Society of Robots - Robot Forum
Electronics => Electronics => Topic started by: drinu on March 05, 2013, 09:09:25 AM
-
Hi,
I was planning to use linear actuators for a robotic arm, but after some research I found out that the ‘cycle’ listed in the actuator’s datasheet is the maximum time that the motor can be on per being off. Most actuators have a cycle of 10%. Does that mean that for every 1 minute on, I have to switch it off for 9 minutes? or am I interpreting it incorrectly?
-
You are interpreting it correctly, although the maximum contiguous time on should also be specified. It may be 1 minute / 9 minutes. Or it may be 10 seconds / 90 seconds. Or it may be 10 minutes / 90 minutes.
-
So basically an actuator is out of the question for a robotic arm right?
Are there actuators with a higher duty cycle (like 50% or 75%) ??
-
Yes, there are actuators with 100% duty cycle. They cost more than actuators with 10% duty cycle.
-
Thanks for the reply.
Do you know of a company which sells these type of actuators please?
I am looking for an actuator with the following specs (approx.):
Force: 6000N
Stroke: 100mm
Speed: 8mm/s
Supply: DC 12V or 24V
Duty cycle: 75% to 100%
-
Does anyone know of any site selling actuators with the specs mentioned in the post above please?
I can only find AC actuators with a 100% duty cycle.
Any help would be appreciated.
-
You are unlikely to find a 6000 Newton actuator with a 12-24V drive and 100% duty cycle. The amount of power you need will generate a lot of heat because of the very high current needed at those voltages, and getting that heat out is likely the problem and why the duty cycle is low with those specs. 6000 N is a *lot* of force -- lifting six overweight men!
If it's an option for you, you could take one that has a too-low duty cycle (but as high as you can find) and look at adding advanced cooling (recirculating refrigerant or whatever) to reach the desired longevity.
Another option might be to use more than one actuator in parallel.
Maybe there is one that can do it, but if it's not available at McMaster, and not in the catalog of makers like Kollmorgen or Thomson or whatever (there are several large names in industrial motion control,) then it probably isn't all that likely to be even semi-affordable...
http://www.kollmorgen.com/en-us/products/linear-positioners/electric-cylinders/ec5-series/ (http://www.kollmorgen.com/en-us/products/linear-positioners/electric-cylinders/ec5-series/)
http://www.thomsonlinear.com/website/com/eng/products/lms.php (http://www.thomsonlinear.com/website/com/eng/products/lms.php)
-
Thanks jwatte for the reply.
So the off cycle is required for the motor to cool down? or are there other factors that causes the short cycle?
Another option might be to use more than one actuator in parallel.
I think that would be the best option.
-
Also, can I operate an actuator using PWM?? or will the pulses cause damage to the actuator's mechanism?
-
I am not a mechanical engineer, but from what I understand, heat is the biggest problem.
There may also be precision/strength design choices, where an actuator is good for X cycles, and has a one-year warranty, and a 100% duty cycle would hit X too soon, but a 10% duty cycle means you won't hit X until after the warranty expires. Mechanics wear out no matter how well you grind, temper, and lubricate it.
The linear actuators I know of all use an inductor coil -- either a solenoid style, or a motor or stepper motor style. All those styles can use PWM to reduce the average current, rather than lowering voltage. However, if the actuator also has a feedback sensor and/or controller, using PWM may be a problem for that logic, if it doesn't have a separate logic power rail.
-
Thanks for the reply.
In some datasheets the duty cycle is listed as "S2-10min". What does that mean?
-
http://letmegooglethatforyou.com/?q=duty+cycle+s2-10+min (http://letmegooglethatforyou.com/?q=duty+cycle+s2-10+min)
-
I think looking into electric linear actuators is going to be a dead end for building an arm. I'm not aware of any existing arms that use these as the primary actuator. It's still probably possible, but since any such implementations whatsoever are rare, it's almost certainly not doable on a hobby budget.