difference between 2.7 and 2.8pre with mpg

The code below works in Lcnc V. 2.7.14 but I can't seem to get it working in 2.8pre.

I see in halmeter all the relevant pins changing state. ie encoder.2---- counts, phase-A, velocity etc.. but it does not jog with the mpg in Z
I have noticed that the pin axis.z.pos-cmd changes with the keyboard jog key but not with the mpg. What I am missing?


# MPG stuff here
#### load ilowpass for mpg smoothing ###

loadrt ilowpass names=ilowpass.mpgX,ilowpass.mpgY,ilowpass.mpgZ

addf ilowpass.mpgZ servo-thread
setp ilowpass.mpgZ.scale 1
setp ilowpass.mpgZ.gain 0.1

setp axis.2.jog-vel-mode 1


#set axis to the required increment ( 0.01)
net jog-inc-z axis.2.jog-scale
sets jog-inc-z 0.01

#encoder Z
net mpg-za encoder.2.phase-A <= hm2_7i92.0.gpio.032.in
net mpg-zb encoder.2.phase-B <= hm2_7i92.0.gpio.033.in

# send mpg-encoder counts to required axis

# Z encoder
net pend-counts-z axis.2.jog-counts
net pend-counts-z ilowpass.mpgZ.in

net encoder-count-z ilowpass.mpgZ.out
net pend-counts-z encoder.2.counts

One missing thing is axis.z.jog-enable
also at least according to my motion man page:
net pend-counts-z axis.2.jog-counts
should be
net pend-counts-z axis.Z.jog-counts
Also the low pass filter is bypassed, not sure if that's just for debugging

PCW wrote:

One missing thing is axis.z.jog-enable
also at least according to my motion man page:
net pend-counts-z axis.2.jog-counts
should be
net pend-counts-z axis.Z.jog-counts
Also the low pass filter is bypassed, not sure if that's just for debugging

Thanks for the help yet again Peter:
#encoder Z
net mpg-za encoder.2.phase-A <= hm2_7i92.0.gpio.025.in
net mpg-zb encoder.2.phase-B <= hm2_7i92.0.gpio.026.in

# Z encoder
net pend-counts-z encoder.2.counts
net pend-counts-z ilowpass.mpgZ.in

#net encoder-count-z ilowpass.mpgZ.out
net pend-z ilowpass.mpgZ.out
net pend-z axis.z.jog-counts

setp axis.z.jog-enable true

I see all the above pins and signals change in halmeter but not reflected in the DRO Z (no movement)

I can't see where I am going wrong. I have cut the ilowpass out for now

Has the system been homed when you are trying to jog?

PCW wrote:

Has the system been homed when you are trying to jog?

Thanks that was it. What fooled me was that it would jog with the keyboard so I expected it to do the same with the mpg.

Now I just need to sort the ilowpass out.

Yes 2.8 is in joint mode before homing (the GUI must know this so jogs via the joint pins with the KB)

PCW wrote:

Yes 2.8 is in joint mode before homing (the GUI must know this so jogs via the joint pins with the KB)

Yes I should have known. If you have the time could you look over the ilowpass for me.

andypugh wrote:

In 2.8 you probably want to connect the mpg to both
axis.L.jog-counts
_and_
joint.N.jog-counts

It appears to work as is on the bench . Do you mean like this:-

# x encoder
net pend-counts-x encoder.0.counts
net pend-counts-x ilowpass.mpgX.in

net pend-x ilowpass.mpgX.out
net pend-x axis.x.jog-counts
net pend-x joint.x.jog-counts

setp axis.x.jog-enable true