Society of Robots - Robot Forum

Mechanics and Construction => Mechanics and Construction => Topic started by: gibsong on July 16, 2015, 08:03:34 AM

Title: Pressure feedback attached to linear actuator or servo
Post by: gibsong on July 16, 2015, 08:03:34 AM
I'm trying to design a bot that will use a motor or linear actuator to move an arm in one direction, apply a set amount of pressure, and then retract. Can anyone help me come up with a way to do this? I already have a prototype that uses a DPDT switch and a linear actuator, but the pressure sensor is not connected to these and must be read manually through an readout on a computer screen.
Thank you in advance!
-Gibson
Title: Re: Pressure feedback attached to linear actuator or servo
Post by: mklrobo on July 16, 2015, 11:52:21 AM
 ;D Hello!
I will try to offer an opinion, where permissible.
To retract the linear actuator, reverse the polarity to the motor; if I am interpreting you
right.(?) :-[
A pressure measurement process, has been to measure the amps the actuator takes when
pressing. You will have to come up with an amperage scale. For example;
When the actuator is pressing, measure the amps; when the correct pressure is reached,
stop the actuator. In other words, make  a amp Vs pressure graph, and then you can program
the actuator accordingly. To find the formula to the graph, see my post in the miscel section,
Solve formula directly from a graph. This is not the best way for controlling the actuator,
but, unfortunately, it is used.  :'(
Title: Re: Pressure feedback attached to linear actuator or servo
Post by: artbyrobot1 on July 17, 2015, 11:45:16 PM
to measure pressure, you want to install a  pressure sensor-aka strain gauge.  Youtube how to install these and how they are used and you should be good to go.
Title: Re: Pressure feedback attached to linear actuator or servo
Post by: gibsong on July 19, 2015, 10:20:46 AM
I'm currently using a FlexiForce thin pressure sensor hooked up to a phidgets 8x8 board. It gives me a readout on a computer screen, which is helpful but still manual overall. This is independent of the DPDT switch that controls the linear actuator (which in turn moves the arm). I'm trying to find a way to automate this so that once I have the sensor properly calibrated I can tell the bot to apply a set pressure, and once that pressure is reached the arm will retract and return to the original position. Sorry for any confusion earlier.
Title: Re: Pressure feedback attached to linear actuator or servo
Post by: artbyrobot1 on July 21, 2015, 06:27:12 AM
Then it sounds like a software issue.  You want that pressure reading to be fed into your code who listens for that reading and makes decisions based on it.  You need to code your robot to have "brains" is what is seems like to me...
Title: Re: Pressure feedback attached to linear actuator or servo
Post by: cyberjeff on July 22, 2015, 08:48:27 AM
Force sensing resistors are widely available. You would need the analog input (and a resistor) on your microcontroller to read this.

http://www.makershed.com/products/round-force-sensing-resistor-2-pack?gclid=Cj0KEQjw_rytBRDVhZeQrbzn_q0BEiQAjnbSHCN-9dRRV3KrBFrkvY_6J8CpPJR2ngcHFYnEv9RIgG0aAoXa8P8HAQ (http://www.makershed.com/products/round-force-sensing-resistor-2-pack?gclid=Cj0KEQjw_rytBRDVhZeQrbzn_q0BEiQAjnbSHCN-9dRRV3KrBFrkvY_6J8CpPJR2ngcHFYnEv9RIgG0aAoXa8P8HAQ)

It costs a bit more to ship them than to buy them!
Title: Re: Pressure feedback attached to linear actuator or servo
Post by: ProgressiveAutomations on August 06, 2015, 04:51:10 PM
You can use an Arduino as brains.

You will also need a motor controller to boost the current to be used by the motor, and then you can implement feedback using your force or current sensor.  Some motor controllers have built in current sensors, so that would give you easy monitoring.

You can measure current, rather then force, because the amount of force an actuator can output is proportional to the current input. One thing you have to be aware of when programming is that when the actuator starts, there will be a current spike to get the motor going.  You'll need to program a slight delay to skip that, else your max current limits will be  wrong.