#robotics Logs

Dec 29 2019

#robotics Calendar

12:28 AM zhanx: ok Super Slam
12:34 AM zhanx: using the leg to map the area, based off leg location
02:14 AM Jak_o_Shadows: ah, dammit.
02:14 AM Jak_o_Shadows: I thouht this chip had a compass
03:05 AM Jak_o_Shadows: Hmm. THis whole sending multiple values to the servos only really works if the calibration is on the uC
03:52 AM rue_mohr: ?
03:53 AM rue_mohr: you want my offset/scale library?
03:53 AM rue_mohr: its PC side
03:53 AM rue_mohr: but an stm32 could prolly do ok
03:55 AM Jak_o_Shadows: nah, not at the moment.
03:55 AM Jak_o_Shadows: It'd have to be STM32 side to take advantage of setting multiple servos to the same value
03:57 AM rue_mohr: no, I translate before I send the commands
03:57 AM rue_mohr: position file in robot degrees -> scale translate-> wire -> controller ->servos
03:58 AM rue_mohr: all my robots have a file with the scale/offset for the servo numbers to make them into robot degrees
03:58 AM rue_mohr: my code can do linear value mixing
03:59 AM rue_mohr: so if you have axis crossover you can do the basic math to seperate them
03:59 AM rue_mohr: or merge them
03:59 AM rue_mohr: thats how I deal with the wrist motion, dispite the wrists being built differently
04:05 AM Jak_o_Shadows: But then you can't take advantage of multiple servos listening
04:05 AM Jak_o_Shadows: side note: this one stupid servo isn't liking the 3.3v pulses from my chip
04:05 AM Jak_o_Shadows: whereas the rest do
04:06 AM Jak_o_Shadows: this particular one only likes the 5V from my little servo tester
04:06 AM Jak_o_Shadows: The biggish chip (the non-motor driver one) only has a 2V pulse signal (when using the 3.3v)
04:08 AM Jak_o_Shadows: It's a right bugger
04:21 AM Jak_o_Shadows: I wonder if the i2c pins i'm using are 5V tolerant, and I can just run the whole thing form that
04:52 AM Jak_o_Shadows: Nope, not that I think
05:05 AM Jak_o_Shadows: bugger it, let's add some capacitance to the uC supply
06:23 AM chisight: or a drive transistor.
06:41 AM Jak_o_Shadows: I gave up and swapped servo
06:41 AM Jak_o_Shadows: Which is a pain - but I guess they're still vaguely useful for something
07:20 AM chisight: next design you know that only a percentage of servos work at 3.3v without assistance and can design the board to compensate.
12:48 PM Tom_itx is now known as Tom_L
01:27 PM rue_bed2: Jak_o_Shadows, yes and no, you will still have co-inncodental positions
11:04 PM rue_mohr: --
11:35 PM mrdata: ++