Hacking a linear actuator?

.:John:.

I am working on an exoskeleton that uses linear actuators. They only go 0.5 inches per second with no load. Could I hack them to run faster without losing torque?

These are the actuators.
www.robotshop.ca/home/products/robot-parts/motors/actuators/firgelli-full-scale-actuators/standard-actuators/firgelli-6-400lb-linear-actuator.html

Post Edited (curious_roboticist) : 5/7/2008 1:08:04 AM GMT

.:John:.

Would increasing power to the motor work?

Mike Green

You're not going to get much more speed out of them. Think about it. These use a gear motor where the final gear works a rack and pinion that provides the linear motion. They get the lifting strength (400 pounds) by gearing down the motor a lot which is why it only moves 1/2" per second. You might get a little more speed out of it, but, if you increase the voltage (and current) much more, you'll burn out the brushes or overheat the motor and there'll be increased friction in the gears and more power wasted there as well as wearing out the actuator faster than normal. If you want more speed, get an actuator designed to provide the speed you want. For a similar motor size, it'll have less load capability. For the same load capacity, you'll need a bigger motor.

.:John:.

www.firgelliauto.com/product_info.php?cPath=83&products_id=144
This is another actuator i just found. 200 pound load capacity. 1.5 inches per second. It will work for my purposes.

erco

These are purely mechanical devices driven by DC motors. They have internal limit switches at each end of their stroke. Only two wires, connect to 12V they go one way, and stop at the end of stroke. Reverse polarity and they extend or retract the opposite way to the limit. They are sealed, lubricated, metal gear motors with either a rack & pinion or worm/leadscrew final drive. Unless you are a master machinist with a full shop and lots of parts, do not attempt to hack the internals of these, don't even open them up to peek. They are rated for 20% (intermittent) use at 12 volts. You could, at some risk, use more than 12 volts for faster travel and greater load. My swag guess is 18 volts is a reasonable maximum, if you drop the duty cycle to 10% and use caution for initial tests to make sure that you won't run past the limit switches at higher speed. Start at 12 volts and work up slowly & methodically.

▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
·"If you build it, they will come."

.:John:.

I may try that to. But if I burn out the motor, 136 dollars down the toilet.

erco

I wouldn't immediately assume that the second actuator you list is better than the first. It moves 3x faster, but has half the force rating and draws twice the current (10A). Only if you are using the 6" linear stroke directly (without any external linkage) and don't care about power consumption would I say it is superior. But in a normal application, where this actuator is mechanically driving a pivoted linkage, as your first exoskeleton post suggests, you can adjust your stroke and mechanical advantage to best advantage. Compared to the the second 200# actuator, the first 400# actuator can (with modified linkage geometry) deliver the same force through TWICE the travel distance at 2/3 the speed, using half the current. That is, the 400# actuator can (with proper linkage) deliver 200# through 12 inches at 1"/sec at 5 amps, while operating off the rated 12V supply.

▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
·"If you build it, they will come."

erco

BTW, $136 for a linear actuator is a tiny fraction of what you're likely to burn through in your quest. You're just starting the experimental design phase and you probably won't get everything right the first time; no one does. Depending on your design loads, you're likely to need very strong steel or aluminum, hardened pivot shafts, bearings or bushings, and very strong hardware in general. That means plenty of machine shop time. Unless you're a qualified machinist with a full shop at your disposal, you'll pay $60/hour or more for precision work. Trust me, the time accumulates alarmingly quickly even if you know exactly what you want. Plus, you're dealing with 5-10 amps per actuator, so you're beyond the current capacity of most low-cost motor controllers. For those currents, your cheapest option is to use replaceable plug-in relays like the 12 volt, 20 amp lighting relays in cars. That's just full on or off, so in that case, a slow-moving actuator is better for fine positional control.

Reality hurts, but you need to know that you're starting an expensive and difficult long-term project: $136 is small potatoes. Time to get a government grant!

▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
·"If you build it, they will come."