#garfield Logs

Feb 23 2020

01:11 AM X704 is now known as tiwake
03:02 AM X704 is now known as tiwake
03:11 AM polprog: good job on that converter rue
03:11 AM polprog: how do you open the transformers without cracking the thing that goes around the coil vertically?
03:20 AM rue_bed: the E cores?
03:21 AM rue_bed: https://i.ebayimg.com/images/g/inMAAOSwsZJaWfzY/s-l400.jpg
03:21 AM rue_bed: their standard E16
03:22 AM rue_bed: funny enough its in the part number of most of them
03:22 AM rue_bed: LOTS of the cores on the ones I had were broken tho
03:23 AM rue_bed: you cant tell till you start to take it apart
03:44 AM Tom_L: you can buy them split like that?
03:45 AM Tom_L: i thought the fingers had to overlap to work correctly
03:45 AM Tom_L: but i know little to nothing about xfrmrs
04:41 AM polprog: oh lol the tiny ones have E shaped cores as well o_O
05:01 AM polprog: im taking apart a 2.00mm raster one (useless on breadboard) and its totally covered in glue, i cracked the core which sucks.. but its a big one so lots of wire
05:09 AM rue_bed: yep, boiling in water or soaking in acetone breaks it up
05:09 AM rue_bed: it can be hard to get the halves free tho
05:09 AM rue_bed: if the middle of the E stick, it'll snap when pulled up
05:09 AM rue_bed: you have to be really delicate
05:12 AM polprog: thats a good tip.. acetone
05:12 AM polprog: ipa doesnt do much i take it?
05:25 AM X704 is now known as tiwake
05:34 AM X704 is now known as tiwake
05:55 AM polprog: im trying boiling water now
06:58 AM X704 is now known as tiwake
07:27 AM X704 is now known as tiwake
10:51 AM Tom_L: http://cq.cx/ladder.pl
11:55 AM rue_bed: pl?
11:55 AM rue_bed: hah
11:56 AM rue_bed: that might be interesting source
11:56 AM rue_bed: I recall projcts like that
11:57 AM rue_bed: its not even 10am yet
11:57 AM rue_bed: wow
11:57 AM rue_bed: I was up till like 6
11:57 AM rue_bed: iirc
11:57 AM rue_bed: its sunday, hmmm
11:59 AM rue_bed: http://www.kiblerelectronics.com/bob/app_notes/note11/images/schematic1_zvs_driver.png
11:59 AM rue_bed: hmm
11:59 AM rue_bed: ok, what am I doing today
12:00 PM rue_bed: I should do things on my ever day list
12:00 PM rue_bed: and assemble more motor controller I guess
12:00 PM rue_bed: the drive supply is working
12:03 PM polprog: ive managed to get a better capture of my 5 ns pulse
12:03 PM polprog: using a higer speed "probe" made from coaxial cable and a 50 ohm terminate
12:03 PM polprog: terminator*
12:03 PM polprog: https://polprog.net/rozne1/ircjunk/5-nanos.png
12:04 PM polprog: you should see a single trace, if not you have old file in cache
01:06 PM rue_mohr: wow, clean
01:07 PM polprog: yeah. i rewatched Mike's video on high speed probing
01:07 PM polprog: https://polprog.net/rozne1/ircjunk/hi-speed-probe.png
01:07 PM polprog: thats a 50 ohm terminator there since my scope has 1M termination only
01:08 PM polprog: not sure if i can call that termination :D
01:11 PM rue_mohr: mine has a switch to internally add a 50 ohm, I'm afraid one day I'm gonna leave it on and blow it up
01:12 PM rue_mohr: so which circuit were you using for that
01:12 PM polprog: i dont think i uploaded it.. gimme a sec
01:13 PM polprog: oh, the old scope screenshot is in the file
01:13 PM polprog: that should be changed
01:14 PM rue_mohr: use Hypertext markup Language to tie all the stuff togethor
01:15 PM polprog: i mean the physical file
01:15 PM polprog: i should prune it... i have like 5 copies of the same circuit now
01:16 PM polprog: i could throw some html page togeth0r
01:16 PM polprog: but then again, what, i get a new scope in a month and have to repeat this all over again
01:16 PM rue_mohr: na, you just change the images
01:16 PM polprog: no, i mean the physical file again :D
01:16 PM rue_mohr: the html deals with rendering ita ll
01:16 PM polprog: i know
01:17 PM rue_mohr: :)
01:17 PM polprog: but i cant html-edit the paper file
01:17 PM rue_mohr: ctrl-F P
01:17 PM rue_mohr: ...
01:17 PM rue_mohr: alt-F P
01:19 PM polprog: ok, lets see if gimp can scan
01:19 PM polprog: no
01:19 PM polprog: the bad thins is this particular samsung printer doesnt work in linux
01:23 PM polprog: https://polprog.net/rozne1/ircjunk/5-nanos-sch.png
01:24 PM rue_mohr: ah ok
01:24 PM rue_mohr: yes, a dual edge detector
01:25 PM polprog: i wonder if just 1 inverter would be enough
01:25 PM polprog: i tried measuring that but i got a run pulse that was half Vcc in height
01:25 PM polprog: the question is, was that real or was it just the limit of my scope
01:36 PM rue_mohr: yea, thats why I'd like to run it thru mine
01:36 PM rue_mohr: 2Gs w/650Mhz
01:37 PM rue_mohr: or was it 500Mhz
01:41 PM polprog: neato
01:42 PM polprog: mine will be 100MHz 200Ms/s
01:42 PM polprog: ill be happy to see your results.
01:42 PM polprog: we could collaborate and make one html file :) that would be neat
01:56 PM rue_shop1: mine is 500M 2Gs
02:17 PM rue_shop1: polprog, I dont have any AC
02:17 PM polprog: shame
02:17 PM rue_shop1: so I did it with HC
02:17 PM polprog: what did you get with HC?
02:17 PM rue_shop1: I'd try LS, but I know its worse
02:17 PM polprog: yeah
02:17 PM rue_shop1: 1 min...
02:18 PM polprog: btw, that tek of yours, what model was it? and how do you transfer the screenshots (i think i asked but i forgot). Floppies?
02:18 PM rue_shop1: 754C
02:18 PM rue_shop1: I do it by serial
02:18 PM rue_shop1: but its painfully slow
02:19 PM rue_shop1: I plan to make a parallel to serial adapter that can do 115200 baud
02:19 PM rue_shop1: with an stm32
02:19 PM rue_shop1: I do it via .eps file cause its smaller
02:27 PM rue_mohr: http://ruemohr.org/~ircjunk/tempimage/polprogPulse1.png
02:27 PM rue_mohr: there is with 50 ohms
02:27 PM rue_mohr: its 2 invert gates, HC
02:28 PM rue_mohr: w/x10 probe
02:28 PM rue_mohr: but its only a 100Mhz probe
02:32 PM polprog: eww, thats a bit mismatched there i think
02:33 PM rue_shop1: 1 gate is closer to 5ns
02:33 PM polprog: hmm
02:33 PM rue_shop1: :)
02:33 PM polprog: that doesnt seem right
02:33 PM polprog: shouldnt you be using 1x probe with 50 ohms inpit?
02:33 PM polprog: input*
02:34 PM polprog: also, can i put that in https://polprog.net/rozne1/ircjunk/5nanos/5nanos.html
02:34 PM polprog: ?
02:48 PM rue_shop1: hmm, good point
02:48 PM rue_shop1: yea enjoy :)
02:49 PM rue_shop1: the scope says the two inverter gates delay the signal 8ns
02:50 PM polprog: yeah, so thats yours, on HCT, with one inverter?
02:51 PM rue_shop1: I think the HC is too slow
02:51 PM rue_shop1: 74HC86
02:51 PM rue_shop1: and 74HC04
02:51 PM rue_shop1: that was two inverters
02:53 PM polprog: did you check that impedance?
02:54 PM polprog: also the resolution on that tek is amazing.. my laptop has a smaller screen.
03:01 PM rue_shop1: :)
03:01 PM rue_shop1: I export is as vector and then use inkscape to convert it to that raster
03:02 PM rue_shop1: I must have not used 10x with 50R cause otherwise it would have died right off right away
03:04 PM polprog: thats a single acquisition on there?
03:04 PM polprog: im getting two different runt pulses o_O
03:04 PM rue_shop1: I was playing and it snubs out the dc state
03:05 PM rue_shop1: I paused it and did the export, they call it "high resolution mode"
03:06 PM rue_shop1: there is a little note at the top about the aq's
03:06 PM rue_shop1: which I'm not totally tuned into just yet
03:07 PM rue_shop1: :)
03:07 PM polprog: hah, im getting a different slew rate for a rising, and a falling edge of the rel osc
03:08 PM rue_shop1: yea, I'm not sure if "9401 Acqs" means samples or overlaped passes
03:09 PM rue_shop1: :)
03:10 PM polprog: that looks like a single pass to me
03:10 PM rue_shop1: I DO have it set to a long pass
03:14 PM polprog: did you get it to show you a non ringing wfm?
03:14 PM rue_shop1: it never does
03:14 PM rue_shop1: heh, I took it apart, working on the motor driver
03:14 PM rue_shop1: we could tune it more later in the week if you like
03:18 PM polprog: sure. thanks :)
03:18 PM polprog: iv eupdated the html file
03:18 PM polprog: ive updated*
04:04 PM rue_shop1: ok, guest is gone
04:04 PM rue_shop1: what was I doing...
04:04 PM rue_shop1: hip 4081
04:52 PM Tom_L: is that the one rif used on his hbridge?
04:53 PM MoonyMoon: Just what the fuck are you talking about ? :D
04:53 PM * Tom_L gives MoonyMoon a 5th of canadian club
04:54 PM MoonyMoon: That would help XD
04:55 PM MoonyMoon: though I'm trying out a sober streak for the next few months :p
05:04 PM Tom_L: i gave that up ages ago
05:43 PM rue_shop1: yes, its what rif used, but he was using an online design
05:43 PM rue_shop1: openservo
05:44 PM rue_shop1: but something is wrong
05:44 PM rue_shop1: the floating supply doesn't look like it and the driver are going right
05:44 PM rue_shop1: oh
05:46 PM rue_shop1: yes!
05:47 PM rue_shop1: ok, I forgot to link the driver to the fets
06:05 PM Tom_L: ok
06:50 PM rue_mohr: so, it looks like the next step is to solder something togethor
06:50 PM rue_mohr: then I can test without all the breadboard issues
07:24 PM X704 is now known as tiwake
08:54 PM rue_shop1: so, irfz44 is 28m ohms, which would be 22w off an H bridge at 20A
08:56 PM rue_shop1: in context, as 2N3055 is 3V drop, which would be 120w loss total
08:57 PM rue_shop1: IRFB1440 is 4m ohms
08:57 PM rue_shop1: which would be 3.2w loss on an H bridge at 20A
08:58 PM rue_shop1: hmm, I want to know more about these motors
09:10 PM rue_shop1: locked rotor looks more like 15A
09:16 PM rue_shop1: 15A on that one too
09:47 PM rue_shop1: ok, the better motors I have, are about 1 ohm
09:55 PM rue_mohr: Tom_L, I need a target for my dc motor servo
09:55 PM rue_mohr: got a datahseet for the steppers you used on your machine anywhere?
10:00 PM rue_mohr: lactose free, hah
10:02 PM rue_mohr: polprog, I wonder if I can find my data on the things I tried for varicaps
10:02 PM rue_mohr: tho, iirc, I was starting to suspect there was leakage that was causing the osc to change for other reasons
10:13 PM Tom_L: yes
10:13 PM rue_mohr: https://www.youtube.com/watch?v=PFqjQ3l1MRA
10:13 PM rue_mohr: oof
10:14 PM Tom_L: not much of one but yes
10:14 PM Tom_L: https://www.ebay.com/itm/3pcs-NEMA23-570OZ-IN-5A-3-8-DUAL-SHAFT-STEPPER-MOTOR/253223632606?ssPageName=STRK%3AMEBIDX%3AIT&_trksid=p2060353.m1438.l2649
10:15 PM Tom_L: 5A iirc
10:22 PM rue_mohr: I need to know its speed/accel characteristics
10:28 PM Tom_L: i'm looking to see if i have that
10:30 PM rue_mohr: yea I cant put anything togethor
10:31 PM rue_mohr: ah, might have something
10:31 PM rue_mohr: er, nope
10:32 PM Tom_L: https://www.automationtechnologiesinc.com/products-page/nema-23/kl23h2100-50-4bm/
10:33 PM Tom_L: nope, that's all there is
10:33 PM rue_mohr: no, there should be charts
10:33 PM Tom_L: https://www.automationtechnologiesinc.com/wp-content/uploads/downloads/2015/04/KL23H2100-50-4BM.pdf
10:33 PM Tom_L: best i can find so far
10:33 PM rue_mohr: with resonance information and things
10:33 PM Tom_L: it's a tough sob though
10:33 PM Tom_L: i've never stalled one
10:34 PM Tom_L: and trust me i've tried
10:34 PM Tom_L: not always willingly
10:35 PM rue_mohr: yea I need its performance data so I can make a dc motor match it
10:35 PM Tom_L: not coming up with anything yet
10:36 PM Tom_L: https://www.cnczone.com/forums/stepper-motors-drives/148604-cnc.html
10:36 PM Tom_L: maybe something there
10:37 PM Tom_L: https://www.cnczone.com/forums/attachment.php?attachmentid=153817&d=1330321455
10:38 PM rue_mohr: not logged in, wont give it to me
10:40 PM Tom_L: http://tom-itx.no-ip.biz:81/~webpage/rue/stepper_chart.jpg
10:40 PM Tom_L: about all i can find on it
10:41 PM rue_shop1: er, thats... not... no...
10:42 PM rue_shop1: I'm hunting for a baldor stepper
10:43 PM rue_mohr: they will ahve a proper data sheet
10:44 PM rue_mohr: ok, their page is broken
10:44 PM Tom_L: that was the biggest nema23 i could find anywhere
10:44 PM rue_mohr: who is a proper stepper manufacurer..
10:46 PM Tom_L: EMI "Empire Magnetics Inc." {Stepper Motors}
10:46 PM Tom_L: Globe Motors {Military Motors / Gear Boxes}
10:46 PM Tom_L: Haydon Switch and Instrument, Inc. ...
10:46 PM Tom_L: JVL Industri Elektronik {Integrated Servo Motors and Stepper Motors}
10:46 PM Tom_L: Kollmorgen {Stepper Motors}
10:46 PM Tom_L: Mclennan Servo Supplies Ltd. ...
10:46 PM Tom_L: MicroKinetics Corp. ...
10:46 PM Tom_L: Nippon Pulse Motor Co., Ltd.
10:46 PM rue_mohr: nippon
10:54 PM Tom_L: you're dealing with a few nimrods at night there
10:54 PM Tom_L: you might get further in the day
10:54 PM Tom_L: unreal seems to be able to build his own junk though
10:56 PM rue_mohr: heh
10:56 PM rue_mohr: the datasheet for the PJE86T is closer
10:57 PM rue_mohr: not by lots
10:57 PM rue_mohr: there should be some other charts
11:00 PM Tom_L: you just don't get much with chinese motors
11:01 PM rue_mohr: I know
11:09 PM X704 is now known as tiwake
11:17 PM X704 is now known as tiwake
11:44 PM X704 is now known as tiwake
11:53 PM * MoonyMoon crawls back into the bush