Messages

1

Third party software and hardware support forums. / Yet another drive compatibility thread. Maxon 4-Q-EC DES 50/5

« on: July 09, 2015, 03:08:32 AM »

Hi all!
I recently acquired (for free) some lab equipment that makes use of some very nice Maxon BLDC servos and controllers. 5 sets of controllers and motors with which I would like to build a small Mach3/4 CNC mill. I have built a Mach3 controlled milling machine before, but it had stepper motors and Kelling drives which were very easy to set up. After using that machine to make hundreds of the same part, I decided I wanted to build another, but with servos instead. My reason for this is that during that run of parts, I continually fought with lost steps due to resonance. I tried everything possible to avoid those missed steps and my motors were more than suitably sized for the application, trust me! I found that accelerating past the resonant range (as suggested in the wiki) ALWAYS resulted in lost steps and so I was limited to a max rapid feedrate of <30ipm. But that is not what this post is about.... just know that I refuse to ever use a non-closed-loop method of motion control ever again after spending 20+ hours troubleshooting the resonance problem. 
Anyway, the drives are Maxon DES-50/5 which can be controlled via RS232 or CANbus. But from reading the very confusing manual for the drives, I am thinking they cannot be controlled by Mach3's step/dir output. I would just like someone here who is smarter than I am to confirm/deny this before I go and spend a bunch of $ on other drives. If the drives are not directly controllable from Mach, the manual mentions the use of a CANbus controller. Perhaps the right CAN controller is compatible with step/dir control?
BTW, all the motors have hall sensors, encoder and zero-backlash planetary gear reduction. Each one (drive and servo with options) is worth close to a thousand bucks, so it would be a shame to not be able to use them!

The specs and manuals for the Maxon 4-Q-EC DES 50/5 can be found here: http://www.maxonmotorusa.com/maxon/view/product/control/4-Q-EC-Servoverstaerker/205679

2

Mach3 and G-Rex / Re: stepper motor stalling in x axis only problem

« on: April 12, 2014, 08:40:02 PM »

My Harbor-Freight mini-mill had the same issue. Does your machine use stepper motors or servos? If it has steppers (and it sounds like it does), then read on.

My setup uses Keling 4042d drives and 425oz-in motors. All three axes are the same and use the same settings, but only the X-axis was stalling... and only when commanded to move the table at 38-41ipm. Above and below this feedrate, no problems. I solved my problem by putting the drives in software controlled mode and running the compatible tuning software that was available. I found that only the X-axis had the problem because it moves the least amount of mass out of the 3 axes and has the least rigid ballscrew and motor mount. The mass (weight) of the Y-axis saddle and the Z-axis head provide enough dynamic dampening and resistance that the problem is avoided. The X-axis motor only has to move the table which weighs a lot less comparatively. Now you are probably thinking: less weight= shouldn't stall as easily. But you would be wrong.

Your X-axis stalls because you are experiencing stepper resonance. Step motors typically move 1 step at a time, even during long rapid moves. It appears to be a continuous movement but it is actually a series of really fast steps. At the end of each step, the motor's rotor over-shoots the stop point a bit due to inertia and then rocks back and forth around the step point before coming to rest. When the rate of the rocking motion equals a multiple of the step rate, the motor resonates causing it to lose almost all of its torque. If the mechanical resistance is greater than the motor's reduced torque, it loses sync with the control pulses and stalls.

There are ways to combat resonance, but it cannot be completely eliminated from a stepper system.

1)
If your motor drives support microstepping, using it will greatly help. Also, increasing the number of microsteps can help. This is because the motor arrives at the end of each step more smoothly instead of jumping from step to step. This means it overshoots the end point less which results in less rocking of the rotor and thus less resonance effects.

2)
Change the motor coupling. "Springy" motor couplers allow the previously mentioned rocking motion to become amplified and actually increase the resonance effect. Change to a coupler that doesn't have urethane cushions. Or use a different type of coupler. If your machine setup allows, using a sleeve with set-screws to hard drive the lead-screw may do the trick. Note that you cannot have any motor to lead-screw mis-alignment to do this. If this doesn't seem feasible, you could possibly put some kind of a harmonic balancer or flywheel on the motor. If your motors have shafts that extend from both ends, this could be an easy thing to implement.

3) Use tuning software. If your stepper drives have tuning software, use it. Once tuned, the resonance RPMs of the motor are avoided internally by the stepper drive regardless of the commanded step speed.

4) Lower your max rapid speed to just below the resonance speed on that axis so the motor never hits the resonance RPMs. For me this was only a temporary solution because I couldn't stand having 37 ipm be the fastest it could move along the longest axis. Especially since my motors can now rapid at 245ipm which is scary-fast for such a small machine.