New Plasma build question

PCW,
Thank you very much, that explains a lot.
So far i have used FF2 only on servo systems for that exact reason.

Thanks again Peter, very helpful.

I am away now till Friday. Could you point me to any spots I need to check, ie would the steps per on the drives cause this. Or the polarity of the dir and pulse settings I am using differential wiring from the 7i96 (I don't think I need to change jumpers for this)

I will start again on my return

Clive s wrote: Ps. Hope to see you Monday ish.

Do give me a call when you land.

The problem is still likely to be bad values the ini file.
Nothing external like wiring would cause a following error.

I used your hal file but had to fix Joint1s stepgen
acceleration in the .ini file as it was the same as the joint
acceleration and also had to set all P terms to 1000.
That fixed all following error issues.

I then changed all joint and axis velocitys to 200, all stepgen
max_vels to 250, all joint and axis accelerations
to 2000 (~ 1/5 G) and all stepgen max_accels to 2500, and this works
fine also. The HalScope plots are from this configuration.

PCW wrote: The problem is still likely to be bad values the ini file.
Nothing external like wiring would cause a following error.

I used your hal file but had to fix Joint1s stepgen
acceleration in the .ini file as it was the same as the joint
acceleration and also had to set all P terms to 1000.
That fixed all following error issues.

I then changed all joint and axis velocitys to 200, all stepgen
max_vels to 250, all joint and axis accelerations
to 2000 (~ 1/5 G) and all stepgen max_accels to 2500, and this works
fine also. The HalScope plots are from this configuration.

Ok had a go at settings as you suggested. Can we just concentrate on the X axis first

The machine homes ok on all axis but I still get a following error on joint_0 if I try and increase the MAX_ACC to 2000 (it is ok at 500)

This is with a nema 34 on a 3:1 so about 1 rev of the motor is about 75mm travel about 9 mtr/min

It errors when I try and do a reverse of the motor at speed.

[AXIS_X]
MIN_LIMIT = -1
MAX_LIMIT = 1300

MAX_VELOCITY = 150
MAX_ACCELERATION = 1000


[JOINT_0]
AXIS = X
MIN_LIMIT = -1
MAX_LIMIT = 1300

MAX_VELOCITY = 150
MAX_ACCELERATION = 1000

TYPE = LINEAR
SCALE = 102


STEPGEN_MAX_VEL = 250
STEPGEN_MAX_ACC = 2500

FERROR = 1
MIN_FERROR = 0.25
DIRSETUP = 5000
DIRHOLD = 5000
STEPLEN = 2500
STEPSPACE = 2500
DEADBAND = 0
P = 1000
I = 0
D = 0
FF0 = 0
FF1 = 1
FF2 = 0.000127
BIAS = 0
MAX_OUTPUT = 0
MAX_ERROR = 0

Can you post the complete hal and ini files (with no changes) that have the following error

PCW wrote: Can you post the complete hal and ini files (with no changes) that have the following error

Thanks.

I don't see anything obvious wrong (and I allready ran your config here with 2000 Accel)

I suspect you will have to plot the commanded joint position and ferror with halscope
to see what going on

Thanks, I have never used halscope

Is this any use

I don't seem to be able to make it trigger sometimes.

Do I trigger it with " joint.0.f-error "

Edit: could you please supply settings for metric halscope ie. timings etc.

That shows something very basic wrong with the feedback
can you post the results of

uname -a

and

halcmd show all *tmax*

( run the last command when LinuxCNC has been running for at least a few minutes )

Thanks