Delta Tau Forums

Full Version: Powerbrick with Rotary Motor and Linear Encoder
You're currently viewing a stripped down version of our content. View the full version with proper formatting.
I have a Powerbrick LV that I am trying to get a rotator motor(6 poles) with hall sensors driving a lead screw stage (2 mm/ rev) with a linear quad encoder configured (0.1 mm resolution). When I go through the basic setup if I configure it as it should be it will not let me get past the hardware interface section. Basic issue is that there appears to be no lead screw length setting to match the linear encoder to help the phase track in between the hall sensors. It seems that I have to either pretend to have a linear stage with the real encoder or the real rotary stage with a fake rotary encoder. When pretending it is a linear stage I used a cycle length of 0.66666 mm (2mm/3 pole pairs).

Either way once I set that up and run the basic test and set the stage fails auto. On manual it sometimes fails the 6 step voltage but even when it does pass it, it seems to fail the open loop test.

Checking the UVW they seem to go in the proper pattern as per the powerbrick manual with the 3-1 with a the correctly set Negative value set for this "backward" direction. Cycle from 3 to 3 distance matches the 0.6666mm cycle length I trued using. I found that the phase value does not track properly but I think this might be caused by the manual testing "passing" values that were not really correct.

I tested the fixed current Locking Method as in the manual. Running the current would look the stage in the 3 UVW state value and would rotate the stage to that state if I started in a different state but the phase would change it appeared. Also when I was doing that test when I activated the motor the encoder monitor would show the encoder changing when the stage was not moving with also made the phase change.

I know the stage is fine as I pulled it from my working ACS based controller system to try to figure out how to configure this.
I doubt that the automatic system setup will be able to configure this for you.

Essentially, this is a dual feedback stage. Halls are used for commutation, and velocity loop. Linear encoder used for position loop.

This can be set up manually following procedures described in the Power Brick LV (ARM or prior) manual.

First, you want to set it up for commutation and velocity using the halls as if they were a quadrature encoder that yields 12 counts per revolution (because of x4). In this step, you would verify current loop, phasing, and open loop functionality.

Next, you would point Motor[x].pEnc to the linear encoder conversion table entry, perform an open loop test to make sure the motor moves in the correct direction of the encoder counting. If not, switch EncCtrl from 7 to 3 or vice versa and try again. Lastly, try to tune the position loop.
Thanks for the reply. There were several electrical problems going on that were making everything much harder than it should have been. The Halls sensors lack pull ups so were not triggering properly. Also their was a grounding issue from the integrator's cable that was making the encoders run wild anytime it was actually powered on.

Once those were sorted I could use the manual setup to get the motor going pretending that it is a linear motor with a linear encoder.