#garfield Logs

Nov 06 2021

12:21 AM rue_mohr: er, carfull eh?
04:26 AM Tom_L: morning
04:35 AM rue_mohr: hah
04:36 AM rue_mohr: I'v not gone to bed yet
04:36 AM rue_mohr: well I had a nap after work
04:36 AM rue_mohr: its only 2:20am
04:36 AM Tom_L: i figured as much
04:36 AM Tom_L: what you workin on?
04:36 AM rue_mohr: I'm having a really hard time trying to be normal
04:36 AM Tom_L: you not been yourself this week
04:37 AM rue_mohr: someone on twitter was doing a christmas decoation project and I got distracted again
04:37 AM rue_mohr: I'm supposed to be working on that power converter
04:38 AM rue_mohr: here is a scary thought
04:38 AM rue_mohr: before I could make pcbs, I re-used pcbs
04:39 AM rue_mohr: if I run out of clad, I guess I'm back to it
04:39 AM rue_mohr: http://ruemohr.org/~ircjunk/tempimage/p1210607.jpg
04:39 AM rue_mohr: http://ruemohr.org/~ircjunk/tempimage/p1210609.jpg
04:39 AM rue_mohr: http://ruemohr.org/~ircjunk/tempimage/p1210610.jpg
04:39 AM rue_mohr: http://ruemohr.org/~ircjunk/tempimage/p1210611.jpg
04:40 AM Tom_L: you running out?
04:40 AM rue_mohr: not currently
04:40 AM Tom_L: you set the yt i put up?
04:41 AM Tom_L: see
04:41 AM rue_mohr: ... no
04:41 AM rue_mohr: 1 min
04:41 AM rue_mohr: https://www.youtube.com/watch?v=YUAqIt_YRQQ
04:41 AM Tom_L: got it goin finally
04:41 AM rue_mohr: nice
04:42 AM rue_mohr: now you need a tailstock ;)
04:42 AM Tom_L: one line of code in the other config was commented out and i'd been focusing on the first file
04:42 AM Tom_L: i have a tailstock for it
04:42 AM rue_mohr: ooo splines
04:42 AM Tom_L: it came with it
04:42 AM Tom_L: got the cad post working too i think
04:43 AM rue_mohr: that looks like a .... vise screw of some kind?
04:43 AM rue_mohr: that its making?
04:43 AM rue_mohr: no
04:43 AM Tom_L: just some test i dreamed up
04:43 AM rue_mohr: its a double helix
04:43 AM rue_mohr: its drilling thru....
04:43 AM Tom_L: canned cycle test too
04:44 AM rue_mohr: in the lead it cut
04:45 AM rue_mohr: looks like twitter wants "lots more" schematics from me
04:45 AM rue_mohr: hmm
04:47 AM Tom_L: http://tom-itx.no-ip.biz:81/~webpage/cnc/rotary/mount/Rotary_unwrappede.jpg
04:47 AM Tom_L: you start with that
04:47 AM Tom_L: and then wrap it around an axis
04:47 AM Tom_L: http://tom-itx.no-ip.biz:81/~webpage/cnc/rotary/mount/Rotary_wrapped.jpg
04:47 AM rue_mohr: huh
04:48 AM Tom_L: https://www.youtube.com/watch?v=ZN9aw9094ag
04:48 AM Tom_L: not a flat spiral
04:48 AM Tom_L: i should have angled it more
04:48 AM rue_mohr: huh
04:49 AM Tom_L: i had to write that post from scratch
04:49 AM rue_mohr: hmm
04:49 AM rue_mohr: I thought you do a lot of that anyhow
04:49 AM Tom_L: yeah some but never on a rotary
04:50 AM Tom_L: i was kinda surprised it worked
04:50 AM Tom_L: http://tom-itx.no-ip.biz:81/~webpage/cnc/rotary/mount/Rotary_test.txt
04:50 AM Tom_L: there's the file
04:51 AM rue_mohr: what did you have to add in
04:52 AM Tom_L: mostly it was just #INDXA stuff
04:52 AM Tom_L: telling it where A should be
04:52 AM rue_mohr: haha
04:52 AM rue_mohr: I got a package of stuff from china today
04:52 AM Tom_L: but had to figure out where to put some of them so like the drills would wait til A moved and stopped before drilling
04:52 AM Tom_L: nice
04:52 AM rue_mohr: and one of the things that arrived were those little tact buttons
04:52 AM rue_mohr: I was confused, I already have a million of them
04:53 AM rue_mohr: so I was just looking
04:53 AM rue_mohr: THEIR NORMALLY CLOSED!
04:53 AM rue_mohr: so I'm gonna check that
04:53 AM rue_mohr: I'm glad I didn't just dump them in with the normal ones
04:54 AM Tom_L: kinda odd they'd be closed
04:55 AM rue_mohr: yea thats why I got them
04:55 AM Tom_L: oh
04:57 AM Tom_L: http://tom-itx.no-ip.biz:81/~webpage/cnc/Smartcam/LinuxCNC_A.TMP
04:57 AM Tom_L: that's what the post looks like
04:58 AM rue_mohr: oooh
04:58 AM Tom_L: alot of the new ones are python i think or something similar
04:58 AM Tom_L: xml maybe?
04:58 AM Tom_L: but i'm able to run it without mods
04:59 AM Tom_L: before i had to mess with the output file but as time went along i got it working with no changes
04:59 AM Tom_L: so i took the good 3 axis and added 4 and that's what you see
05:00 AM Tom_L: it will need a few tweaks i'm sure
05:01 AM Tom_L: you're doing alot more on twitter now i see
05:05 AM rue_mohr: yea
05:05 AM Tom_L: http://tom-itx.no-ip.biz:81/~webpage/cnc/rotary/mount/Rotary_unwrapped_2.jpg
05:06 AM rue_mohr: being the only one alive on irc and trying to always spark up some life was gettin me down
05:06 AM Tom_L: i did 2 paths in the video you saw
05:06 AM Tom_L: i know i could tell
05:06 AM rue_mohr: ah, that explains it
05:06 AM Tom_L: i wanted one path to force the rotary to reverse and still follow x
05:06 AM rue_mohr: k
05:06 AM Tom_L: same with the drills
05:07 AM Tom_L: http://tom-itx.no-ip.biz:81/~webpage/cnc/rotary/mount/Rotary_wrapped_2.jpg
05:08 AM rue_mohr: http://ruemohr.org/~ircjunk/tempimage/p1210614.jpg
05:08 AM rue_mohr: http://ruemohr.org/~ircjunk/tempimage/p1210615.jpg
05:08 AM Tom_L: those buttons?
05:09 AM rue_mohr: under my finger there
05:09 AM rue_mohr: but their normally closed
05:09 AM Tom_L: yep
05:47 AM rue_mohr: llorens
10:19 AM aandrew: woo, furnace room disinfected/deodorized, just waiting for the room to clear so I can turn on my networking gear again
03:37 PM rue_mohr: 1874
03:37 PM rue_mohr: is that up or down, doesn't matter but I'm being frustrated that I cant remember the last number
03:50 PM Tom_L: the last number was earlier today
03:51 PM Tom_L: you should associate it with a key search word
03:54 PM Tom_L: <rue_mohr> 1870 followers, I dont know if thats up or down
03:54 PM Tom_L: and the home team is up by 4
04:53 PM rue_mohr: :)
06:52 PM Tom_L: added an 'indexer' section to the post
06:54 PM Tom_L: https://www.youtube.com/watch?v=5xIwtoPkHY0
07:05 PM rue_mohr: how do you use it.
07:05 PM rue_mohr: that looks like it has more tha 4 sides
07:05 PM rue_mohr: was that 8 sided?
07:06 PM Tom_L: hex head
07:07 PM Tom_L: line, index, line, index....
07:07 PM rue_shop3: I'm having a problem understanding transformer saturation
07:07 PM Tom_L: well don't ask me for sure
07:07 PM Tom_L: :)
07:08 PM rue_shop3: I thought I had it, and then what I'm trying is proving me wrong
07:08 PM Tom_L: i bet different types produce different results
07:08 PM Tom_L: torroid vs wafer vs wafer different shape etc
07:09 PM Tom_L: so my post is pretty complete i think
07:47 PM rue_shop3: saturation is the same thing tho
08:31 PM rue_mohr: aandrew, polprog
08:31 PM rue_mohr: ya there?
08:31 PM rue_mohr: this is good to know, inductor saturation
09:16 PM Tom_L: i should try one of your tubes on the rotary for the heck of it
09:34 PM rue_mohr: did I send you a tube?
09:34 PM Tom_L: i think so
09:54 PM aandrew: rue_mohr: I'm here, what's up
10:16 PM rue_shop3: inductor saturation
10:40 PM aandrew: yeah don't do that
10:40 PM rue_mohr: oh but I did!
10:40 PM rue_mohr: and I learned a LOT!
10:41 PM rue_mohr: https://twitter.com/RueNahcMohr/status/1457156850550394884
10:43 PM rue_mohr: using what I learned I made an oscillator thats timed by the inductor saturation
10:50 PM aandrew: I saw that
10:50 PM rue_mohr: this is really cool
10:51 PM rue_mohr: when an inductor saturates, its polarity changes
10:53 PM rue_mohr: http://ruemohr.org/~ircjunk/tempimage/p1210652.jpg
10:53 PM rue_mohr: this circuti looks like it shouldn't work
10:53 PM rue_mohr: the transistor looks like it should never turn off
10:54 PM rue_mohr: but it does, when the inductor saturates, the polarity flips and triggers the opto to turn off the drive
10:54 PM rue_mohr: from there, its held off by the inductors flyback
10:54 PM rue_mohr: untill its discharged
10:54 PM rue_mohr: then the opto turns off, and the driver turns on
10:56 PM rue_mohr: without knowing about the reversal thing, I'v have never guessed
10:56 PM Tom_L: what causes it to reverse?
10:56 PM rue_mohr: yea
10:56 PM rue_mohr: :) I'm still trying to grasp all this
10:57 PM Tom_L: aliens?
10:57 PM rue_mohr: well, its part of the saturation behaviour
10:58 PM rue_mohr: http://ruemohr.org/~ircjunk/tempimage/p1210641.jpg
10:58 PM rue_mohr: to understand it, I took an inductor and shoved a pulse into it
10:58 PM rue_mohr: the top trace is the drive, so most of that diagram is the "on" phase
10:58 PM rue_mohr: the current goes up linear, as one would expect
10:59 PM rue_mohr: till the system saturates (its actually the transistor cant supply any more current)
10:59 PM rue_mohr: then the voltage across the inductor collapses
10:59 PM rue_mohr: then it over-collapses
10:59 PM rue_mohr: and finishes ringing
11:00 PM rue_mohr: but the core is still magnetized, so when the transistor lets up, there is still flyback voltage that here gets dumped into the leds
11:01 PM rue_mohr: linear discharge, till the inductor is out of steam, then the voltage drops out and rings
11:01 PM * Tom_L finishes up a bowl of grapes
11:01 PM Tom_L: i had a good day productivity wise
11:01 PM rue_mohr: yea, I'm excited to see ya cut soemthing
11:02 PM Tom_L: thing about your tube is the edges of the slot (i haven't dug yours out yet to look) will be perp to the point it is cut at not like it would be non rotated
11:03 PM rue_mohr: you have 3 axies
11:03 PM Tom_L: i have 4
11:03 PM rue_mohr: so, its really just indexing
11:04 PM rue_mohr: heh, ok not what i meant,but yea
11:04 PM rue_mohr: its silly simple, cut slot as 2 axis, rotate, cut other slot
11:04 PM Tom_L: it might be tough to program without a 3d model to unwrap
11:05 PM Tom_L: how do you cut the big slots?
11:05 PM rue_mohr: you could hand-bang in the A axis index
11:05 PM rue_mohr: they are both done with a jig for the router
11:05 PM rue_mohr: aka wood router table
11:05 PM Tom_L: where's the fun in that?
11:05 PM rue_mohr: 2 bits, one 15mm one 10? mm
11:05 PM Tom_L: ok
11:06 PM Tom_L: i bet that makes alot of noise
11:06 PM rue_mohr: bolt on the alignment block, push end into router
11:06 PM rue_mohr: change bits
11:06 PM rue_mohr: flip over, push into router
11:06 PM rue_mohr: obviously we batch them
11:06 PM rue_mohr: so its not that bad
11:06 PM rue_mohr: about 5-7 at a time
11:07 PM Tom_L: that actually might be a difficult part to us A axis on
11:07 PM rue_mohr: how so
11:08 PM Tom_L: because it needs flipped and nothing to hold it when it's squeezed unless i make a plug for the ID
11:08 PM Tom_L: and to get it indexed in alignment with the other half of the slot
11:08 PM rue_mohr: set up path for one side ( a U path ) insert rotate command, set up path for second side ( U )
11:08 PM Tom_L: the path isn't the problem
11:09 PM Tom_L: same path for both big slots but aligning it might be
11:09 PM rue_mohr: hmm
11:09 PM Tom_L: so the way you do it may be quicker and easier
11:09 PM rue_mohr: score a line downt eh side and align it with the edge of a jaw?
11:10 PM Tom_L: how wide is that little saw cut?
11:10 PM rue_mohr: thickness of the head, so 1/8"
11:10 PM Tom_L: i could use that slot cutter i got for that
11:11 PM rue_mohr: :) then it just remains to work out how to do ones that are 36" long ;)
11:11 PM Tom_L: i'm wondering if being unsupported like that it would snap cutters due to vibration on the end
11:11 PM rue_mohr: it would be pretty dodgy even at 3"
11:15 PM Tom_L: stuff a wood dowel inside to help vibration
11:15 PM Tom_L: at least on the end
11:21 PM rue_mohr: all these transformer based oscillators suddenly make sense now
11:22 PM rue_mohr: http://ruemohr.org/~ircjunk/tempimage/p1210561.jpg
11:22 PM rue_mohr: ignore the diode and 100R resistor, they are just flyback control
11:23 PM rue_mohr: as soon as you know the polarity reverses on the coil when it saturates, the whole operation of it makes sense
11:23 PM Tom_L: wonder how the first guy figured it out
11:24 PM rue_mohr: I think he threw bannanas at it till something new happened
11:24 PM rue_mohr: then arranged the mess into the simplest working example
11:24 PM rue_mohr: wrote about it, got a prize, etc, etc etc
11:25 PM rue_mohr: but its kinda taught me something else too
11:26 PM rue_mohr: designing them isn't about core saturation, its about transistor drive saturation
11:26 PM rue_mohr: which explains things
11:26 PM rue_mohr: they say that over a certain votlage, royer isn't a good arch
11:26 PM rue_mohr: yea, because you cant get your transistors to saturate right
11:27 PM rue_mohr: having a circuit kick the collector of an on transistor from 0-13V is one thing
11:27 PM rue_mohr: having a circuit kick a transistor up to 145V is another
11:27 PM rue_mohr: you dont want to do that when their on
11:28 PM rue_mohr: the instantanious power is pretty far over the ouch region for the die
11:29 PM rue_mohr: in PC power supplies they use a current sense transformer
11:29 PM rue_mohr: to detect the current spike before the voltage goes crazy
11:31 PM * Tom_L is having bagels now. grapes weren't enough
11:32 PM * rue_mohr slides the haloween basket up beside tom
11:33 PM Tom_L: she was gonna go on a trip leaving around 10pm and it got cancelled so now she's not the slightest bit sleepy
11:34 PM Tom_L: so i may be somewhat awake all night
11:34 PM Tom_L: she'd gone to sleep around 3