#hazzy Logs
Jul 14 2020
#hazzy Calendar
02:04 AM Joco[m]: Bit of a technical question on mesa MPG. Does the MPG just read the single rising edge of the AB input? Have noticed the 5mu glass scales are giving a 0.02mm resolution So it is looking very much like only part of the AB TTL signals are being read. Still super useful for what I am doing as a test. But just wanted to check if I was missing a trick.
02:04 AM Joco[m]: * Bit of a technical question on mesa MPG. Does the MPG just read the single rising edge of the AB input? Have noticed the 5mu glass scales are giving a 0.02mm resolution So it is looking very much like only part of the AB TTL signals are being read. Still super useful for what I am doing as a test. But just wanted to check if I was missing a trick. I think once I get to a proper h/w encoder this will go away.
02:07 AM TurBoss: my encoders sends 4 pulses per step
02:07 AM TurBoss: sek I'll check config
02:08 AM Joco[m]: TurBoss: but those will be h/w encoders? I am using (temporarily) the mode2 MPG encoders on the 7i76
02:08 AM TurBoss: yup same
02:08 AM TurBoss: 7i76 mode 2
02:08 AM TurBoss: brb shop
02:08 AM Joco[m]: Rgr
02:08 AM Joco[m]: Cheers
02:12 AM TurBoss: i just use enc0.count pin
02:12 AM Joco[m]: Am watching the count pin and its only ticking over on 0.02mm increments
02:14 AM Joco[m]: And I know that a signal is going to both A & B legs as it knows when I change direction.
02:14 AM TurBoss: what kind of encoder wiring a+b+a-b- or just ab
02:15 AM Joco[m]: Given is mode 2 there is inly AB. The wiring on the glass scale is also just AB
02:15 AM Joco[m]: * Given is mode 2 there is only AB. The wiring on the glass scale is also just AB
02:16 AM Joco[m]: * Given is mode 2 there is only A+ B+. The wiring on the glass scale is also just AB
02:16 AM TurBoss: sorry duno
02:16 AM Joco[m]: That’s ok
02:18 AM Joco[m]: I would say its not registering the drop in the TTL. I.e. to get the right res I think it should be registering the rising and falling edges of the TTL
02:18 AM TurBoss: use halscope
02:18 AM Joco[m]: Which would be your 4 pulses
02:18 AM TurBoss: hmmmm not sure if posible to attach to ab
02:19 AM Joco[m]: Yeah - the encoder is fronting things
02:19 AM Joco[m]: AB not visible I think. But will have a good snoop
02:20 AM Joco[m]: It’s not the end of the world as I will be getting a proper encoder add on card for this.
02:21 AM TurBoss: not visible
02:21 AM Joco[m]: AB not visible? As I suspected
02:21 AM TurBoss: yes
02:21 AM Joco[m]: All good. Thanks for checking
02:22 AM * TurBoss has to engrave a little plate
02:22 AM TurBoss: but begin lazzy
02:22 AM Joco[m]: Lol
02:22 AM Joco[m]: I’m about to go watch TV
02:22 AM * TurBoss uploaded an image: VectorImage_2020-07-14_091306.jpg (323KB) < https://matrix.org/_matrix/media/r0/download/jauriarts.org/vBrDbcODlAsSGbTQyJCIbndi >
02:22 AM Joco[m]: So being very lazy
02:23 AM Joco[m]: Like the X+ and Z- on the tape. :-)
02:23 AM TurBoss: yup no posible fail lol
04:51 AM JT-Cave: morning
09:39 AM hazzy-m: Morning
09:39 AM * hazzy-m uploaded an image: ima_fd44043.jpeg (138KB) < https://matrix.org/_matrix/media/r0/download/matrix.org/AcRcpkIFxNWvTFRTSCTwqPqq >
09:40 AM hazzy-m: People keep giving me junk lol
10:29 AM hr_ram: Hi, guys can anyone point me in the right direction to turn off the machine table size display in the probe basic plot. I'm having trouble viewing the actual tool movement when zoomed in on the plot. Graduated scale just covers the whole thing.
11:01 AM hr_ram: I'm new to linux any help is really appreciated.
11:18 AM JT-Cave: you can turn them on/off in the ini file
11:20 AM JT-Cave: damn he didn't stay 2 minutes lol
12:21 PM TurBoss: I think that the plot needs better defaults
12:55 PM JT-Shop: I agree
02:53 PM Joco[m]: hazzy: I could do with junk. I end up having to buy all my steel and parts.
03:44 PM hazzy-m: Joco: if shipping weren't so experience you could have all the 1000mm x 500mm x 140mm NEMA enclosures you wanted
03:44 PM * hazzy-m uploaded an image: ima_51ed662.jpeg (179KB) < https://matrix.org/_matrix/media/r0/download/matrix.org/CkgNeyUXpPSEpAZxBKgSGXZH >
04:08 PM Joco[m]: Oh dear. That’s not fair.
04:11 PM Joco[m]: I’m about to get a sheet of 2mm aluminium and fab a custom control box to go on the side of the mill stand. Will be welded. Don’t have a brake.
04:16 PM hands[m]: oh nice.. I could use a few ;-)...
04:16 PM hands[m]: (jk)
04:29 PM * Joco[m] cry’s in his coffee.
04:29 PM Joco[m]: Sometimes being at the butt end of the world is a real pain
04:36 PM roguish[m]: hazzy: how much to CA
06:32 PM hazzy-m: roguish: I can check, could probably fit 8 on a pallet
06:33 PM hazzy-m: Or since FABTECH is in Vegas this year, drive a truck down and you can have the lot of NEMA boxes and the other small enclosures will be running
06:34 PM hazzy-m: I hate to just throw them away
06:36 PM TurBoss: hello
06:52 PM hazzy-m: TurBoss: hey
07:56 PM * TurBoss uploaded an image: 20200715_024650.jpg (255KB) < https://matrix.org/_matrix/media/r0/download/jauriarts.org/YcLqAquANtzdsyhYZxcQSsYZ >
09:06 PM Joco[m]: That looks very old school
09:24 PM TurBoss: a sega mega drive
09:24 PM TurBoss: or sega genesys
09:25 PM TurBoss: just created a branch with new 3dPlotter defaults
09:25 PM TurBoss: without labesl and ticks
09:25 PM TurBoss: * without labels and ticks
09:26 PM TurBoss: from this
09:26 PM * TurBoss uploaded an image: image.png (323KB) < https://matrix.org/_matrix/media/r0/download/jauriarts.org/NFwnavwSClYHaOnsCcRKJWvO >
09:28 PM * TurBoss uploaded an image: Captura de pantalla de 2020-07-15 04-18-24.png (21KB) < https://matrix.org/_matrix/media/r0/download/jauriarts.org/WClitItNtzCrqoqaisCUgAuM >
09:30 PM hr_ram: TurBoss Thanks. I think you got my question correctly. I had to go away yesterday. This is exactly What I needed. Please kindly keep in mind that I'm a newbie to linux. I have retrofitted a large size okuma cnc mill. larger table size causes all the issues with the plot.
09:30 PM TurBoss: Hello hr_ram:
09:30 PM TurBoss: do you use metric ? what sizes?
09:31 PM hr_ram: can you kindly point me how to make the plot look like that
09:31 PM hr_ram: i'm using metric
09:31 PM TurBoss: yes
09:32 PM * TurBoss uploaded an image: Captura de pantalla de 2020-07-15 04-22-44.png (13KB) < https://matrix.org/_matrix/media/r0/download/jauriarts.org/MctWIuCbZqzGEGSiqMjCvAzm >
09:33 PM TurBoss: hr_ram: just add a section to your ini called [VTK] and set true or false as you like
09:33 PM TurBoss: new release will have that by default
09:33 PM hr_ram: Huge Thanks.
09:35 PM TurBoss: hr_ram: what dimensions are yours?
09:35 PM hr_ram: If you guys need mill for testing or some feedback I'm really happy to do it. Probe basic was one of the reason why I jumped to linux rather than centroid and controllers like ADTECH.
09:36 PM TurBoss: cool thanks!
09:36 PM hr_ram: My travels are 850mm on X 500mm on Y and 400mm on Z
09:36 PM TurBoss: and does wired things?
09:36 PM TurBoss: * and does wierd things?
09:37 PM hr_ram: It's a really heavy mill. Has box ways on all axis and a geared head
09:38 PM hr_ram: You mean the plot right? wired mean all the fonts cover the plot. they don't scale down when I zoom in. I think thats normal
09:39 PM TurBoss: yes yes
09:39 PM TurBoss: ok thanks!
09:39 PM TurBoss: I tried to reduce font size but i wasn't able so hide them is the only option for now
09:40 PM hr_ram: That's totally ok. By the way my plot was somewhat slower than the plot on axis gui.
09:40 PM hr_ram: I'll turn off the scales and check today
09:41 PM TurBoss: great!
09:41 PM TurBoss: my onboard GPU is slow
09:41 PM TurBoss: it lags
09:42 PM TurBoss: but on my desktop it works very nice
09:42 PM hands[m]: hr_ram.. I'm in the process of dumping ADTECH too..
09:43 PM TurBoss: I don't know ADTECH does have something interesting?
09:43 PM hr_ram: I'm using a office pc. I might need to upgrade that. That's no issue.
09:44 PM hands[m]: If by "interesting" you mean a 1/2 ass implementation of Macro_B and a firmware that gets worse everytime you ask for an upgrade, and it taking 3 weeks and multiple videos to show that it has a bug.. then, yah, I guess it has something..
09:44 PM TurBoss: on my machine pc i have to stay with the onboard gpu
09:45 PM TurBoss: wow
09:45 PM TurBoss: interesting
09:45 PM hr_ram: ADTECH has a controller bit similar to Generic Fanuc at a very low cost.
09:45 PM TurBoss: gona see some vids
09:45 PM hr_ram: It's routine practice to retrofit with ADTECH or GSK controllers on where I live.
09:46 PM hands[m]: and very low quality... (4650 and 9960 both couldn't handle writing proper probing routines without crapping out.. then when they fixed the issues with variable assignments in the MacroB, they broke the spindle control...)
09:46 PM hr_ram: But I Never saw the point with those retrofits.
09:46 PM TurBoss: ops
09:47 PM TurBoss: looks like a PLC HMI
09:47 PM hr_ram: For the Exact reasons hands[m] describe
09:47 PM hr_ram: All those retrofitters do is get the cheapest controller and sell it.
09:48 PM hands[m]: yah... hence the rebuild to LinuxCNC for me.. Luckly this is just a Hobby 🙂
09:50 PM TurBoss: mine is also hobby
09:50 PM TurBoss: little self made engraver
09:50 PM hands[m]: <span class="d-mention d-user">TurBoss</span> I did a bit of convention based programing in the __init__.py in the widgets folder.
09:51 PM * hands[m] sent a long message: < https://matrix.org/_matrix/media/r0/download/matrix.org/cCqqgmYKiQwwlIELlZmtMFYB >
09:51 PM TurBoss: ops the message los ident
09:51 PM hands[m]: as long as the folder name for the widget matchs the widget's .py name and the name of the widget class, this will "discover" the widget..
09:51 PM TurBoss: amazing
09:51 PM hands[m]: oh that is dumb.. let me push and I'll send a link to it on git
09:51 PM TurBoss: great
09:51 PM hr_ram: For me this machine is going to do some production run.
09:52 PM hr_ram: heavy metal cutting. I guess this will be a real test.
09:53 PM hands[m]: https://github.com/jhandel/probe_basic/blob/vertical_ui/widgets/__init__.py
09:53 PM TurBoss: got it
09:53 PM hands[m]: I figured if I was going to be breaking down these components I wanted something that was less manual..
09:54 PM hands[m]: I should probably "compile" a skin and make sure it all works as a compiled skin and not just in Dev.... Time to read more docs on how to do that and set up that test..
09:55 PM TurBoss: this init method can be applied to QyPyVCP also
09:55 PM TurBoss: is the same mechanism
09:56 PM hands[m]: <span class="d-mention d-user">TurBoss</span> I am still treating QyPyVCP as the big scary engine... Baby steps 😉
09:57 PM TurBoss: :)
09:57 PM hands[m]: but yes, if that is how it manages all of its widgets, totally agree we could move this code there for better reuse.
09:57 PM hands[m]: I just literally haven't looked under its hood at all..
09:57 PM TurBoss: I'm seraching for the relevanc parts
09:57 PM TurBoss: yes thats okay
09:58 PM TurBoss: any way i don't want to scare you with all thoose code
09:58 PM hands[m]: (yet)
09:58 PM TurBoss: https://github.com/kcjengr/qtpyvcp/tree/master/qtpyvcp/widgets
09:59 PM TurBoss: bo
09:59 PM TurBoss: :D
09:59 PM TurBoss: I mostly know about widgets and some plugins...
09:59 PM TurBoss: but the rest is hazzy's domain lol
10:00 PM TurBoss: but... seriously what amount of work! amazing!
10:02 PM hazzy-m: hands: looks good at first blush, I'll test tomorrow when I'm awake
10:02 PM TurBoss: :)
10:04 PM TurBoss: I love the pluging factory
10:04 PM hands[m]: cool thanks @hazzy, looks like you are doing "similar" things in the widgets area of the engine.. If you have a pattern you like better or more, I am open to adjusting... I know I am coming in a bit like a wrecking ball in probe_basic working on breaking apart larger pieces and such...
10:05 PM hands[m]: Time to spend some quality time with my lovely wife.. night guys... I'll probably start cutting things apart tomorrow evening (or taking in any feedback that shows up here :-).. )
10:05 PM hazzy-m: hands: no, what you're doing is exactly what I would like to do if I had more time! Really appreciate it
10:06 PM TurBoss: just reading
10:07 PM hazzy-m: hands: same here (except for the lovely wife). night
10:09 PM hr_ram: guys do you know how much I can trust the "run from here"
10:09 PM hr_ram: The firs time I tried I just started at the top of the program
10:09 PM TurBoss: not much for now
10:10 PM hr_ram: I think its better to avoid it right
10:10 PM TurBoss: we are working on a new text editor wich may do it better
10:10 PM hr_ram: If this machine crashes it not going to be pretty
10:12 PM hr_ram: I have had some crashes on my smaller machine before with mach 3 starting from a tool change position. Almost had a similar indecent with this one few days ago.
10:12 PM TurBoss: uff
10:12 PM TurBoss: thats scary
10:13 PM hr_ram: DTG saved the day.
10:14 PM hr_ram: if I was in plain axis gui it might not be the case. having distance to go is really helpful.
10:18 PM hr_ram: Thanks all for the help. I'm leaving for some work.
10:19 PM TurBoss: cu