Hi.
Im after some advice on how to make a linear actuator sense an obstruction on closing. I plan to run the actuator of an arduino which has a 7" touchscreen display which will have a timer app/program displayed on it. The actuator will be programmed to open 3 times a day for 1/2 hr then close. The only 2 motor controllers that i have found that sense an increase in current are the AS20 and MAC -3100 which both reverse for a small period if they hit an obstruction on closing before return to the close postion. Both of these are qwuite expensive and the Mac you have to use the dakota digital actuators which once again are quite expensive. I already have a figerelli actuator and want to make it work with this. Is there an easier way to make this happen without having to buy the seperate motor controllers. I have no ezperienxe with arduino as of yet and up till now have just built a prototype using a timer with a spdt relay. All this works fine and is cheap but there is no safety built in if something was to get in the way of the hatch whilst closing. Any ideas??
Current sensor in the motor supply line; pressure sensor on the hatch; optical or ultrasound proximity sensor. Various options.
Which one is most suitable depends on the size of the hatch, how fast it closes, with what force, what your linear actuators are exactly, what obstructions may be present, how sensitive it has to be, what the consequence of failure to detect an obstacle may be, etc.
Do you have feedback on the actuator? Or some other kind of sensor on the hatch?
If you drive it towards closed for 10 seconds and it doesn't hit the 'closed' switch then it's probably time to stop driving in that direction.
If that means that by then instead of one cat you have two halve cats, it's too late to start reversing...
No, just the cat is a little narrower in the middle and a little more angry.
No i have no feedback from actuator or sensor on the hatch. The set up now is very basic as is my electronic knowledge ha ha. Im suprised if got this far. There are no open or closed switches either i have just positioned the actuator accordingly so fully open the lid is close to 90 deg and fully closed is a nice tight seal. With the As20 you can set the sensitivity and make it reverse with as little or as much force as needed. Im interesyed how the inline current sensor that was mentioned above. Could this provide the same thing as the As20 which is 250 pound to buy with programmer. The actuator closes quite slow and i thought of pressure sensors. I would have to run them all around the hatch though which is nearly 2.0 x .9 m in size
The idea of this current sensor is that the motor current goes up the moment the motor encounters extra resistance.
Whether it works for your device depends on the device itself, and the required sensitivity.
It doesnt have to be super sensitive but i guess the current draw is quite large when the actuator starts up so can that be overidden?? I think the max current draw is 5 amps so could it be set to reverse for 5 secs if it say goes to 6 or 7 amps? After reversing for 5 secs it then has to continue back down. Is this all done in the arduino programming??
If the max draw is 5A it'll never go to 6-7A. Make sure you know the actual values, so you can get an appropriate current sensor.
Other than that, sure, no problem to program that in an Arduino.
Hi,
You need to put a sprung hinged edge on the door with one or a number of micro-switches under neath it.
The spring keeps the edge away from the door and the switches.
When the door encounters an obstruction the edge moves towards the door edge and trips the micro-switches.
Tom.... 
So with a current sensor and correct programming thos may be achievable. I guess i would need to add a relay board then to reverse polarity if i did away with the expensive AS20 actuator controller??
That, or an h-bridge driver. Where possible I would go for solid state solutions rather than relays.
But again it depends on your actual actuator.
The Sparkfun Monster Motor Shield is a good example of a H-bridge motor controller with current feedback.
If you are using the standard Arduinio PWM frequency, add an extra 1uF capacitor on the feedback pin. They didn't put enough filtering on the Sparkfun board.
The thing is, current feedback doesn't tell you a lot about the motor. It also gives you lots of strange readings when starting to move. Current cannot tell you the difference between "door is blocked" and "we only just started moving a few milliseconds ago". So you have to put a lot of time-based logic into your current sensing.
It usually is better to use switches. Like if it absolutely has to NOT squeeze the cat, then put a switch on the edge of the door, maybe with a wide 'spacebar' on it, so that it can feel the cat blocking the door and cut the motor off well before the current starts to go up.
Ueah good point. Thanks for everyones input it has been very helpful. Time to buy some new parts and start playing again. Much appriciated