#robotics Logs

Aug 22 2020

12:21 AM rue_shop3: I had noodles for lunch
12:27 AM rue_shop3: at, like 10pm
12:52 AM Jak_o_Shadows: I was makng the robot micro-controller board, and power distribution, but then it rained
01:15 AM rue_shop3: ok...
01:15 AM rue_shop3: so, I have code that uses timer1 to divide 8Mhz by 623, but the result is 62.5Hz
01:15 AM rue_shop3: so
01:23 AM rue_shop3: I dont get it
01:26 AM rue_shop3: another chip does the same thing
01:26 AM rue_shop3: so its me
01:32 AM rue_shop3: huh
01:32 AM rue_shop3: works for phase correct pwm
01:35 AM rue_shop1: numbers still aren't right
01:35 AM rue_shop1: oh, its running at 20Mhz!?
01:36 AM rue_shop1: on internal!?
01:41 AM rue_shop1: ok, turning on clkout says 9.6Mhz
01:41 AM rue_shop1: so, wtf
01:46 AM rue_shop1: aha, giveaway, duty isn't 50%
01:50 AM rue_shop1: ok, phase correct is not the way to go
01:56 AM rue_shop1: ok, use timer overflow instead of compare
01:56 AM rue_shop1: should be the same, but ok
01:57 AM rue_shop1: wt
01:57 AM rue_shop1: wtf
01:59 AM rue_shop1: I can put any number is OCR1A and it makes no difference
02:01 AM rue_shop1: wtf
02:01 AM rue_shop1: nothing changes the interval of the interrupt
02:25 AM rue_shop1: no change to the top value on any of these will cause the interval to change
02:25 AM rue_shop1: W T F
02:25 AM rue_shop1: I switched to atmega8 to be sure its not the chips
03:00 AM Jak_o_Shadows: My calculations for the STM32 didn't quite match either
03:00 AM Jak_o_Shadows: so I just measured and did a few more counts
03:04 AM rue_shop3: this isn't 'not quite matching' this is WILL NOT CHANGE
03:05 AM rue_shop3: no matter what value, and which method I use to set the counter TOP, its having NO effect
03:05 AM rue_shop3: I had to resort to presetting the count when the interrupt is called
03:05 AM rue_shop3: the avrs (atmega8 and atmega88) WILL NOT change the count cycle
03:05 AM rue_shop3: I really dont get it
03:06 AM rue_shop3: I"v just taken like 4 hours to get a pin doing 6.4kHz on interrupts
03:06 AM rue_shop3: I need food
03:35 AM veverak: I recently got interested into attiny85
03:35 AM veverak: shit seems cheap and capable
04:03 AM rue_bed2: wont say cheap
04:03 AM rue_bed2: and, for the io space, a tiny26 beats a tiny13
04:03 AM rue_bed2: its 2am
04:03 AM rue_bed2: I been up sinnce 6am
04:03 AM rue_bed2: I should go to sleep
05:31 AM kerfluffy: anyone know how a blood pressure monitor works when it deflates to find the blood pressure? there seems to be just one solenoid inside, but once it's found both the diastolic and systole blood pressure, it deflates much faster. here's a teardown https://youtu.be/8-dEvfMvQc0?t=151
05:31 AM kerfluffy: how is it able to have two deflation speeds?
06:15 AM Jak_o_Shadows: dual position?
06:15 AM Jak_o_Shadows: well, three
11:52 AM Miyu is now known as hackkitten
01:00 PM SpeedEvil is now known as Guest22568
02:16 PM BitEvil is now known as SpeedEvil
09:21 PM rue_mohr: probably two vent solinoids
09:21 PM rue_mohr: different size oriphicees
09:32 PM Jak_o_Shadows: Is there such a thing as tri-position solenoids?
09:33 PM zhanx_workshop: jak for boilers yes but not sure what you need
11:13 PM Jak_o_Shadows: kinda answering kerfluffy's question from earlier
11:13 PM Jak_o_Shadows: About how a blood pressure monitor has two deflation rates with only a single solenoid
11:14 PM kerfluffy: could the pump itself leak once it stops?
11:14 PM kerfluffy: maybe that's what drives the slow deflation rate?
11:18 PM zhanx_workshop: jak pwm