#hazzy | Logs for 2018-09-18

Back
[02:46:49] <jesseg> hazzy, you around? just got back from a 10+ hour hike in very rugged terrain -- trying to get to a 100+ year old copper mine way up in the mountains: https://drive.google.com
[02:47:11] <jesseg> anyway I'm beat, going to bed earlier than normal :P Night folks
[06:31:23] <jthornton> morning
[10:35:59] <hazzy-m> jesseg: Very nice! What do you use to log your position?
[10:46:44] <jesseg> Good Morning.
[10:46:55] <jesseg> hazzy, Garmin Etrex 20
[10:47:18] <jesseg> but it's not super accurate when it's upside down in my knee pocket LOL.
[10:47:41] <Lcvette> hola from ground zero
[10:48:14] <hazzy-m> Lcvette: Hurray! your still alive!
[10:48:41] <hazzy-m> jesseg: Thanks! Map data looks good
[10:48:41] <Lcvette> yes! hurray! for now
[10:48:56] <hazzy-m> Do you have power yet?
[10:49:04] <Lcvette> river has not yet crested and isn't expected to crest until ZThursday and is already pretty high
[10:49:22] <hazzy-m> that is not good
[10:49:27] <Lcvette> just got power back this morning and internet
[10:49:50] <Lcvette> ubt we are without fuel, food, water or access to the outside world
[10:49:52] <hazzy-m> sweet!
[10:50:11] <hazzy-m> oh :(
[10:50:20] <Lcvette> completely cutoff with all roads being under water and impassable
[10:50:41] <Lcvette> and the looming threat of the slow rising river
[10:51:06] <hazzy-m> so sounds like the situation it not really improving much
[10:51:34] <hazzy-m> do you have enough food and water?
[10:51:49] <Lcvette> well, now we can worry with internet and information and lights
[10:51:49] <hazzy-m> I mean potable water, I know you have plenty of the other kind :D
[10:52:17] <Lcvette> running low on both drinking water and food
[10:53:15] <hazzy-m> Not good
[10:53:35] <hazzy-m> how are the older folks doing? Are they with you?
[10:54:19] <Lcvette> mom and dad are here with me
[10:54:33] <Lcvette> we are better now with power for my dad's oxygen pump
[10:54:40] <Lcvette> he was running low on bottle
[10:55:45] <Lcvette> everyone here is on edge
[10:55:47] <Lcvette> its not much fun
[10:56:05] <hazzy-m> good, I was worried about him running out, especially if you ended up having to do some diving
[10:56:20] <hazzy-m> doesn't sound fun at all :(
[10:56:34] <hazzy-m> Is there anything I can do?
[10:56:42] * hazzy-m wishes he had a helicopter
[10:56:59] <Lcvette> how are things there in GA?
[10:57:19] <hazzy-m> beautiful, and bone dry
[10:57:37] <hazzy-m> din't get any rain from Florence, and we needed it
[10:58:49] <Lcvette> well, be glad thats not the kind of rain you want
[10:59:15] <jschi[m]> lcvette - glad to hear from you - sorry for your burden out there
[10:59:37] <Lcvette> thanks Jeff!
[10:59:49] <Lcvette> its a mess here!
[10:59:56] <jschi[m]> sounds like it
[11:00:49] -!- Roguish has joined #hazzy
[11:01:00] <Lcvette> I drove out to try and find some food and water yesterday before we had any new or internet to know what was open or closed, everything was word of mouth, you should see I-40, it looks like its a river ratgher than an interstate
[11:01:30] <Lcvette> water in both direction from treeline to treeline as far as the eye can see
[11:03:37] <hazzy-m> wow, just looked up pictures or I-40, unbelievable
[11:12:08] <Lcvette> yeah its nuts
[11:12:09] <Lcvette> most roads around here are like that
[11:12:24] <Lcvette> we are blocked in on everywhere
[11:28:38] <Lcvette> hazzy are your updates mergable yet?
[11:32:07] <hazzy-m> getting very very close
[11:32:20] <hazzy-m> I have most of the buttons you needed, and then some
[11:38:18] <jschi[m]> I got my machines up off the shop floor - nothing's running yet but I can almost walk around the shop now, compressor and air lines installed yesterday
[11:40:14] <jschi[m]> Found my box of arduinos and started laying down code for the touch/TFT display to automate my surface grinder - more projects than time in the day
[11:41:45] <hazzy-m> How do you plan to automate the grinder? Is it hydraulic feed or will you add servos?
[11:42:48] <jschi[m]> Thinking steppers ATM but might go clearpath - the surface grinder is the base that Tormach uses for the PSG-612 - same castings etc
[11:43:12] <jschi[m]> just focusing on table travel ATM - z-height left as exercise for later
[11:43:51] <jschi[m]> got a nice little gui lib setup on the touch screen - running on a cortex processor - so plenty of juice
[11:45:31] <hazzy-m> Nice!!
[11:47:11] <hazzy-m> even slight stepper resonance can cause problems with surface finish on a grinder, Tormach had some trouble with that initially
[11:47:30] <hazzy-m> I think they solved it by going to 3ph steppers
[11:47:44] <jschi[m]> ahh, yeah, the cheapskate in me doesn't want to spend as much on servos as I paid for the machine
[11:48:01] <jschi[m]> but .... finish is everything
[11:48:26] <jschi[m]> the joy of twirling handles vanished quickly
[11:49:21] <hazzy-m> 3ph steppers are quite affordable these days, not mush more than 2ph, might be worth it for a grinder
[11:49:39] <jschi[m]> * will investigate
[11:53:17] <jesseg> wow 3 phase steppers? tell me more! or I could search on ebay :P
[11:53:33] <jesseg> are they closed loop?
[11:53:52] <hazzy-m> Yes, and awesome!
[11:53:56] <hazzy-m> very fast and quiet
[11:54:05] <hazzy-m> can't loose steps
[11:54:28] <jesseg> in what way are they better than a closed loop 2/4 phase stepper?
[11:54:52] <jesseg> or is 3ph stepper another term for brushless servo motor? :D
[11:54:55] <hazzy-m> very low resonance, so much smoother motion
[11:55:07] <jesseg> fascinating
[11:55:21] <hazzy-m> they are basically a high pole count BL servo
[11:55:26] <Roguish> jesseg: try dampers on the steppers.
[11:55:48] <hazzy-m> Roguish: That is no fun :D
[11:55:52] <jesseg> Roguish, huh? I hate steppers. I'm a closed loop / servo fanatic.
[11:56:03] <Roguish> like a harmonic balancer on a car engine crankshaft
[11:56:30] <jesseg> I even had some 5 phase german steppers running closed loop - it was working amazingly
[11:56:40] <Roguish> i don't care for steppers either. servos rule. but if steppers is what ya got, run 'em.
[11:57:24] <Roguish> all, any, can get vibs, and dampers can really help for critical applications.
[11:58:09] <Roguish> you can even add a little friction. it also works as a vib damper.
[12:00:07] <jesseg> Roguish, I actually was experimenting with a step-stalling scheme where the stepper basically operated like a stepping motor but the microcontroller simply monitored the encoder and would stall incoming step commands until the stepper was caught up to the point where it could accept them without losing step -- and if you forced the motor, it would counter "step back" so it never lost count. So no PID or anything, but it worked quite nicely.
[12:00:44] <Roguish> interesting study.
[12:02:50] <Roguish> I guess if 'time' isn't a critical thing, that would be ok.
[12:03:45] <jesseg> I don't understand
[12:04:12] <jesseg> There would be no time delay, except what was mechanically needed by the dynamics of the motor -- in other words, only in cases where a step would be lost..
[12:04:38] <jesseg> and as soon as the stepper mechanically moved to where it could accept the queued step it would receive it, so it's much better than open loop
[12:05:18] <jesseg> it's every bit as quick as open loop, and it has the added bonus that you can actually over step a few steps (128 in my case) and the motor will catch up as soon as it can with no step loss.
[12:06:49] <jesseg> That was on a 5 phase stepper. But I also did an experiment where I had a regular 2/4 phase stepper running as a BLDC motor with an optical encoder. I found that I could have the microcontroller advance the timing to 180 degrees once the stepper was above a certain speed. In other words, it could run either way - and it runs really fast that way. Of course if the speed gets pulled down below a certain point it'd switch back to 90 degrees advanced.
[12:08:22] <hazzy-m> The closed loop steppers on my mill are that way. If I stall or force the motor shaft the driver stops or steps back to stay in sync with the rotor, it works very well, and it always knows where it is so there is no such thing as a lost step
[12:09:14] <jesseg> yeah exactly
[12:15:27] <hazzy-m> The nice thing it that that unlike regular steppers which are constant current, the drive only applies as much current as needed to maintain a reasonable following error, so the motors don't get hot, and for short periods of time it can send much more current than the stepper is rated for. The motors effectively have about twice the power/torque (and speed) as they did when run open loop
[12:17:41] <jesseg> it's probably much more than twice because you can load them so much more heavily without the fear of losing a step
[12:20:29] <hazzy-m> you are right, I was able to get to twice the rapid speeds, but since the torque curve drops way off with steppers at the same speed it probably has many times the usable torque
[14:14:32] <Not> [02QtPyVCP] 07Kurt Jacobson pushed 038 commits to 03actions [+0/-0/±12] 13https://gitlab.com/Hazzy/qtpyvcp/compare/76c0d19e65519cfb5b4e98c146a66d9bc88cbe1b...47235db84b502d8dcd18b5678c42061cce37e6b6
[14:14:34] <Not> [02QtPyVCP] 07Kurt Jacobson 03b3b9f9a - add machine.override-limits action
[14:14:35] <Not> [02QtPyVCP] 07Kurt Jacobson 039308f40 - fix bindOk errors with jog actions
[14:14:37] <Not> [02QtPyVCP] 07Kurt Jacobson 03c188653 - widget should be a keyword arg to bindOk
[14:14:38] <Not> [02QtPyVCP] ... and 5 more commits.
[17:16:00] <Not> [02QtPyVCP] 07Kurt Jacobson pushed 032 commits to 03actions [+1/-0/±5] 13https://gitlab.com/Hazzy/qtpyvcp/compare/47235db84b502d8dcd18b5678c42061cce37e6b6...332647287e45ea6283e2c3c812c2b1c07293de31
[17:16:01] <Not> [02QtPyVCP] 07Kurt Jacobson 0302da43d - add action slider
[17:16:03] <Not> [02QtPyVCP] 07Kurt Jacobson 033326472 - get spindle OR action working with slider
[17:22:18] <hazzy> TurBoss: When you get a chance some time in the next few days, would you mind testing the actions branch? It is only the brender UI that I have updated to use them
[17:23:40] <TurBoss> now is the time
[17:24:02] <TurBoss> Hey Lcvette
[17:24:31] <TurBoss> 10 mins and I'm with it
[17:27:33] <Lcvette> Jschi, grab some of these
[17:27:34] <Lcvette> https://rover.ebay.com
[17:27:50] <Lcvette> Turboss!
[17:27:57] <Lcvette> What's up man!!
[17:28:56] <TurBoss> not much
[17:29:22] <TurBoss> we where worried about you xD
[17:38:51] <TurBoss> hazzy: how can I issue G54 ?
[17:38:52] <TurBoss> is posible?
[17:49:18] <TurBoss> should be a sim issue but I can't home X and Y again
[19:24:11] <hazzy> hmm, what sim are you using?
[19:24:37] <hazzy> I have only tested the action branch with brender
[19:26:46] <TurBoss> hazzy: sim/brender.ini
[19:59:38] <hazzy> ok, that should work, I'll check it
[20:56:31] <Not> [02QtPyVCP] 07Kurt Jacobson pushed 032 commits to 03actions [+0/-1/±5] 13https://gitlab.com/Hazzy/qtpyvcp/compare/332647287e45ea6283e2c3c812c2b1c07293de31...40ee6986ba985830cd957a8625d7a894dd4ded01
[20:56:32] <Not> [02QtPyVCP] 07Kurt Jacobson 03b25a1e0 - minor clean up on actions __init__.py
[20:56:34] <Not> [02QtPyVCP] 07Kurt Jacobson 0340ee698 - remove old action_button.py
[21:03:16] <Not> [02QtPyVCP] 07Kurt Jacobson pushed 032 commits to 03actions [+0/-2/±2] 13https://gitlab.com/Hazzy/qtpyvcp/compare/40ee6986ba985830cd957a8625d7a894dd4ded01...5ed2e919c4646c00907587513f9c8b32f433c839
[21:03:17] <Not> [02QtPyVCP] 07Kurt Jacobson 03e1d6320 - delete axis_button.py
[21:03:19] <Not> [02QtPyVCP] 07Kurt Jacobson 035ed2e91 - delete jog_button.py
[21:26:17] <Not> [02QtPyVCP] 07Kurt Jacobson pushed 032 commits to 03actions [+0/-0/±2] 13https://gitlab.com/Hazzy/qtpyvcp/compare/5ed2e919c4646c00907587513f9c8b32f433c839...8718d34d9c045a4d998c65c0c5acc4e591635cc4
[21:26:18] <Not> [02QtPyVCP] 07Kurt Jacobson 036b13ea7 - delete old actions dict experiment
[21:26:20] <Not> [02QtPyVCP] 07Kurt Jacobson 038718d34 - user new actions for menu homing actions
[21:55:41] <Not> [02QtPyVCP] 07Kurt Jacobson pushed 033 commits to 03actions [+0/-0/±4] 13https://gitlab.com/Hazzy/qtpyvcp/compare/8718d34d9c045a4d998c65c0c5acc4e591635cc4...d01b25e9054ed7699f2aad48dac698a7eff6cb56
[21:55:43] <Not> [02QtPyVCP] 07Kurt Jacobson 0374c7464 - a little clean up in main_window.py
[21:55:44] <Not> [02QtPyVCP] 07Kurt Jacobson 03e00686e - add rest of mysterious feed-hold actions
[21:55:46] <Not> [02QtPyVCP] 07Kurt Jacobson 03d01b25e - enable/disable mode buttons properly
[21:56:11] * hazzy-m is on a roll
[21:56:12] <hazzy-m> :)
[21:56:22] <TurBoss> wow
[21:57:23] * hazzy-m goes afk
[22:42:46] * hazzy-m is back