Index Homing odd offset

OT-CNC wrote: I know it's off topic, but while you are looking at my hal file, on the wish list is a limit component to ignore the jog increments above .001" so my mpg only uses the jog increments that are selected below .001" for movement.

Why not just remove those increments? They are in the INI [DISPLAY] section

Also, The spindle has not been implemented. I probably will need help with that as well.

Have a look here for a complete spindle HAL file, including spindle orient:
git.linuxcnc.org/gitweb?p=linuxcnc.git;a...756e24039816;hb=HEAD
Not 100% plug-and-play as the loadrt pid.... lines need to all be in the first HAL file loaded.

PCW,

Thanks for the input.

What is the range for P? 0-100 for closed loop? 1000 only for open loop?

I'm trying the servo calibration screen and don't see a distinct change.
Below around 75ipm I don't get the following errors, above that its not immediate but after a few inches of travel. Sometimes it happens when I release the jog key, usually not on initial accel.
Do I have all the right components in the hal file?? Some stuff is # out.
Also, in the ini, what are safe max acceleration settings that should work with 5i25? Stepgen_Maxaccel?
I also changed the pid max error to .005 in the hal, what about the ferror and min_ferror in the ini? I have them set at .5 and .005 now.

The range of P depends on how well and how fast the drive follows the step commands so is only guessable now
(thats why you adjust it upwards until it oscillates and then back off)

Did you change these parameters?

MAX_VELOCITY = 5.0
(5 IPS = 300 IPM)
STEPGEN_MAXVEL = 6.0
(25% headroom)

if not, a following error would be _expected_ at just about 75 IPM
since the stepgen_maxvel was set for 1.25 IPS = 75 IPM

Stepgen_Maxaccel should normally be set for 25% more than the machines acceleration setting
(which depends on what the loaded drives are capable of)

The 5I25 has no acceleration limits, it the mechanics/drive that do


When you use the calibration screen, you must press the test button to use a new value that you have typed in

If you still have issues you will probably have to plot the following error and velocity with HALScope to determine whats going wrong

andypugh,

Thanks for the spindle file. I will probably start a new topic if I get stuck. That will be later...

All the step increments I want to keep on the axis gui for keyboard arrow jogging. My mpgs use the same selected increments but I want to eliminate only the ones above .001 yet keep them for the arrow jogging. I will dig up the topic from a while back and continue there as I don't want to get off topic.

www.linuxcnc.org/index.php/english/forum...rements?limitstart=0


Anything obvious look wrong in the hal or ini to address the homing issue?

Thanks again for all of your time, much appreciated.

PCW wrote: Did you change these parameters?

MAX_VELOCITY = 5.0
(5 IPS = 300 IPM)
STEPGEN_MAXVEL = 6.0
(25% headroom)

if not, a following error would be _expected_ at just about 75 IPM
since the stepgen_maxvel was set for 1.25 IPS = 75 IPM

Yes, I changed them.

Okay, so the P parameter is not 0-100%. I will try again.

At what feed should I test the increase in P? For now it's 60 ipm, I can go as far as 10 to 1000 on the P, below P 10 the motor sounds off.
P at 100 and 90ipm, I get joint following error with violent oscillations. This is on the x with no load. I don't get the oscillations on the Z with heavy slide and same settings.

Come to think of it, the DMM's have their own internal pid that should get adjusted to the load of the axis first. I need to look at that via their software.

OT-CNC wrote: P at 100 and 90ipm, I get joint following error with violent oscillations. This is on the x with no load. I don't get the oscillations on the Z with heavy slide and same settings.

The PID settings depend on the mechanics of the system, so the settings will be different for a heavy axis than a light one.
(In principle you ought to re-check the PID with a heavy workpiece....)

It sounds like P=100 is too high for X, though you might be able to kill the oscillations with some D. (typically D is a very small number)

I would expect the controller to mainly be using FF1 and [strike]I gain.[/strike]. Sorry, for a velocity drive like you have, it will be Pgain.

I changed a value in the ini file and don't have the joint following errors any more.
I changed the MAX_OUTPUT=1.25 to MAX_OUTPUT=0 (This is the setting between deadband and encoder_scale under axis_2.)
Is 0 an acceptable value for the way I'm controlling my drives? I'm now up at 300 ipm with no following errors. I left the P at 50 for now.
Still need to spend time with halscope to see what's going on.

OT-CNC wrote: I changed a value in the ini file and don't have the joint following errors any more.
I changed the MAX_OUTPUT=1.25 to MAX_OUTPUT=0 (This is the setting between deadband and encoder_scale under axis_2.)

That sounds very unlikely to be correct, but it is hard to say without knowing what that parameter is being used for in the HAL file.

So, what is that parameter being used for in the HAL file?

At a first guess you are turning your PID loop completely off...

Actually that's OK.
A max_limit of 0 means the output is not limited at all
a max_limit of 1.25 means a 75 IPM limit on PID output (velocity)