#hazzy | Logs for 2020-07-12
Back
[00:25:54] <Joco[m]> hazzy: would seem to be a much slicker implementation. Worth pursuing I think
[00:39:43] <Joco[m]> Re done my compensation settings on X axis. Getting the process pretty smooth now with some macros in libre calc to help move data about. Results looking good. There are error graphs on the commanded v's actual position. Measure going from 0 to 480m and from 480mm to 0.
[00:39:53] * Joco[m] uploaded an image: x_test_neg.png (20KB) < https://matrix.org >
[00:39:55] * Joco[m] uploaded an image: x_test_pos.png (20KB) < https://matrix.org >
[00:46:14] <Joco[m]> It's looking a heck of light better than it was when I started out.
[00:56:41] <hazzy-m> Joco: that looks very good indeed
[00:57:49] <hazzy-m> 1am, time for sleep
[00:58:24] <Joco[m]> nite
[01:30:57] <Joco[m]> * It's looking a heck of lot better than it was when I started out.
[01:36:34] -!- c-log has quit [Ping timeout: 240 seconds]
[01:41:36] -!- c-log has joined #hazzy
[02:36:38] -!- c-log has quit [Ping timeout: 258 seconds]
[02:40:41] -!- c-log has joined #hazzy
[02:51:46] -!- mozmck has quit [Ping timeout: 260 seconds]
[04:22:35] -!- theresa[m] has joined #hazzy
[04:44:38] -!- c-log has quit [Ping timeout: 272 seconds]
[04:46:58] -!- c-log has joined #hazzy
[06:08:02] <JT-Cave> morning
[10:02:15] <hazzy-m> Morning
[11:29:09] <hands[m]> joco did you run it at the same speed for your tests? you may want to run at different speeds so you can tell if the issue is related to RF/noise or are related to the ball screw.. Especially how consistent the data is.. (just finished a personal project where I was measuring forces and noise/hz coming out of the wall and through the DC power supply was causing issues with measurement.. given the sizes you are measuring it
[11:29:09] <hands[m]> it might be worth making sure time isn't a component of the .04mm movement noise).. By the way .04mm is amazing error rate!!!!
[11:40:03] <roguish[m]> morning.............
[11:53:50] <JT-Cave> yo
[11:57:55] <hazzy-m> JT-Cave: did you see I worked on your plain text edit?
[11:58:22] <hazzy-m> I didn't realize it was so close to being finished, fantastic work
[12:00:10] * JT-Cave has been neck deep in chickens lol
[12:00:44] <JT-Cave> glad you got it done, I'll be glad when all I have to do is feed, water and pick up poop
[12:17:20] <hazzy-m> JT-Cave: How are the automation projects coming?
[12:22:42] <JT-Cave> slow, finally got all my components in the other day... just need to put it together
[12:23:24] <JT-Cave> Rusty the truck tied me up for a week trying to sort out why the rear calipers were dragging after replacing the pads and van has been in the shop for 2 weeks
[12:23:34] -!- c-log has quit [Ping timeout: 240 seconds]
[12:23:45] <JT-Cave> well luxury weekend nap time here
[12:25:07] -!- c-log has joined #hazzy
[13:09:31] <JT-Cave> well that didn't last long
[13:09:50] * JT-Cave thinks he finally got screen blanking turned off in X
[14:55:07] <Joco[m]> hands: good question re speed. I have been running at 150mm/min feed rate. As i needed to ensure was not over stressing the mesa encoder that is really designed for MPG. But running at say 100 or 200 should be ok for a test. A big variable in this homing consistency. I have run a few checks with rehoming as a key element. The error band does move a little but is still pretty tight. I.e holding the bulk of error points within
[14:55:07] <Joco[m]> the 0.02mm band. After work I’ll try the different speed tests.
[14:58:32] <Joco[m]> <hazzy-m "I didn't realize it was so close"> I didn’t know it existed. Like a pure QT implementation. Scintilla is sooo finiky.
[15:04:01] <Joco[m]> Quick mesa question: given I think there is some good mesa knowledge here ... what card would be best/most appropriate to add h/w high speed encoders to a 7i76e?
[15:08:44] <Joco[m]> * Quick mesa question: given I think there is some good mesa knowledge here ... what card would be best/most appropriate to add h/w high speed encoders to a 7i76e? Im thinking a 7i85 but someone was poopooing the db25 connects being used.
[15:10:46] -!- c-log has quit [Ping timeout: 256 seconds]
[15:14:40] -!- c-log has joined #hazzy
[15:24:34] <hands[m]> Joco, I was just talking to Peter at Mesa about this, he suggested the 7i89 (but I also need a toon of encoders)..
[15:29:18] * hands[m] uploaded an image: unknown.png (11KB) < https://matrix.org >
[15:29:22] <hands[m]> Before I go down a rabbit hole of refactoring, is there a good reason NOT to break the large reusable components (maybe even whole tabs) into self-encapsulated widget using .ui files?
[15:29:22] <hands[m]> for instance something like:
[15:30:03] <hands[m]> I did a quick test with a single control with the jog rate, and it seems to be getting the updates just fine even though the slider was in the main form..
[15:39:25] <Joco[m]> You mean completely encapsulated UIs sub parts? All laid out and ready to drop into place?
[15:47:32] <hands[m]> yes
[15:47:49] <hands[m]> DRO, Jog, Offsets, etc..
[15:48:56] <hands[m]> would make feature creation easier as every template could "consume" the same components.. composing as needed... (within reason)
[15:54:03] <Joco[m]> hmmm ... conceptually I can't see why not. But in all honesty if anyone is likely to see a "why not" it will be the likes of hazzy or JT-Cave
[15:54:20] <Joco[m]> TurBoss: as well
[15:55:00] <Joco[m]> I've spent the last 20 years managing technical delivery teams. I've been away from the tools too long
[15:55:22] <hands[m]> yah it was a general question... I was answering your question about encoders, then I switched topics 😛
[15:55:37] <Joco[m]> lol
[15:55:48] <Joco[m]> I saw your comment re the 8 encoder board
[15:57:44] <Joco[m]> I think on the kit I have I will probably stick with stepper drivers but adding a linear encoder as a fine adjustment is what I am looking into . HOWEVER, having the 8 encoders would allow me to move to either servos or encoder backed steppers in the future.
[15:58:01] <Joco[m]> while still keeping use of the scales.
[15:58:08] <Joco[m]> just for giggles. lol
[15:58:26] <Joco[m]> I think I have said this before: the mill IS the hobby
[15:58:56] <hands[m]> best machines in the world (that aren't linear servos) are glass scales + servos with encoders..
[15:59:04] <Joco[m]> O.o - about to hit a tunnel. Will go dark for about 8-10mins
[16:32:26] <Joco[m]> Monday morning. Work. Blah. Got to win the lottery.
[16:35:42] * JT[m] uploaded an image: VectorImage_2020-07-12_031518.jpg (5455KB) < https://matrix.org >
[16:35:44] <JT[m]> Finally got the run enclosed
[16:37:48] <Joco[m]> Nice. Always a good feeling to get something done.
[18:02:53] <roguish[m]> JT: will it keep all the hungry critters out?
[18:03:13] <roguish[m]> anyway to electrify?
[18:03:24] <roguish[m]> old car coil?
[18:03:40] <roguish[m]> when the chicks are inside, of course.
[18:03:48] <roguish[m]> inside the house
[19:01:08] <JT-Cave> there will be some very HOT wires on the outside lol
[22:55:58] * hazzy-m uploaded an image: image.png (121KB) < https://matrix.org >
[23:16:50] <Joco[m]> looking good. How are you showing what line is being executed?
[23:17:54] <hands[m]> Nice!
[23:30:34] * hazzy-m uploaded a video: gcode-edit.mp4 (4432KB) < https://matrix.org >
[23:30:46] <hazzy-m> just highlighting the current line for now