#flex-dev Logs

Nov 19 2025

04:25 AM Tom_L: morning
04:38 AM JT-Chicken: morning
06:11 AM JT-Cave: 48-65f
07:13 AM lcnc-relay: <AllAboutJake@> I think action_estop() in actions.py needs to set the estop checked state. If you use the menu to open/close the estop, the button state doesn't change on the UI.
07:17 AM Tom_L: 41F 65F
07:21 AM lcnc-relay: <AllAboutJake@> or maybe it somewhere in the status update function?
07:40 AM JT-Cave: hmm got a bit of catch 22 going on but time to start my day
08:19 AM JT-Cave: that was interesting
09:09 AM * JT-Shop heads to town to ship a package
11:55 AM Tom_L: PR #412
11:59 AM Tom_L: probe enable button works now with just estop
11:59 AM Tom_L: power off
12:00 PM Tom_L: on touch, not a regular gui
12:03 PM Tom_L: regular gui requires estop, power, home before probe enable button works
12:06 PM JT-Shop: ok
12:27 PM JT-Cave: it's a mystery to me why tp3 does that but I guess it's something to do with either led or probe enable being a hal pin
12:28 PM JT-Cave: it's the hal pin part
12:28 PM * JT-Cave heads to River Ridge
12:45 PM lcnc-relay: <AllAboutJake@> Tom_L: Tom is this in the project’s main repository? I’ll take a look
01:18 PM lcnc-relay: <AllAboutJake@> OK I found it in your repo. So what's happening is just a wierd confuence of things.
01:19 PM lcnc-relay: <AllAboutJake@> 1. tp3 has hal_pin dynamic properties on probing_enable_pb which the normal touch-probe3 does not.
01:19 PM lcnc-relay: <AllAboutJake@> 2. This causes the button to get added to BOTH home_required and state_estop_reset
01:21 PM lcnc-relay: <AllAboutJake@> 3. the state_setop_reset list gets enabled/disabled very early in update, so it gets enabled when estop is reset.
01:24 PM lcnc-relay: <AllAboutJake@> 4. home_enabled only gets evaluated on a mode change, so it never really gets changed back once its enabled by state_estop_reset
02:00 PM Tom_L: i added hal_pin to touch-probe3 probe enable button and it behaves the same as it did previously
02:01 PM Tom_L: wait..
02:05 PM Tom_L: it won't let me create the pin_name
02:05 PM Tom_L: comes back saying 'probez_btn' is already in use but it isn't
02:05 PM Tom_L: comes back saying 'probe_btn' is already in use but it isn't
02:09 PM Tom_L: even though it gets ignored, the behavior changes to like the touch one acts
02:09 PM Tom_L: estop on -> probe enable can be checked
02:09 PM Tom_L: so it does it in either ui file
02:10 PM Tom_L: i don't get that warning in tp3-touch though
02:13 PM Tom_L: so with dynamic properties function:hal_pin and pin_name:touch_btn both .ui behave the same
02:14 PM Tom_L: probe_btn... sry
02:18 PM lcnc-relay: <AllAboutJake@> probing_enable_pb is a special case button, right, I think its not specifically the hal/pin settings, but its asisgning hal stuff to a special case button like probing_enable_pb.
02:24 PM lcnc-relay: <AllAboutJake@> so weird, I get that error two about it being already created, but the string "probe_btn" doesn't exist ANYWHERE in the *.ui file
02:24 PM lcnc-relay: <AllAboutJake@> (other than the probe enable button I just added it to)
02:26 PM Tom_L: if you change the name, the error follows the name too
02:29 PM lcnc-relay: <AllAboutJake@> yeah I think the hal pin is getting created earlier and it adds it to "parent" and then it does it again later.
03:02 PM JT-Mobile: Heading to the barn
04:13 PM lcnc-relay: <Lyn@> the metric probing routines are stopping randomly without any error message for me. the single move probes work, but the center finding routines just stop on the line G90 G1 Z #2021. sometimes it works for x and then stops on y on the bar probe routine
04:21 PM lcnc-relay: <Lyn@> no idea how to even begin troubleshooting this. sometimes it completes the first 2 probes successfully and gets to the second set of probe movements
04:21 PM lcnc-relay: <Lyn@> then stops at the same place
04:23 PM lcnc-relay: <Lyn@> updated with the latest scripts from GitHub, didn't help
04:46 PM lcnc-relay: <Lyn@> it executes the line G90 G1 Y #2023 and then exits the subroutine.
05:16 PM Tom_L: hole or bar?
05:16 PM Tom_L: are you running a parallel port?
05:17 PM Tom_L: lyn...
05:17 PM lcnc-relay: <Lyn@> bar, p_lx_out_m, and the corner probe routines all stop the same way. yes i have a parport breakout card.
05:17 PM Tom_L: that could be the problem and we haven't been able to find the issue yet
05:18 PM Tom_L: i experienced random stops on mine even in inch
05:18 PM Tom_L: i'm running a parallel port mesa card
05:18 PM Tom_L: i reverted to an older flex deb and it works
05:18 PM lcnc-relay: <Lyn@> on subroutines specifically? the machine runs for hours perfectly normally
05:18 PM Tom_L: i'll have to check what version
05:18 PM Tom_L: yes
05:19 PM Tom_L: i've only run into this on the parallel port
05:19 PM Tom_L: and i don't know why
05:19 PM Tom_L: i just walked in.. gimme a few and i'll check what flexgui version i'm running on that pc
05:20 PM lcnc-relay: <Lyn@> very interesting. thanks for the info, you caught me as i was typing up a forum post. i have a mesa card and one of those cheapo ones connected to the pc's parallel port
05:20 PM Tom_L: 7i90 is what i have
05:20 PM Tom_L: i'll go check the version if you give me a couple min
05:26 PM Tom_L: i reverted to 1.2.1 in order to get them to work on the pc with the parallel port
05:27 PM Tom_L: neither me or JT can figure out what broke it yet
05:27 PM Tom_L: https://github.com/jethornton/flexgui/releases
05:28 PM Tom_L: if you would be kind enough to help us, i'd like to verify that works on your setup as well
05:31 PM lcnc-relay: <Lyn@> trying now
05:31 PM Tom_L: they don't always stop in the same spot either
05:34 PM lcnc-relay: <Lyn@> it stopped again on 1.2.1
05:36 PM Tom_L: revert to 1.2.0
05:36 PM Tom_L: i was able to run 1.2.1
05:36 PM Tom_L: what mesa card?
05:38 PM Tom_L: i'm 90% certain it's not the subroutines
05:40 PM lcnc-relay: <Lyn@> it worked once on 1.2.0, the whole bar routine. first time it's worked
05:40 PM lcnc-relay: <Lyn@> 6i25 and 7i85s
05:42 PM lcnc-relay: <Lyn@> need to revert some config changes as the theme isn't loading and the icons aren't working
05:42 PM Tom_L: yeah unfortunately...
05:42 PM lcnc-relay: <Lyn@> qss under display
05:43 PM Tom_L: i run 6-8 configs and it gets hard to keep track of them
05:44 PM lcnc-relay: <Lyn@> yeah it seems to be totally working on 1.2.0 thanks so much
05:44 PM Tom_L: right now that's the best answer i can give
05:45 PM xxcoder: interesting
05:45 PM lcnc-relay: <Lyn@> I have Xbox controller jogging now so no need for keyboard
05:45 PM Tom_L: you might make an issue on his git page with your setup and what version works
05:46 PM Tom_L: i'd like to get to the bottom of it but i don't know if i can
05:48 PM lcnc-relay: <Lyn@> will do
05:53 PM Tom_L: double check all the settings for your setup. i had them set for mine and i generally don't run a metric config
05:54 PM Tom_L: the probe & ops settings
06:26 PM Tom_L: interesting the 6i25 and 7i85s aren't parallel port cards
06:27 PM Tom_L: unless he's also running a parallel port bob
06:31 PM Tom_L: Lyn@, it would be interesting to see if you had an extra pin on one of those cards to put the probe on
06:31 PM Tom_L: then try a newer deb version
06:32 PM Tom_L: i assume you have the probe on the parport BOB
06:33 PM Tom_L: although that shouldn't affect motion and that's where the problem is
06:39 PM lcnc-relay: <Lyn@> i do have it on the parport bob, as well as spindle speed control and limit switches
06:41 PM Tom_L: rethinking, it's not the probe tripping that's the problem, it's something in motion
06:41 PM Tom_L: we looked at it a while back and couldn't figure it out