Created Date: September 13,2012

Created by: Tinker

I haven't done that sort of app, nor have I used the Sureservo brand at all, but I do have a suggestion, study the manual very carefully before buying the system. And unless you have a budget where losing ~$1100 is not a big deal, handle all the parts and packaging carefully and plan to evaluate the system within the 30 return period.

“My initial thought is to use a sure servo in torque mode “

I’m not so sure that would work, I’d think position mode with a torque limit would be better. Since the motor and your chuck will have inertia, if you spin the cap on at any reasonable speed it will likely be over tightened by the inertia of the moving parts before the drive’s torque limiting takes effect. I suspect that is part of you problem with the friction or hysteresis clutches which I assume are driven by a constant speed motor, while the clutch decouples the motor, the chuck and parts of the clutch might still slam the cap over tight.

Once you have the control of a servo system, I’d think spinning the cap on at relatively high speed until just before it bottoms out , and then creeping up to tight at a relatively slow speed would be more effective.

Created Date: September 13,2012

Created by: Adisharr

The main thing you'll run into using a servo in torque mode to tighten to a specified torque is that the feedback from the amp section to the control section is generally too slow to give any kind of useful precision. The cap is going to be over tightened before you can do anything about it.

Another way to do it uses the servo motors position error to kill the motor when the error exceeds a specified threshold. The encoder loop is extremely fast - unlike the torque feedback loop. You would keep you 're torque levels consistent and then just change the error threshold to give different torque levels for your cap. Some experimentation will be required to find the ideal values.

This technique works best when the motor uses sinusoidal commutation. Other commutation methods will produce torque ripple which will affect your results. I don't know what the SureServo uses.

Created Date: September 13,2012

Created by: allukes

I 've used the Sure servo with the torque limiting feature in Position Mode.

You can program the percentage torque using the Modbus interface. Its been a few years since I worked on that machine, but I am pretty sure when it reaches the torque limit it just sits there. Its a different setup from the torque alarm. Like Adishar says, you can set your position error limit if you want it to fault. If you have time, you can read the position over the Modbus link and look for it stop moving. You might be able to monitor the motor current also. Look at the end of the Sure Servo Manual for all of the Modbus registers.

What about a load cell?

Created Date: September 14,2012

Created by: automa

Good thoughts here. I do believe the inertia of the current system is a portion of the problem. It is driven by a constant speed motor that operates at 160rpm. The bottle and cap are double threaded so the number of rotations till the cap seats can range from one to three, depending on how the operator applies the cap. The position error is an interesting idea and may work well especially since the cap socket will be directly attached to the armature, so the 95% of the centrifugal will be generated from the armature. I will have time to clear the error between bottles. I could even apply the brake if it will respond quickly enough to remove some of inertia from the equation.(doubt it). How would you go about using a load cell allukes?

Created Date: September 14,2012

Created by: Bob S BN

The position error is an interesting idea and may work well especially since the cap socket will be directly attached to the armature, so the 95% of the centrifugal will be generated from the armature.

I will have time to clear the error between bottles.

I'm not sure how the SureServo works, but I did a deburr application with a 50 watt servo (Parker) and a 1 " end brush on the armature, and the motor/brush assembly mounted to an air slide.

The "following error " was a variable that could be monitored. It ran about 20 counts when the motor wasn't loaded, and would increase to about 30 when the brush was in contact with the part. I monitored this variable, and when it increased over a set limit threshhold, reversed the air slide as I had confirmation that the brush contacted the part.

It didn't "set an error ", it was just a variable that could be monitored. Like I said above, I don't know if the SureServo has this "following error " as a variable that can be monitored, but the concept may work for your application.

Good Luck,

Created Date: September 12,2012

Created by: automa

I am looking to build to precision capper for small glass vials. My initial thought is to use a sure servo in torque mode to control the cap torque. Has anyone used the sure servo system for this type of app? How well does the controller control the output torque from the servo? Torque is 6in/lb for one vial and 8in/lb for another. I plant to use a 400w system. We have a hard rime getting consistency and longevity out of a friction clutch or hysteresis clutch setup.

Thanks!