Hello-
Here’s what I’m trying to do:
I have a firgelli automations FA-Po-150-12-4 linear actuator with a built in potentiometer. I’m trying to build a machine where the following happens: There is a “car” that moves on a straight 15’ track. Depending on where the car is on the track, the linear actuator needs to be extended to a specific length, and that length needs to change as the car travels.
For example, when the car is at, say, 6’ from one end of the track, the actuator will be extended 2.2"; when the car is at 6.5’ the actuator needs to be extended to 2.3" and so on. I would like to be able to dial in the actuator position so that I can control exactly where the actuator will be at a bunch of points of car travel (maybe every 6" determine a specified actuator position). Ideally I want the actuator to transition smoothly between these indexed points (so that for example, 1/3 of the way from point “A” to point “B,” the actuator is 1/3 of the way between the “A” and “B” set distances).
The motor on the actuator is 12vdc. According to factory specs, the motor can draw a peak of a little over 3A. the built-in pot on the actuator is 0-10k. The car has wheels, so maybe some sort or rotary encoder could measure car travel distance? Everything in the system can be wired and powered externally. I want the system to be stand-alone if possible (not wired into an external computer). To complicate everything the whole system will be exposed to water, but I expect I’ll just figure out how to waterproof all components.
I would really appreciate any and all advice on hardware, software, coding, online resources etc. I’m a total automation novice, I’m fairly mechanically inclined but I’ve never done any automation and haven’t done coding since I was a kid in school ( a long time ago).
Thanks very much!!
-CB
Thanks JB-
The breakout of subsystems makes perfect sense.
The actuator is controlling the tilt of a grinding disk that needs to grind concave surfaces at specific angles. Because of the funky geometry involved nothing is linear. Also, I don’t know what the actuator distance needs to be at any given point of car travel until I get the machine up and running and do some experiments with it.
The actuator starts at a zero position, goes to it’s maximum travel around the mid-point of the car’s travel, then goes back to zero as the car continues to the end of the track. The actuator travel at intermediate distances are not linear or even a simple curve. But after experimentation I will eventually know actuator extension at distances, so I can plot those and maybe approximate them with a formula. I have mathematician friends who are easily bribable with beer.
This whole assembly is part of a larger machine and the linear actuator is the only part of the machine that needs automation. The car itself is controlled by off-the-shelf inverter/controllers (someone else is installing that system). It will be operated via manual controls in all likelihood. It will probably have a 3hp 3-phase motor, or something thereabouts. The track is approximately 17’ long. It is vertical. The car will weigh approximately 200lbs and will be counterweighted.
As far as ways to accurately measure travel of the car, the car is driven along by a sprocket and chain, so I was wondering if some kind of rotary encoder would be a good way to measure that. Do you agree?
Also, I have a power supply and push-button controller for the actuator from Firgelli Automation that I’ve been using to play with the actuator. I’m not sure if that would be a helpful part of a built automated system.
I wish I could get you drawings but due to IP reasons I’m pretty limited.
I’m just totally lost as far as what hardware to start with. I know i need to dive in but I don’t know where to dive. I really appreciate you taking the time to help me.
Best,
Cory
Let’s look at the logic of your system and split this into subsystems. After that we can look at the implementation and hardware you can use.
1- Quantify the car’s position.
2- Translate the car’s position to a actuator pot value.
3- Output the desired position to the linear actuator.
This can be done with some simple math since we can assume that there is a “linear” correlation between the car’s position and the actuator’s position. You start off by finding correlations:
1- (Car’s minimum position, Actuator’s minimum position pot value)
2- (Car’s maximum position, Actuator’s maximum position pot value)
Now you have 2 numerical points to make a graph using the linear slope equation y= mx+b.
mathforum.org/cgraph/cslope/slopeintercept.html
So with a given car position, we plug it into our equation and we get the actuator’s position as output. Easy as pie
The trick will be finding a method of determining the car’s position on a linear scale. Can you provide more details about the car and track such as precise dimensions, weight, drawings, etc?
The wheel encoders would work well because you can calculate distance traveled and use that value to determine what the actuator’s position should be. You will need to do some custom programming on microcontrollers and/or PCs.
Here is what I envision, maybe it will give you some ideas:
Ideally a microcontroller on the mobile gathers data from the encoders on the wheels and also possibly the start position on the track. Sensing the start position could be done using a reed switch and a magnet or even an IR sensor. Feel free to get creative.
On the track side in your command center (you will have a command center right?) you can set up a PC or a microcontroller to control the linear actuator portion and gather data on your experiment.
Linking the two via wireless completes the system. You could use cables but this gets really messy and the cables might get twisted 
If you were to use a slip ring (rotating connector) you might get unwanted noise on the communication lines.
The theory of operation would be to have the mobile collect and send it’s “state” to the command center that in turn decides where the actuator should be in its stroke. This is all done in a matter of milliseconds.
Which technology you decide to actually use depends mainly on your familiarity with PC and microcontroller programming. If you have PC programming experience, microcontrollers will be a breeze to master and open up a world of possibilities for your projects.
The only limiting factors in your system are the speed of the mobile and the speed of the actuator which will probably need to be synchronized.
If you were to use a slip ring (rotating connector) you might get unwanted noise on the communication lines.