InputShaping-Plugin-0.2.0-pre3 released
-
@jay_s_uk did you configure your accelerometer? does it show any error?
-
@mfs12 it was an accelerometer issue. It didn't show any errors.
I've managed to get it to run. I did get an error telling me the auto populated frequency of 7 was too low. Increasing this to 10 allowed the test to run.
I also noticed that the progress counter lagged behind by one. So when the 3 run was being recorded, it showed it as being on 2/4 rather than 3/4 -
- can you describe the problem you solved with you accelerometer?
- on my machine 7 is the lowest frequency i can enter. what RRF version are you using?
The process counter shows the number of completed test runs. Not the current one.
-
@mfs12 also, the plugin is fine for capturing data on Z on a delta but its no good for capturing data on X and Y.
This is because it homes the printer and then results in the printer trying to move out of range.
I think automatically homing the printer to run the test is a bad idea -
@jay_s_uk interesting feedback. I have no experience with delta printers. I will check this.
-
@mfs12 I'm running the STM32 version. I can run with 8 but not 7.
also, I created 2 sets of initialise for X and Z. I ran Z and it completed successfully. X wouldn't complete due to the issues described above. I then went back to the Z to run it again at a lower frequency from 10 and it had changed the axis from X to Z
And, when I had them all set at 7 and it wouldn't run, I changed them all to 10. It ran the first test but then threw the too low error, indicating its not respecting the value in the box when its overridden.@mfs12 said in InputShaping-Plugin-0.2.0-pre3 released:
The process counter shows the number of completed test runs. Not the current one.
then should it say completed rather than progress?
-
@jay_s_uk can it be that you need to lower the z height first?
On a delta in homed position (zmax) you can not make a big move in x/y, because one or two rods will reach the end of the tower. -
@cosmowave you don't get a choice. the plugin automatically homes and then runs the test
-
@jay_s_uk Ah ok. Sorry for that.
-
@cosmowave no need. I obviously wasn't clear enough in my original post
-
@jay_s_uk if i understand you correcntly because of the implicit homing testing on a delta doesn't make sense right now. So for delta or in general the initial homing should be optional.
-
@mfs12 yes. the previous plugin was much better in that it allowed you to set the position of the head yourself. optional homing would be great
-
@jay_s_uk said in InputShaping-Plugin-0.2.0-pre3 released:
it allowed you to set the position of the head yourself.
which parameters would you like to set? Having also a start position shouldn't be such a big deal.
-
@mfs12 to ensure you get full movement in Y, its best that the effector is at 0 in X.
Don't forget on a delta, the bed is round so only has full movement across the centre.
You also need to be able to set Z to an appropriate height to ensure the towers don't come into contact with the endstops -
@mfs12 also, the input shaping type doesn't appear to being passed through when I get the F is too low error
-
@mfs12 Consulting with @gloomyandy he said
The Duet3 boards use a step clock of 750KHz we use (as in STM32) 1MHz the Duet 2 and Maestro boards use 937.5KHz. This gives a min F value of 5.72, 7.63 and 7.15 respectively. So I suspect that the Duet 2 and Maestro won't work with an F value of 7 either
You therefore may want to consider upping the default frequency to 8
-
sorry that I have to ask that. I'm doing something wrong. I now get to the point where he tells me that "InvalidOperationException in InstallPlugin: Command cannot be executed because the plugin service has not started".
-
@heartleander81 sounds like you're on an SBC and don't have the plugin service installed
see instructions here https://github.com/Duet3D/DuetSoftwareFramework/wiki/Changelog-DSF-3.x#version-330 -
@jay_s_uk yes I use sbc. I look at the link thanks
-
@jay_s_uk many thanks. It work and install is done