estop-out is not following the E-Stop status

More
3 years 2 weeks ago #282739 by ruffle
estop-out is not following the E-Stop status [SOLVED] was created by ruffle
Elsewhere in this August Forum, someone said that estop-out should automatically follow the E-Stop status... but it doesn't seem to.

I'm running rodw's fine build for bullseye on a RPi 4 which means Linuxcnc 2.10-pre0.

A simple config built with PNCConf produces the following things to do with estop:

Code:
[code]cnc@lathe:~/linuxcnc/configs/Test$ grep -i estop * Test.hal:# --- ESTOP-OUT --- Test.hal:net estop-out  =>     hm2_7i76e.0.7i76.0.0.output-05 Test.hal:# --- ESTOP-EXT --- Test.hal:net estop-ext     <=  hm2_7i76e.0.7i76.0.0.input-15 Test.hal:#  ---estop signals--- Test.hal:net estop-out     <=  iocontrol.0.user-enable-out Test.hal:net estop-ext     =>  iocontrol.0.emc-enable-in cnc@lathe:~/linuxcnc/configs/Test$ Triggering the external E-Stop button shows ESTOP on Axis but the state of estop-out doesn't change. Attached images showing without and with the External EStop button triggered but Halwatch shows no change to estop-out. Is this a PNCConf thing? a 2.10.-pre0 thing or am I totally misunderstanding the function of estop-out? [attachment=56488]Screenshot_2023-10-11_12-36-37.png[/attachment][attachment=56489]Screenshot_2023-10-11_12-37-09.png[/attachment] Thanks.
[/code]

Please Log in to join the conversation.

More
3 years 2 weeks ago #282754 by chris@cnc
Replied by chris@cnc on topic estop-out is not following the E-Stop status
You have latency problems and in this case, the EMG status does not change.
Run this and check where the problem is.
Code:
​​​​​​​latency-histogram --nobase --sbinsize 1000 halcmd show param hm2*tmax* sudo chrt 99 ping -i .001 -q [your_card]

Please Log in to join the conversation.

More
3 years 2 weeks ago #282757 by ruffle
Replied by ruffle on topic estop-out is not following the E-Stop status
Ah.... maybe that didn't help but...

I'd used Pncconf to generate a simple config and took the defaults including a servo period of 1000000.

Changing this to 2000000 (as suggested for an RPi 4 by PCW) gets rid of the warning but estop-out still doesn't change from FALSE no matter what the E-Stop status actually is.

Please Log in to join the conversation.

More
3 years 2 weeks ago #282758 by Aciera
Replied by Aciera on topic estop-out is not following the E-Stop status
could you please attach your hal and ini files?

[edit]
Also maybe try again with this line commented out in your hal file:

net estop-out  =>     hm2_7i76e.0.7i76.0.0.output-05

Please Log in to join the conversation.

More
3 years 2 weeks ago #282763 by ruffle
Replied by ruffle on topic estop-out is not following the E-Stop status
No difference with the

net estop-out  =>     hm2_7i76e.0.7i76.0.0.output-05

line commented out.

hal and ini files attached.

[edit]
FYI, nothing in the other hal files. As I said it was a simple config generated by Pncconf.
Code:
cnc@lathe:~/linuxcnc/configs/Test$ ls -l total 44 -rw-r--r-- 1 cnc cnc 6630 Oct 11 18:33 Test.hal -rw-r--r-- 1 cnc cnc 3233 Oct 11 16:41 Test.ini drwxr-xr-x 2 cnc cnc 4096 Oct 11 12:02 backups -rw-r--r-- 1 cnc cnc  103 Oct 11 12:02 custom.hal -rw-r--r-- 1 cnc cnc  108 Oct 11 12:02 custom_gvcp.hal -rw-r--r-- 1 cnc cnc  111 Oct 11 12:02 custom_postgui.hal -rw-r--r-- 1 cnc cnc 1666 Oct 11 18:32 linuxcnc.var -rw-r--r-- 1 cnc cnc 1666 Oct 11 18:32 linuxcnc.var.bak -rw-r--r-- 1 cnc cnc  105 Oct 11 12:02 shutdown.hal -rw-r--r-- 1 cnc cnc  209 Oct 11 12:02 tool.tbl cnc@lathe:~/linuxcnc/configs/Test$ 

Please Log in to join the conversation.

More
3 years 2 weeks ago #282792 by phillc54
Replied by phillc54 on topic estop-out is not following the E-Stop status
iocontrol.0.user-enable-out does not follow the estop state, it is commonly used as an output from GUI's for estop.

github.com/LinuxCNC/linuxcnc/blob/2.9/sr...ask/ioControl.cc#L14

If you are looking to follow the estop state then you could use hauli.estop.is-activated.
 

Please Log in to join the conversation.

More
3 years 2 weeks ago #282805 by ruffle
Replied by ruffle on topic estop-out is not following the E-Stop status
Thanks for that pointer (wouldn't have found it in a million years!)

hauli.estop.is-activated does indeed follow the E-Stop state as you say.

However, I can't get it to turn the output on my 7i76E on :-(
Code:
net hauli.estop.is-activated => hm2_7i76e.0.7i76.0.0.output-05

doesn't actually change output 5 on when hauli.estop.is-activated changes. I'm using the same constructs for other 7i76E outputs (net pause-on => hm2_7i76e.0.7i76.0.0.output-06) and they work.

I'm missing something obvious but I'm in danger of wearing out Mr Google trying to find it :)

Please Log in to join the conversation.

More
3 years 2 weeks ago #282806 by phillc54
Replied by phillc54 on topic estop-out is not following the E-Stop status

Code:
net hauli.estop.is-activated => hm2_7i76e.0.7i76.0.0.output-05

You forgot a name for the net:
Code:
net estop-active hauli.estop.is-activated => hm2_7i76e.0.7i76.0.0.output-05

Please Log in to join the conversation.

More
3 years 2 weeks ago #282815 by ruffle
Replied by ruffle on topic estop-out is not following the E-Stop status
I think that was one of the permutations I tried while flailing around but it results in an error:

Putting this in Test.hal:

net estop-active hauli.estop.is-activated => hm2_7i76e.0.7i76.0.0.output-05

Linuxcnc errors out with:
Code:
./Test.hal:28: Pin 'hauli.estop.is-activated' does not exist

Please Log in to join the conversation.

More
3 years 2 weeks ago #282816 by tommylight
Replied by tommylight on topic estop-out is not following the E-Stop status

net estop-active hauli.estop.is-activated => hm2_7i76e.0.7i76.0.0.output-05
Typo, should be "halui"
Code:
net estop-active halui.estop.is-activated => hm2_7i76e.0.7i76.0.0.output-05

Please Log in to join the conversation.

Time to create page: 0.097 seconds
Powered by Kunena Forum