#garfield Logs

Apr 24 2021

12:14 AM rue_bed2: I feld asleep
12:21 AM rue_bed2: I'm gonna work on my debug led
01:06 AM rue_mohr: ok, debug led and then serial port
01:07 AM rue_mohr: then everything else
01:46 AM rue_mohr: I'm gonna babble in #robotics for a while
06:55 AM rue_mohr: ok!
06:55 AM rue_mohr: so the new lcd functions work
06:58 AM rue_mohr: there is an init screen, send clear command, blit screen and position cursor
06:58 AM rue_mohr: should be all the io I need with the screen
06:58 AM rue_mohr: oh except custom characters, want to set something up for that
07:01 AM rue_mohr: and there is screen busy check
08:04 AM Tom_L: ur up early
12:01 PM rue_mohr: just woke up
02:32 PM rue_mohr: OK
02:32 PM rue_mohr: so I'll do up some recycling and stuff
02:33 PM rue_mohr: the scolling lcd is a milestone
02:33 PM rue_mohr: I can go thru all the old notes, collect togethor all the code that it replaces
03:48 PM aandrew: so
03:48 PM aandrew: this fucking st25r3911 (NFC chip) has a goddamned active-high, push-pull IRQ output
03:48 PM aandrew: seriously what the fucking fuck
04:41 PM rue_shop1: ? output?
04:41 PM rue_shop1: high?
04:47 PM aandrew: yes, when the chip wants to signal the MCU it sets the IRQ pin high
04:47 PM aandrew: and it's push-pull too
04:53 PM rue_shop1: 2N7000 makes it active low open collector :)
05:27 PM aandrew: yep that is *exactly* what I did, even the same part
05:28 PM aandrew: just frustrates me that a) they implemented an IRQ pin that way and b) I didn't catch it
05:41 PM c-log is now known as Guest25509
05:56 PM polprog: i caught my siggen generating signals it should not be generating
06:16 PM rue_mohr: ok, recycling done
06:16 PM rue_mohr: no time to go up to the landfill today
06:17 PM rue_mohr: its open tommorow
06:17 PM rue_mohr: polprog, yours yours or -brand- yours?
06:18 PM polprog: my off the shelf one
06:18 PM polprog: not homemade
06:18 PM polprog: the 6MHz voltcraft i got the other day
06:18 PM polprog: it's generating some emi at 580ish MHz :c
06:18 PM polprog: the spectrum shape is intersting, and it depends on the set frequency
06:19 PM polprog: it looks like mixing
06:25 PM polprog: friend suggests it might be parasitics oscillating
06:35 PM rue_mohr: cool
06:35 PM rue_mohr: how many db?
06:35 PM polprog: 10-15 above noise level
06:35 PM rue_mohr: oof
06:35 PM polprog: so not the end of the world
06:35 PM polprog: https://cdn.discordapp.com/attachments/823223032822104094/835638099303464980/IMG_20210424_235638.jpg
06:35 PM rue_mohr: I wonder if they didn't latch the binary before the dac
06:35 PM polprog: whos they
06:36 PM rue_mohr: the makers of it
06:36 PM polprog: noo its fully analog
06:36 PM rue_mohr: does it use a resistor based dac?
06:36 PM rue_mohr: what?
06:36 PM rue_mohr: wow
06:36 PM polprog: sin, square and triangle is all you get
06:36 PM rue_mohr: I was going to use mixing for mine
06:36 PM polprog: the LCD display is just a counter, you cant type in a frequency
06:36 PM rue_mohr: its more logical to generate a sine wave, then turn that into square and others than to go the other way
06:37 PM polprog: yeah. Maybe that's what they are doing
06:37 PM rue_mohr: rf oscillators are quite stable
06:37 PM polprog: which would explain the lack of emi on the other two shapes
06:37 PM polprog: it just doesnt pass thru the circuits
06:38 PM polprog: so i was wondering. The TV RF mod seems to be using an oscillator module to generate the RF signal
06:38 PM polprog: its tunable between 530 and 630 MHz
06:38 PM rue_mohr: :) yea
06:38 PM polprog: which is not in any ISM band :(
06:38 PM rue_mohr: thats 100Mhz
06:38 PM polprog: yeah. impressive!
06:38 PM rue_mohr: if you subtract 530Mhz with a mixer, you have 0-100Mhz
06:39 PM polprog: i wonder if i can modify it to move it to the 433 MHz band
06:39 PM polprog: hm, i was thinking about a mixer
06:39 PM rue_mohr: last time I looked they had too many tuned stages
06:39 PM polprog: a 100MHz mixer would give me 430-530 + 630-730 ?
06:39 PM polprog: yhh
06:39 PM polprog: a mixer + 100MHz signal*
06:40 PM polprog: now my plan is to go grab the bag of demodulators and try to make a scanner out of them
06:43 PM polprog: http://www.rfcandy.biz/communication/supertuner.html
06:43 PM rue_mohr: and then you filter off the high end
06:44 PM polprog: yeah
06:44 PM polprog: i need to make a 12->33V booster
06:44 PM rue_mohr: thats so easy
06:44 PM rue_mohr: for me anyhow
06:44 PM polprog: :P
06:44 PM rue_mohr: srry
06:44 PM rue_mohr: I dont know why they wont work for you
06:44 PM polprog: ive got some buck IC's, ill try to make a boost converter with one
06:45 PM polprog: i have to try them with these new cores
06:45 PM polprog: ive got a maths exam on monday so i cant play much
06:45 PM rue_mohr: 1 pn2222, 1 resistor, 1 transformer, 2 diodes
06:45 PM rue_mohr: 1 cap
06:45 PM rue_mohr: 1:3 ratio
06:45 PM polprog: do you have like a uni in vc?
06:45 PM rue_mohr: hu?
06:46 PM rue_mohr: a degree in voltage conversion?
06:46 PM polprog: i wish my uni was over so id have more free time
06:46 PM polprog: do you have a university in vancouver
06:46 PM polprog: :p
06:46 PM rue_mohr: oh a bunch of "universities" but they are just money suckers for fools
06:47 PM polprog: ill tell everyone im doing research in boost converters, they will pay for my ticket to vancouver and we can play
06:48 PM rue_mohr: heh
06:48 PM rue_mohr: exchange student
06:48 PM polprog: :)
06:49 PM rue_mohr: do you have that last one I drew up for you?
06:49 PM polprog: then you come to warsaw and we play with lazers
06:49 PM rue_mohr: heh
06:49 PM rue_mohr: use real equip and build the laser xor gate
06:49 PM polprog: that would be cool
06:49 PM rue_mohr: bring computing up to the THz it should be at
06:50 PM rue_mohr: you know you can make any gate with or and xor, right?
06:50 PM polprog: yes
06:50 PM polprog: i think you can do everything with nand
06:50 PM rue_mohr: optics like or and xor
06:50 PM rue_mohr: esp monochromatic
06:51 PM polprog: yeah
06:53 PM rue_mohr: so whats the propigation delay for a 1km roll of fiber?
06:53 PM polprog: hmm..
06:53 PM polprog: idk. depends on which mode
06:53 PM rue_mohr: totally single
06:53 PM polprog: its not the delay thats the problem. its the dispersion
06:53 PM polprog: i guess it's gonna be.. 0,75c?
06:53 PM rue_mohr: delay isn't a problem at all, its memory
06:54 PM polprog: you take the index of refraction and use that
06:54 PM rue_mohr: question is, at say, 2THz, how much memory is it?
06:54 PM polprog: i think it's n = c/v
06:55 PM polprog: im guessing that velocity because were halway thru the undergrad and they havent told us that
06:55 PM rue_mohr: sure
06:56 PM polprog: but 1st year was just all the basics at once
06:56 PM rue_mohr: so I wonder if I should attack 32k rom or serial I/F first
06:56 PM polprog: id say serial so you can have a bigger display on the computer
06:57 PM rue_mohr: units 1km/0.8c us
06:57 PM rue_mohr: * 4.1695512
06:57 PM rue_mohr: units 2THz*4.16us
06:57 PM rue_mohr: Definition: 8320000
06:57 PM rue_mohr: 8.3Mbits
06:58 PM rue_mohr: 1Mbyte
06:58 PM polprog: hmmm
06:58 PM polprog: delay line memory?
06:58 PM rue_mohr: lesson: run the freaking optic computer faster than 2THz!!!
06:58 PM rue_mohr: yup
06:58 PM rue_mohr: what else you gonna use on a THz core?
06:58 PM polprog: heh, true
06:59 PM rue_mohr: so
06:59 PM rue_mohr: an optic XOR based blocking oscillator
07:00 PM rue_mohr: employing a delayed feedback path at 0.8c
07:00 PM rue_mohr: 2THz, whats the length of the feedback path?
07:00 PM rue_mohr: few cm?
07:00 PM polprog: dunno
07:01 PM rue_mohr: hmm I think its 0.12mm
07:02 PM rue_mohr: kinda tight
07:03 PM polprog: i have to adjust the timebase in the analog scope
07:03 PM rue_mohr: it happens
07:03 PM rue_mohr: can you do it per range?
07:03 PM polprog: ive gotta read the manual
07:03 PM polprog: it sits just below the SA, so i can use a short BNC to connect the IF output from the analyzer to the scope
07:04 PM polprog: that would make a nice setup :)
07:06 PM polprog: why didnt i notice its off though
07:06 PM polprog: its off by like 20%
07:06 PM polprog: a 1khz signal is like 4/5 of a div :<
07:07 PM polprog: otherwise its a super cool 100mhz analog scope
07:07 PM rue_mohr: ok, first I'll confirm that the 8251 can hardware-issue a DTR signal
07:08 PM rue_mohr: DTR operation has never worked for me
07:09 PM polprog: ive always used that as gpio
07:10 PM rue_mohr: heh
07:10 PM rue_mohr: on that 8250 it is,
07:10 PM rue_mohr: there is no automatic operation of the handshake lines
07:10 PM rue_mohr: I'd like my z80 to be able to handle 115k
07:10 PM rue_mohr: even tho it cant take that rate
07:11 PM rue_mohr: I have to use an advanced version of the 8251 to get up that high
07:13 PM polprog: i should play with serial on the 8051
07:13 PM polprog: goodnight for now
07:14 PM rue_mohr: gn
07:18 PM rue_mohr: ok I need the 8253 anyhow, I'll start with that
10:31 PM aandrew: https://www.reddit.com/r/Whatcouldgowrong/comments/mxszst/nice_parking/ tht's so bad... so so so so so bad
10:31 PM aandrew: my god
10:48 PM rue_shop1: oh yea I was just waiting for the spriklers to run out of air
10:48 PM rue_shop1: that'd be the fire alarm in full swing
10:48 PM rue_shop1: its ok, the fire department is on the way
10:52 PM aandrew: you know the "how'd you get the beans up on top of the frank" scene in There's Something About Mary, where the police, fire, etc all come to laugh at the guy?
10:52 PM aandrew: that's waht I see happening when the FD gets to this scene
10:54 PM aandrew: the other great line from that scene is when the gf's father calls his wife in and the guy is like "no!" and the dad says "oh don't worry, she's a dental hygenist, she'll know what to do."
11:09 PM rue_shop1: hmm, well "simply install a 8253 to generate a uart baud clock" isn't going too good
11:14 PM rue_shop1: wait wait wait, this says the 8250 has a 16 bit divier?
11:15 PM rue_shop1: oh there it is
11:25 PM aandrew: did hte LED at least go smoothly?
11:27 PM rue_shop1: hahah I took a detour somewhere, I dont want the 8250, I want the 8251!!!
11:27 PM rue_shop1: I got the led working
11:27 PM rue_shop1: back in 1994, for easy access, I moved the io chip to 0x5000 in memory space
11:28 PM rue_shop1: which there was a tiny comment about, it was supposed to be 0x1000 in io space
11:28 PM rue_shop1: ok 8251
11:31 PM rue_shop3: ok, the dividers are 1, 16, and 64
11:32 PM rue_shop3: so, for 115200, I need a max of
11:32 PM rue_shop3: 7.3Mhz
11:32 PM rue_shop3: ok
11:33 PM rue_shop3: I have an 8254 that can go up to 10Mhz