I manufacture Linear Actuators for the RC Plane guys and I am Developing one for the Robotic guys now. I need some help on finalizing what you guys need.
This current actuator is not proportional and was designed for retracting large scale model air plane landing gear. So it goes to one extreme and stops. It has about 36 lb. of force at stall and takes about 4 seconds to go 1 inch.
The new Robotic version will be proportional like a servo. The controller will be built right into the actuator, it will connect up just like a servo. It should retail for $100 each.
So what I need to know from you guys is the following:
Could you use this?
How much throw would you like to see?
How fast, remember the faster it is the less force?
How much force?
If we sold them in a package, how many would you like to see?
Hi Doug, I took the liberty to add the url and some images from your web site to your post. When you post again they will actually show up.
On the speed question. I think robot builders require two different products. Slow and powerful, say 4 seconds for full throw, and a fast but less powerful one, about as fast as a servo. A slow servo moves 60° in around .25 seconds, so for 180° it can take .75 seconds. So how much force can be generated with this speed, and is it enough to do real work? I’m looking forward to having an affordable linear actualtor available for the robot building community.
nice to see some options coming out i,m shure this would clean up the ankles on the bipeds !!! ,is the power or speed gear change on the bottom or inside ,it would be sweet if u could swap a couple of gears to customize for the travel/force/speed combination u need for you particular app and “ADD HEAT SINKS” and or overtemp,overvolt protection (make it “goober proof” for me)
Yhea it would be a gear change, and that would be hard to make adjustable by the user. But I will make two versions for sure, one for fast movemen and one for high power slower movment.
I started the below thread a while back concerning the use of linear actuators. Beyond arm use, they would be useful for jobs that need power instead of speed. Just for fun I’ve been wondering what it would take to make an ROV out of my snapper riding lawn mower. Positionable linear actuators would be handy for the stearing, gear shifting, and clutch/breaking task.
You know this is readily available, right? Warner Electric for one makes several different types or linear actuators and they have position feedback. They really are not SES scale like I suspect this LADO product will be, but for a riding mower or go cart they would certainly be in the ballpark. We use some to shift an automatic transmission in an automobile (P,R,N,D) and in some other products. One of them has an 8" stroke that will lift 75 lbs. warnerlinear.com
The actuators I am making would be much smaller. Something that could be used in place of a servo but have much more power and a long linear stroke. Do you guys see a use for this?
Just looking for some clarification; are we talking the same size as what is shown with perhaps a different throw and gear setup, a scaling of the design up or down, or a total redesign?
The linear actuators may have an advantage in power consumption. Worm gear types of drives resist reverse movement of the gearing much better than standard gear drives. A linear actuator should hold the final position under load using very little power compared to a servo with standard gearing in the same type of application. The Firgelli site says they have what appears to be a servo controller compatable actuator comming out this spring, but they may be having problems supplying their current stock, so it may be vapor hardware. If possible, I’d make actuators with a throw distance of a couple inches at least to take better advantage of their capability. In the gearhead discussion I mentioned making a larger arm using actuaors which I think would be an interesting project.
I was messing with a worm gear drive servo a few months ago but got discouraged by the cost of the parts to make one. One of the down sides to keep in mind about worm gear drives is they generally have really poor efficiency and what you make them from is really important because of wear. This is something LADO really needs to consider when shifting their design from the relatively sparse duty cycle of an r/c landing gear actuator to the nearly continuous duty of an actuator used for locomotion of a robot. In the $100 a pop range you are competing with higher end rotational servos not only for torque output but mechanical reliability as well.
I can see some potential for use of linear actuators in leg designs for hexapods and octapods too. I could see some nice powerful leg designs with the right actuator. There would probably need to be one or two new SES brackets made for this to work though.
At this point I am open to suggestions. I am considering a slightly smaller design that would use a gear motor that would allow me to have 3 version in relation to speed and power. This unit would be about the same height but thinner.
Also as zoomkat said linear actuators hold position under load very well. When I did my force tests, even when I removed power the actuator still held the full 36lb. of force with no power!
In regard to EddieB’s statement:
In the final stages of testing for the model airplane version I ran 6 actuators through 2000 cycles to see if I could get ware on the screw. I was surprised I didn’t see any at all. One thing that I found was the tolerance on Linear Actuators must be much higher then you would think. With that said, I am very curious to see how they hold up in a real Robot environment. As EddieB said Robotic usage is much different then landing gear.
We own all our own CNC equipment at Lado. So making changes and new designs is no problem!
Thanks for all the feedback guys, keep it coming. This will really help me make a great product for all the Robotic guys.
Also if you guys want to see a video of my actuator bring a large landing gear up, check out my site. www.lado-tech.com
The actuator is running at about 60% speed, this is to make the retracts look scale. Much higher speeds are possible.
Depending on limb movement method, wether directly attached, or indirectly thru cables and wires, mechanical advantage is capable of being used to both speed up actuation vs that actuators real speed, or to multiply the strength of the actuator “past” it’s stock capacity.
I am very interested in these actuators. Here are my thoughts:
More speed is needed for walking bots. It sounds like they will be plenty strong anyway. In my opinion servos are pretty slow. Watch a video and see how painfully slow something like a Robonova-1 walks. I would say you definitely should aim to have a version that significantly beats the fastest servos.
They should work in a fairly wide range of voltages, say 6v - 12v (2S or 3S lipo). It’s annoying that most servos can’t handle 8.4v (2S lipo) without cooking under load. Look at the walking robots like those from Kondo, Kyosho, etc - they are using higher voltage servos.
Consider offering adapter brackets to work with servo mount patterns. If people need to fabricate brackets, you won’t sell very many.
They would need to have position feedback and current sensing. Again, all the modern robot servos are starting to have these features built in. Take a look at the Bioloid servos to see what I’m talking about.
You need to bring these to market for around $100 or so each, which is about the price of a high end digital servo.
Packs of multiple actuators at a reduced per-unit price would be appealing. Maybe 4 packs would be nice to see.
To get people to move away from servos, you need to beat servos. But if you beat servos, people will adopt actuators, because we all want to build better robots! Good luck - We’ll be watching.
…though allowing them to respond to servo-type pulse-width signals would make them backward-compatible, and thus able to be used with existing controllers, and letting people upgrade to linear actuators incrementally, rather than having to jump into a whole new technology cold. Whether responding to PWM natively, or by jumper selection or flash configuration, allowing potential users to utilize their existing programming and electronics with the new actuators - or using a mixture of linears and traditional servos, would allow for more flexibility in design, as well as making the transition easier, both financially, and from a control and programming standpoint.
It’s a lot like the SSC-32 versus the Mini-SSC II; the SSC-32 has a lot more features than the MSSC II did, but it’s still capable of responding to the other controller’s command set. A user can upgrade to the SSC-32, use it with their existing programming, and then rewrite to take advantage of the new command set as they become comfortable with it.
,is the power or speed gear change on the bottom or inside ,it would be sweet if u could swap a couple of gears to customize for the travel/force/speed combination u need for you particular app and “ADD HEAT SINKS” and or overtemp,overvolt protection (make it “goober proof” for me)
More speed is needed for walking bots. It sounds like they will be plenty strong anyway. In my opinion servos are pretty slow. Watch a video and see how painfully slow something like a Robonova-1 walks. I would say you definitely should aim to have a version that significantly beats the fastest servos.
