#garfield Logs

Dec 25 2020

12:45 AM rue_mohr: 5Khz
12:52 AM rue_mohr: 3.3V
04:39 AM rue_shop1: ok, the little switching converter works fine with the error amp
06:22 AM Tom_L: Merry Christmas
11:40 AM polprog: Merry Christmas to you too :)
11:40 AM polprog: rue_shop1: oh, nice. boost converters again?
12:24 PM rue_mohr: its that forward converter, for -5V
12:27 PM polprog: i need to build one at some point
12:35 PM rue_mohr: the first trick is finding a good core
12:35 PM rue_mohr: you want a primary of about 2mH
12:36 PM rue_mohr: at about 20 turns
12:36 PM rue_mohr: which I think puts 10 turns at .35mH or better?
12:38 PM rue_mohr: polprog, did I link you with the page that moon found me on inductors?
12:38 PM polprog: that toroidal calculator?
12:39 PM rue_mohr: yea
12:39 PM rue_mohr: but its not
12:39 PM rue_mohr: its just an inductor calc
12:39 PM polprog: yeah, i have it somewhere
12:39 PM rue_mohr: you leave the dimensions the same
12:39 PM polprog: yeah
12:39 PM rue_mohr: you just tell it how many turns and the inductance you got
12:39 PM rue_mohr: I been meaning to re-trim it for calculating how many turns you need after doing a test-winding
12:40 PM polprog: i should build a measurement jig for these
12:40 PM polprog: maybe some oscillating filter or something, to get a wide range
12:49 PM rue_mohr: hmm
12:49 PM rue_mohr: I got one of the china things, but its range is not a good fit
12:49 PM rue_mohr: If you have something that can go from about 5uH to 10mH its good
12:50 PM rue_mohr: you can use a scope and pulse generator
03:50 PM polprog: my idea was to make up a generator where the DUT would be the part of the oscillator circuit, and then measure the frequency or throw a micro and a small LCD on it
03:51 PM polprog: that would just measure the frequency and spit out a measurement in uH
04:11 PM rue_mohr: lots of things throw out the freq
04:11 PM rue_mohr: I think a scope and a pulse generator are a better idea, look at the current slope
04:25 PM polprog: hmm, could be a better solution
05:04 PM rue_shop1: these forward converters are really good on parts
05:04 PM rue_shop1: power density is good too
05:05 PM rue_shop1: turns out I need 7mA for this regulator board for my bench supply
05:05 PM rue_shop1: I was double-checking the charge pumps
05:05 PM rue_shop1: but they have awefull regulation
05:06 PM rue_shop1: the forward converter, at 6 components doesn't need regulation feedback, and yet the regulation is good
05:26 PM polprog: ive put the frequency counter back on the bench
05:27 PM polprog: i should see if i have any 10M xtal generators, internally it uses an RC generator to clock itself but there is an external 10M clock input
05:28 PM polprog: i could make up a small box that would clock it, probably could even tap into the power rails inside to avoid needing another power adapter
05:28 PM polprog: just a couple caps, the xtal generator and a female BNC
05:53 PM rue_shop2: hmm, or a 10Mhz module
05:53 PM polprog: that would be it, just a way to power and plug the 10M clock into the input
06:14 PM rue_mohr: you know the oscillator moduels for computers?
06:15 PM rue_mohr: darkrose, any more progress on if you should resort to moving to canada?
06:15 PM polprog: you mean the clock gen and a distribution IC?
06:15 PM polprog: i mean a 4 pin can with a crystal inside
06:15 PM rue_mohr: yea
06:15 PM polprog: i can never remember the acronyms
06:15 PM polprog: thats what i wanna use
06:15 PM polprog: im not sure if i have a 10M one, if not then ill just go get oe
06:15 PM polprog: one*
06:23 PM rue_mohr: take one from a network card and put a 2.5 divider on it
06:23 PM rue_mohr: er, or are they 50Mhz
06:24 PM rue_mohr: I think a 7490 can do 50Mhz
06:24 PM polprog: hmm
06:24 PM polprog: good idea
06:24 PM polprog: i have a 20M one for sure
06:24 PM rue_mohr: they were imported into TTL from the radio era
06:24 PM polprog: i have a shitton of 100M nics.. i should do something with them
06:24 PM rue_mohr: thats why they dont have the right pinout
06:24 PM polprog: 7490s?
06:24 PM rue_mohr: yea
06:25 PM rue_mohr: but they can divide by 5
06:26 PM polprog: right
06:26 PM polprog: interesting IC
06:32 PM rue_mohr: why do I have a problem getting up and doing things
06:32 PM rue_mohr: ar
06:32 PM rue_mohr: g
06:32 PM polprog: same
06:33 PM rue_mohr: polprog, do you have any magnet wire?
06:33 PM polprog: i think so
06:33 PM rue_mohr: or misc small wire?
06:33 PM polprog: i bought some
06:33 PM polprog: yeah, i have some
06:33 PM rue_mohr: k
06:34 PM rue_mohr: somehting simple to sit down with
06:34 PM polprog: .1 mm
06:34 PM rue_mohr: put 10 turns on different ferrite cores you have for inductance testing
06:34 PM rue_mohr: I was re-using the same peice of wire and it was stupid
06:34 PM rue_mohr: from 10 turns, you get know what you need
06:34 PM polprog: id have to set the measurement setup all up
06:35 PM polprog: its 1:30
06:35 PM rue_mohr: no , just put the test-turns on
06:35 PM polprog: ah
06:35 PM polprog: let me grab the stuff
06:35 PM rue_mohr: yea I mean if you find yourself in a cant-get-doing anything mode
06:35 PM rue_mohr: just put 10 turns on a bunch of different cores
06:35 PM rue_mohr: I had to go thru about 15 cores to find what I wanted
06:36 PM polprog: hmm
06:36 PM rue_mohr: as it turns out, the oscillator cores in compact flourescent lights are great
06:36 PM polprog: might as well move the laptop to the other desk, ill have to do that anyway when the labs begin
06:36 PM rue_mohr: I think thats what it was from, might have been a tracklight power converter
06:37 PM rue_mohr: I'm gonna strip down 2 more phones
06:37 PM rue_mohr: I think the 4x20 lcd is the only usefull thing in them
06:37 PM rue_mohr: I need a bigger box for lcds
06:38 PM polprog: how cool is that, i have space for two monitors and the laptop
06:39 PM rue_mohr: I'm supposed to redo this computer, it needs a new kernel and to have the new video card put in
06:39 PM rue_mohr: I'm not so enthusiastic about making a big change like that to it
06:39 PM polprog: ok, got 4 different cores
06:40 PM rue_mohr: k, about what size?
06:41 PM polprog: one is a flat flex choke, 1 5/8" length, 5/8" depth
06:41 PM rue_mohr: k
06:41 PM rue_mohr: I think those are high perm.
06:41 PM polprog: another is a rod, .5" dia, 1 3/8" length
06:41 PM rue_mohr: rods are always low inductance, but high saturation current
06:41 PM rue_mohr: rod/spool
06:42 PM polprog: arg, wrong measurement
06:42 PM polprog: .25" dia 5/16 depth
06:42 PM polprog: length
06:42 PM polprog: ermagherd
06:42 PM rue_mohr: k, do you know what its from?
06:42 PM rue_mohr: radio/power supply?
06:42 PM polprog: the rod is from an output stage of an atx psu
06:43 PM rue_mohr: ok, so lower freq higher ind.
06:43 PM polprog: most of this is from psus
06:43 PM rue_mohr: ok
06:43 PM rue_mohr: you know the beads they put around communication cables?
06:43 PM polprog: the toroids are 1) 0.5 outer, 0.25 inned
06:43 PM polprog: inner*
06:43 PM rue_mohr: they are often high permability ones
06:43 PM polprog: yeah, that would be the flat from the flatflex
06:44 PM rue_mohr: and odl usb cales
06:44 PM rue_mohr: old
06:44 PM polprog: yeah
06:44 PM rue_mohr: or good power cords
06:44 PM rue_mohr: or laptop adatpers
06:44 PM rue_mohr: the power filters, (linked inductors) are also high permiability
06:45 PM rue_mohr: they make good 1:1 transformers
06:45 PM polprog: i have a common mode choke here too
06:45 PM rue_mohr: if used in push-pull
06:45 PM rue_mohr: yea
06:45 PM polprog: what do you want me to do with the wire
06:46 PM rue_mohr: just saying that, if your watching a movie, you could put 10 turns on them
06:46 PM polprog: :P
06:46 PM rue_mohr: 10 turns is a fine test winding
06:46 PM polprog: i guess i can prepare something for tomorrow
06:46 PM rue_mohr: later you can measure the inductance with it
06:46 PM rue_mohr: then you know how many turns to put on to get what you want
06:46 PM rue_mohr: I'd like to make a direct converter
06:47 PM rue_mohr: L = something * n^2
06:47 PM rue_mohr: maybe just L = K*n^2 for all it matters
06:48 PM rue_mohr: so with 10 turns K = L/n^2
06:48 PM rue_mohr: er K = L/100
06:48 PM rue_mohr: then n = {math here}
06:48 PM rue_mohr: n = sqrt(L/K)
06:49 PM rue_mohr: I need to check that, it all seems to simple at the moment
06:51 PM polprog: whoops
06:51 PM polprog: barely got 6
06:52 PM rue_mohr: ok, lets say I have a 5V power supply and a 40uH inductor
06:52 PM rue_mohr: polprog, as long as you know how many it is
06:52 PM rue_mohr: I = fn(v, L, t)
06:52 PM rue_mohr: didn't you write a program for this?
06:53 PM rue_mohr: ./rclib --help
06:53 PM rue_mohr: ./rclib: invalid option -- '-'
06:53 PM rue_mohr: ./rclib: invalid option -- 'h'
06:53 PM rue_mohr: ./rclib: invalid option -- 'e'
06:53 PM rue_mohr: ./rclib: invalid option -- 'l'
06:53 PM rue_mohr: ./rclib: invalid option -- 'p'
06:53 PM rue_mohr: ah, "no"
06:54 PM rue_mohr: i(t) = fn(L, V, t)
06:54 PM * rue_mohr flips thru his little black book
06:56 PM rue_mohr: I = v*t/L
06:56 PM rue_mohr: so, for I to hit, 100mA
06:57 PM rue_mohr: t = I*L/v
06:57 PM rue_mohr: l = 40/1000000
06:58 PM rue_mohr: 8e-something
06:58 PM rue_mohr: .8us
06:59 PM rue_mohr: if this was done at 1V
06:59 PM polprog: this vinyl tape i got at a store nearby is pretty neat
06:59 PM rue_mohr: 4us
07:00 PM rue_mohr: so, idealy, the tester should have a low voltage supply
07:02 PM polprog: this CDrom has belts problem and i dont care to fix it, i have a plethora of them
07:02 PM polprog: MT1155PE
07:02 PM polprog: main cntrl
07:03 PM polprog: 25f1024 soic8 flash?
07:03 PM rue_mohr: 25, serial?
07:03 PM rue_mohr: yea
07:03 PM rue_mohr: I think multi channel serial, 1M
07:03 PM polprog: heh, the pcb says RXD TXD 3v3 on the bottom silk
07:03 PM rue_mohr: ooo
07:04 PM rue_mohr: flyback knows how to talk to them
07:04 PM polprog: who?
07:04 PM rue_mohr: yea
07:04 PM polprog: never met him
07:04 PM polprog: or her
07:05 PM rue_mohr: uh, troll major of freenode?
07:05 PM polprog: rings a bell but -_-
07:05 PM polprog: lets try to read it
07:06 PM polprog: maybe for one the soic clip wll be able to read something in circuit
07:08 PM rue_mohr: hmm
07:09 PM polprog: yesterday's hacks https://polprog.net/rozne1/ircjunk/various/packageonpackage.jpg
07:09 PM rue_mohr: if I'm charging an inductor at only 0.7V, how do I detect current without tearing off a significant percentage of the votlage
07:09 PM rue_mohr: I ownder if a hall is fast enough
07:09 PM polprog: u = di/dt
07:09 PM rue_mohr: 4us?
07:10 PM polprog: have a steady di from a vccs
07:10 PM polprog: then its sorta voltage independent
07:10 PM polprog: no wait
07:10 PM polprog: ugh
07:10 PM polprog: i should think more
07:10 PM polprog: l = di/dt...
07:10 PM rue_mohr: heh
07:10 PM polprog: ^H^H^H^H^H^H
07:11 PM polprog: overcurrent protection
07:11 PM polprog: what is the clip for even
07:13 PM rue_mohr: it looks like .1 ohms would not really cause much loss
07:13 PM rue_mohr: gonna go play with that...
07:14 PM rue_mohr: and take phone apart
07:14 PM darkrose: rue_mohr: biggest problem with canada is the neighbours
07:14 PM rue_mohr: its ok, they got a new leader
07:15 PM polprog: hi darkrose
07:16 PM darkrose: hi
07:18 PM polprog: how are you?
07:19 PM rue_shop1: darkrose, actaully, if the economy over here tanks hard fast, you might have a dollar advanatge that works really nicely
07:19 PM rue_shop1: but it would be a window
07:23 PM polprog: ok, so the in circuit programming problem seems to be that the power rails are shorted to ground when off
07:23 PM polprog: and when the thing is powered up the rom is accessed anyway so no dice
07:23 PM polprog: still idk when is the soic clip useful
07:39 PM polprog: goodnight
07:51 PM rue_shop1: ok, so its feeding this test inductor with about 0.754V
07:52 PM rue_shop1: at 30us, the current is... about...
07:53 PM rue_shop1: 120mA
07:55 PM rue_shop1: t = I*L/v
07:55 PM rue_shop1: L = t*v/I
07:56 PM rue_shop1: 188uH
07:57 PM rue_shop1: and I got a measurement of ~200uH from before
07:57 PM rue_shop1: thats close enough for me
08:01 PM rue_shop1: 8mA 300ms
08:01 PM rue_shop1: 300us
08:02 PM rue_shop1: 2.8mH
08:02 PM rue_shop1: I had it rated for 2
08:38 PM rue_mohr: well, seeing the saturation info was interesting
09:06 PM Tom_L: https://www.mikestools.com/Sherline/Sherline%20Accessories/Sherline_3074a.jpg
09:06 PM Tom_L: supposed to be one of those
09:06 PM Tom_L: threads fit nice though
09:07 PM Tom_L: first try on the small one was a fit
09:07 PM Tom_L: large one i
09:07 PM Tom_L: large one i'd already done before
09:15 PM rue_shop1: could it just be the alignment of you flipping it?
09:16 PM Tom_L: likely
09:16 PM rue_shop1: I think you should try it in one mounting
09:16 PM rue_shop1: female thread them both
09:16 PM rue_shop1: use a stub for the chuck
09:16 PM rue_shop1: but can your thread cutter reach past the large thread
09:16 PM Tom_L: probably
09:56 PM rue_mohr: hmm, I wonder what my resolutions where
10:12 PM Tom_L: better figger em out, that day is soon approaching
11:13 PM rue_mohr: http://tom-itx.no-ip.biz:81/~tom-itx/irc/logs/%23garfield/2019-12-25.html
11:13 PM rue_mohr: Merry christmas!
11:14 PM Tom_L: not for much longer
11:17 PM rue_mohr: haha
11:17 PM rue_mohr: its been a year since I got the keypad working for the water dispencer
11:17 PM rue_mohr: last week I got the water meter on it to be consistant
11:19 PM Tom_L: save something for next year..
11:19 PM rue_mohr: ooo got the tek scope just 1 year ago
11:21 PM rue_mohr: "05:00 PM rue_mohr: my goals for this year are to upgrade the shop drillpress, finish the large cnc (hahaha "finish") and get the mecha rings assembled "
11:21 PM rue_mohr: http://tom-itx.no-ip.biz:81/~tom-itx/irc/logs/%23garfield/2020-01-01.html
11:21 PM rue_mohr: the drillpress is upgraded
11:21 PM rue_mohr: no work on the other stuff
11:21 PM rue_mohr: this week I been trying to clear space to get the cnc on the deck
11:22 PM Tom_L: oh the big one?
11:26 PM rue_mohr: yea
11:26 PM rue_mohr: I did a tiny bit for it, not much
11:28 PM Tom_L: i used the numbers from that site yesterday for that 3/8 thread and it was spot on first time
11:28 PM rue_mohr: nice
11:47 PM rue_mohr: ok I got the phones stripped down, the deck is getting closer
11:47 PM rue_mohr: I could squeeze the machine on there now
11:48 PM rue_mohr: hoping moon comes by and can give me a hand getting it up there