scales vs axis.N.motor-pos-fb vs axis.N.joint-pos-fb

More
16 Dec 2018 17:54 - 17 Dec 2018 01:12 #122485 by Nitram
Hi.

From what I can see, the usual HAL file configuration for feedback from servo encoders is along the lines of:
net x-pos-fb <= hm2_7i76e.0.encoder.00.position
net x-pos-fb => axis.0.motor-pos-fb

Which of these two pins (axis.0.motor-pos-fb or axis.0.joint-pos-fb) provides the data to LCNC as definitive actual joint position? Am I incorrect in saying that the value of axis.0.motor-pos-fb drives the GUI screen position display, or is it ultimately driven by axis.N.joint-pos-fb?

When the motor encoder fb pin is connected to axis.0.motor-pos-fb, from what I can see, when the axis applies a backlash offset, it moves the motor/encoder the required amount but freezes the display given that the backlash move does not create any axis positional change, hence even though the encoder turns to take out the BL, the position and readout understandably stay static as this is a compensation move not one which changes joint position.

From the above, what I would be keen to know is when scales are also attached to an axis which has dual feedback via its motor encoder, if we assume that scale readings are an absolute axis position which should never really need any form of backlash, does it make more sense to connect their output to the axis.N.joint-pos-fb pin and leave the motor encoder to axis.0.motor-pos-fb?

Will doing this have any unforeseen negatives or is this even possible?

Thanks,
Marty.
Last edit: 17 Dec 2018 01:12 by Nitram.
More
17 Dec 2018 02:15 #122501 by cmorley
axis.0.motor-pos-fb in an input pin, fed from the feedback device.
axis.0.joint-pos-fb is an output pin.

For a trivial trajectory machine they should be the same.

Chris M
The following user(s) said Thank You: Nitram
Time to create page: 0.065 seconds
Powered by Kunena Forum