force torque sensor manual calibration

Toby

Hey All,

Our application involves a payload that is quite large, and can vary in its CG / total mass. I was curious how the FT150/300 calibrates itself, so that we can replicate that in our code to run it when needed. I attempted a simple statics calculation to account for CG, but the output was not correct.

Best,
Toby

louis_bergeron

@Toby

rq_set_zero is a function defined in the file accessor_capt.script.  In this file, the function is defined like this:

 

def rq_set_zero():
  	if(socket_open("127.0.0.1",63350,"acc")): 
  		socket_send_string("SET ZRO","acc")
  		sleep(0.1)
  		socket_close("acc")
  	end
  end

To replicate it, your socket connection must be done on the 63350 port and the real command for the sensor is "SET ZRO".

Toby

additionally, I tried sending 'rq_set_zero' through the socket, but this didn't change the FT300's readings at all

URTrent

Toby, I haven't used socket commands, but to verify functionality have you tried using the "rq_set_zero()" through just a simple script block in Polyscope?  I've been using this extensively to compensate for the weight of tools I pickup when trying to apply specific forces or set heavier parts down gently with success.

Toby

@louis_bergeron Thank you, this works well.

I use python-urx: https://github.com/SintefRaufossManufacturing/python-urx for the connection, and use this python code to reset the FT300:

def reset_FT300(rob, wait=True): <br>&nbsp;&nbsp;&nbsp; prg = '''def prog():<br>&nbsp;&nbsp;&nbsp; if(socket_open("127.0.0.1",63350,"acc")): <br>&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp; socket_send_string("SET ZRO","acc")<br>&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp; sleep(0.1)<br>&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp; socket_close("acc")<br>&nbsp;&nbsp;&nbsp; end<br>end<br>'''<br>&nbsp;&nbsp;&nbsp; programString = prg<br>&nbsp;&nbsp;&nbsp; rob.send_program(programString)<br>&nbsp;&nbsp;&nbsp; time.sleep(0.25)

Afterwards, I can take into account the location of my weight by taking my mass vector: [0, 0, -9.81*mass] and rotating it into tool-frame, and then subtracting from my force readings.