Advanced Search

OK that's a good start, all we need is someone to confirm with actual hardware.

I looked through the EL5101 docs and - did I get it right? - it measures frequency by counting number of edge rises during a 10 ms window? Could be where 10 ms comes from. Seems low resolution also, frequency will be in steps of 100 Hz.

Its been quite meany years now but I wonder if anyone has got these handy little tools working on a more recent version on Python3.x ?

Yeh, that looks OK. If Ethercat was going to fail due to unsupported kernel, you would not have got that far...

PCW wrote:

Pretty strange, seems to be a issue with the 2.9.4 binary unless it's something subtle
wrong with the hal file syntax.

Can you post you hal/ini files?

 

2.94 is just a tag in git at a (recent) given point in time. You git checkout v2.94 to get the source. Perhaps there is a dependency missing from the .Deb build instructions.
 

theres 14, 20, 26, 36, 50, 68. 50 and 68 were the only ones commonly used in computers i think.

Hmmmm there is no 26 pin SCSI connector that i can think of, they are always 50 pin IDC style or over 70 pin connectors with pretty damn small raster.
Again, pictures?
Of the connector now, at least.

well, I guess I'm waiting another 2 weeks. but I think I can at least get the servos moving and linuxcnc/ethercat running. And then maybe the rs485 spindle.

Point me to where I need to read about configuring that?

pics later. already have 2 issues.

1: the 200w servos have 11mm shafts. wft is that? I ordered couplings but they wont arrive for 2 weeks.

Second issue is i am having a hard time cinding the CN26P cables for the io from somewhere economical that doesnt take 2 months to ship. (nohting on amazon, digikeys pricing is insane, and the actual cables from omron or yaskawa are both expensive AND take 2 months.

sigh.

its just a scsi connector, youd think every computer store would have some in the back room collecting dust.

opi5 plus has 2x2.5G lan.
This board can run linuxcnc with ethernet mesa card?
Supported by mainstrean kernel?
Has anyone tested it?

Just in terms of naming, something like seem a bit more readable

net home_y2_z hm2_7i96s.0.inm.00.input-03-not
net home_y2_z joint.2.home-sw-in  # assuming joint 2 is Y2
net home_y2_z joint.3.home-sw-in

also of course Z should home first (and I assume there is a Y switch as well so you can use gantry homing)

 

preamble:
i am converting an mr1 to linux cnc via 7i96s. the y axis is parallel stepper motors.  y2 and z home switches are nc type and in series. the box i am making is all new drivers/power supplies only thing i will be reusing out of the factory box is the vfd for the spindle. i want this to be a plug and play so i can swap back to stock box if needed. dealing with code is not completely forign to me but the last time i did anything serious was in msbasic, and a little dabbing into megasquirt engine control.

im not sure the best way to set up the homeing switches. i tried
1. assigning net home-y2     <=  hm2_7i96s.0.inm.00.input-03-not and
    net home-z     <=  hm2_7i96s.0.inm.00.input-03-not
   did not work gave errors

2.checking home is shared on y2 and z and only asigning an input to z.  did not work either

3. tied 2 inputs together and asigned them each as y2 and z
  worked but don't seem right to me

4. in z axis joint3 i put
net home-y2    =>  joint.3.home-sw-in and home is shared

that works but does not seem right.

what is the best way to do this, or is there another way

Each config should have it's own folders, and yes you can delete those, but if you edited the same config do not delete stuff before making a backup or copy the folder somewhere else.

ihavenofish wrote:

Woo servos arrived, among other things.

We don't believe you !
We want proof!

Not much there...