servo madness

[jensenjim]

my final attempt for the day, I cant get the graph to zoom out

[m_c]

That is going the correct direction, but obviously still needs work.

What you have there, is the output rises and hits it's limit (P * max output = 100), with the actual position lagging behind the commanded position. Although you can't see the rest of the plot, what you can see is enough to work with.

Increase J(erk) up to 4000000, as it's extremely low. Jerk is how quickly acceleration is applied, and is visible in your last plot, as the initial curve in each plot, before things hit peak acceleration (where the plots increase is essentially a straight line).


Then do as I mentioned in my previous post about continually increasing P and D, then adding a little bit I.

[jensenjim]

[jensenjim]

[TomKerekes]

Hi jensenjim,

Please post the IIR Filters Screen for this Axis.

[jensenjim]

[TomKerekes]

Hi jensonjim,

Ok filters are all clear. I'm at a loss at how you reversed the drive direction without setting the Output Gain to -1 as we requested. I suppose you changed something without telling us.

Anyway, the next step would be to add a 2nd order low pass filter and increase the D Gain to something like 10 or higher. A D value of 0.001 will be no different than none at all. Please read this article on Acceleration mode servos like yours carefully.

Please remove Deadband by setting range to 0 and gain 1.

[m_c]

Now you're making progress.

Hi jensonjim,

Ok filters are all clear. I'm at a loss at how you reversed the drive direction without setting the Output Gain to -1 as we requested. I suppose you changed something without telling us.

Reply #37, Jim swapped the output wires from the Drive.

[TomKerekes]

Hi m_c,

Reply #37, Jim swapped the output wires from the Drive.

Ahh, thanks I missed that.

[jensenjim]

ok first I would like to say thanks for all the help, its been a few hours with this. here is my crossroads
1. the kflop servo setup is going to be to difficult to get the 2 servos running smoothly, the drivers having pots and no real documentation. also one drive is wired backwards so they both pull the same
I spent a few hours trying to get the master/slave setup and its not happy. I could try to figure a drive setup that uses only one sery for both sides of the gantry this beast is long and heavy so only one side drive would not be enough.

I might look for better servo drivers, the servos in this setup are high quality and good shape but still will have setup issues with the master/slave on single axis

my other option is to use the geko stepper drivers I have with stepper motors and figure a controller out

Either way tonight I throw in the towel with this servo setup, I cant waist another week without any progress

again thanks for the help, I will look for a mill that needs an upgrade for this hardware, I believe it could be stable in a different configuration