a stepper motor you are only controlling the magnetic field (via winding current) and the shaft itself will be generally pulled in that direction. In a servo you are directly controlling the motor shaft.

The deeper you dig the more advantages you find for servos over steppers, but the truth is that in most machines the motors contribute a tiny fraction of the total error. Unless you are right on the bleeding edge, money is better invested in the frame, spindle, linear motion, etc..
r691175002
 
Posts: 242
Joined: Mon Dec 13, 2010 5:05 pm

Re: Clearpath Servo Motor

Postby terjeio » Tue Jul 28, 2015 4:00 pm

@r691175002 As I understand it servo motors are linear motors and thus have no inherent resolution per se, resolution will be a function of the encoder and the servo loop. A belt driven system, which will inevitably have some backlash, will not gain any precision with a servo system unless the encoders are placed on the axes directly - not so if the encoders are mounted on the motor spindles. I would therefore think that servo motors on an Ox (with spindle mounted encoders) are a waste when it comes to enhancing resolution, if speed is what is desired the I can understand. Problem is that increasing the speed on a belt driven system will reduce resolution no matter what... Servo motors (or rather the servo loop) must also be tuned to function properly - not easy I understand. So maybe better to stick to steppers (IMHO), of course depending of the intended use of the mill. Previous disclaimer still apply ;)
terjeio
 
Posts: 24
Joined: Mon Oct 27, 2014 11:22 am

Re: Clearpath Servo Motor

Postby r691175002 » Tue Jul 28, 2015 7:36 pm

As I understand it servo motors are linear motors and thus have no inherent resolution per se, resolution will be a function of the encoder and the servo loop.

This is correct. The resolution of a servo is the resolution of its encoder. A stepper motor basically maxes out at 400/800 half-steps per revolution whereas encoders are cheaply availaible up to 10,000ppr (with virtually no upper bound on resolution if you have the $).

A belt driven system, which will inevitably have some backlash, will not gain any precision with a servo system.

In applications like 3d printing and laser cutting where there are very low cutting forces sometimes the additional resolution of a servo will help. The correct timing belt profiles don't really have backlash, just stretch and poor rigidity.

[...] unless the encoders are placed on the axes directly [...]

Moving the encoder off the spindle shaft is an advanced topic. Generally encoders on the load are only useful when compensating for very "slow" errors such as thermal growth and screw pitch. Chatter and cutting forces are too fast and unpredictable. If your machine is not rigid an expensive servo won't help.

Problem is that increasing the speed on a belt driven system will reduce resolution no matter what.

This is only true if you are trading speed for resolution mechanically (for example by using larger pulleys). A servo will have higher resolution and maximum speed when compared to a stepper.

Servo motors (or rather the servo loop) must also be tuned to function properly - not easy I understand.

Modern servos are generally pretty easy to setup. Most have software so you just plug them into a computer and click autotune.

So maybe better to stick to steppers (IMHO), of course depending of the intended use of the mill. Previous disclaimer still apply

This is almost always true. As a general rule I would never recommend servos unless you have linear block and rail.
r691175002
 
Posts: 242
Joined: Mon Dec 13, 2010 5:05 pm

Re: Clearpath Servo Motor

Postby macona » Tue Aug 04, 2015 5:44 am

r691175002 wrote:Thanks for bringing them up, they are actually really neat. The marketing is pretty much bang on: they hit an amazing price/performance ratio that actually beats most Chinese systems in the Nema 23+ size range. I have been planning a build that uses leadshine easy-servos but will probably switch to the clearpath motors instead.

Unfortunately I doubt many hobbyist builds can make use of their increased performance. Unless your build is rigid enough to hold <0.005" under full load I wouldn't recommend a servo.

There are a huge number of differences between stepper motors and servo motors, but the most important ones are difficult to explain. Here is a quick overview:

[list][*]Surprisingly, stepper/servo motors are electrically identical and completely interchangable - there are only so many ways you can wind a brushless motor. Stepper motors often have higher pole counts, but the real difference is purely the presence of an encoder+control system.


No, servos and steppers are not electrically identical. They are two (bipolar), four(unipolar), or five phase motors. The rotors and stators are constructed differently as well, the serrated rotor and stators that direct the magnetic field to do the stepping. Feedback is what makes a servo a servo. That can one or more of a multitude systems. encoders, resolvers, hall effect, potentiometers, etc.

r691175002 wrote:[*]Servo motors often achieve a higher maximum speed but lower torque.


Servos have more running torque and speed, especially at high speeds. The torque rating you see listed for a stepper is not the same kind of torque for a servo, it is the step torque, the amount necessary to move the motor one step while the coils are energized at the rated parameters of the motors. Steppers are great for move and hold type situations.

r691175002 wrote:[*]Servo motors can achieve enormously higher resolution. Realistically you cannot expect accurate motion below half-step resolution out of a stepper - there is just too much sticktion. Servo motors can get to within a few counts even at 10,000+ PPR.


Yes, servos, especially modern ones, have insanely high encoder resolutions. The mitsubishi ones I am using on my telescope one come standard with 131072 line encoders and these are two generations old. I believe the current gen, the J4, has 2 million line encoders.

r691175002 wrote:[*]The biggest difference is rigidity. Magnets are really springy and if you put a large lever on a stepper you will see just how much play exists at the step level. High frequency control systems on a servo can be exceptionally rigid (similar to direct metal on metal contact).


r691175002 wrote:[*]This difference is also reflected in motion accuracy/reasonance. When you "step" a stepper motor you are only rotating the magnetic field - the shaft itself will lag and can over/undershoot. A servo can move exactly as commanded, especially in control systems that integrate lookahead.


When you advance a stepper you are really not rotating the field like you would with a three phase stepper. They are not wound like that, the cold cause a shift in the field which causes the teeth to index to the next spot, it is similar, but not the same. When you start micro stepping things change a bit and you are running them closer to a brushless motor and thats why you start loosing accuracy in between full steps.

r691175002 wrote:The deeper you dig the more advantages you find for servos over steppers, but the truth is that in most machines the motors contribute a tiny fraction of the total error. Unless you are right on the bleeding edge, money is better invested in the frame, spindle, linear motion, etc..


Servos are awesome, I am not sure about the chinese ones. Tecnik I dont know much about. In the past they have been very customer unfriendly.

I prefer to buy used Mitsubishi and Yaskawa motors and drives off ebay. The newer drives have incredibly good auto tuning and harmonic elimination that makes them real easy to use. Tuning is the bane of servo motors. Good drives have it built in, cheap ones you have to set manually and it can take quite a bit of time on a new machine, there are so many variables to get smooth servo operation. I had a heck of a time getting my laser cutter tuned, I chose not to use mitsubishi servos on that.

The teknic servos do look neat and the price is not terrible, the resolution is pretty awful though.
macona
 
Posts: 363
Joined: Tue Aug 17, 2010 10:30 pm


Return to General Discussion

Who is online

Users browsing this forum: No registered users and 10 guests

cron