Cross Coupled and Skew Controlled Gantry Calculator

[Sina.Sattari]

Usually setting up a gantry in cross coupled mode or skew controlled mode, can be a challenging task if you don't want to use the default motor numbers explained in the application notes available on Delta Tau's website. The following Excel spreadsheets will generate all the necessary settings for these setups.

[JohnR]

Sina, Can you verify that the skew control methid works as described in the spreadsheet ? I can not get the axis to opperate. Thanks john

[bradp]

[quote='JohnR' pid='299' dateline='1264280093'] Sina, Can you verify that the skew control methid works as described in the spreadsheet ? I can not get the axis to opperate. Thanks john [/quote] It does work but has some limitations. They are that he amount of correction is limited to the maximum value of he DAC output. So it can only correct for a limited amount of skew. Also, since it runs from actual position the axes must start with the same actual position. Both of these make it difficult to get going. We are fixing both of these problems in a new version. It was put on hold for awhile but we hope to have it done this week.

[JeffLowe]

>It does work but has some limitations. They are that he amount of correction is limited to the maximum value of he DAC output. So it can only correct for a limited amount of skew. Also, since it runs from actual position the axes must start with the same actual position. Both of these make it difficult to get going. We are fixing both of these problems in a new version. It was put on hold for awhile but we hope to have it done this week. [/quote] Am I correct in assuming another limitation is that this does not work with PMAC Sine Commutated motors? Do you have an example user written servo that takes the two filter outputs and does the math for R + theta, and R-theta and outputs thes to the default commutation routines?

[bradp]

No, both will work with commutated motors (sine or PWM). The addition point of the cross coupled terms into the "main" or "other" control loops is before the DAC command is sent to the commutation algorithms.

[JeffLowe]

[quote='bradp' pid='379' dateline='1271709982'] No, both will work with commutated motors (sine or PWM). The addition point of the cross coupled terms into the "main" or "other" control loops is before the DAC command is sent to the commutation algorithms. [/quote] Thanks Brad. On more careful reading I see that this approach uses the Tcomp node as the injection point. Previous versions that I was aware of used the DAC bias node which, of course did preclude sinusoidal commutation. Not that I really need them for my application, but would Torque comp tables still be useable as well? J

[bradp]

Right, a torque compensation table should not be used with this setup. If someone needed both commutation and TCOMP we could work out another way, but this method is good for most setups and does not get so complicated.