#avr Logs

Sep 10 2017

12:24 AM day is now known as daey
07:12 AM Emil: tpw_rules: the algo is the subtractio one iir
07:12 AM Emil: iirc*
07:13 AM Emil: successive subtraction until result is smaller than modulo
08:01 AM HelloShitty is now known as HelloKitty
08:02 AM HelloKitty is now known as Guest30696
08:03 AM Guest30696 is now known as HelloShitty
08:07 AM HelloShitty is now known as DaRkV0rt3X
09:12 AM tpw_rules: Emil: really?! if so that's absolutely ridiculous
09:12 AM tpw_rules: unless they do it by tens or something
10:37 AM andrzejku: hi
10:37 AM andrzejku: :)
10:45 AM polprog: hi
11:06 AM Emil: tpw_rules: it's not ridiculous
11:06 AM Emil: tpw_rules: the algo is very small and applicable to every single parameter
11:07 AM Emil: It's not the fastest, true, but you should know better than use modulo operators ;)
11:14 AM enh: hi
11:14 AM Emil: Hi
11:22 AM day is now known as daey
11:54 AM enh: https://imagebin.ca/v/3ZvPf3wFSk1a
11:55 AM enh: Telemetry and the module capacity matrix.
11:55 AM enh: Telemetries start with ### and have longs separated by #
11:56 AM enh: The capacity matrix has capacities on lines and slots on columns. In this case capacity 9 is accel sensor, 10 means gyroscope, 11 means magnetometer and 12 means barometer
11:57 AM enh: This means that slot #4 has all these sensors.
11:58 AM enh: This is the current capacity list: https://pastebin.com/ydBwCSqQ
11:58 AM enh: Amazing!
11:58 AM enh: Nice to see these things working
12:00 PM Emil: enh: hmm
12:00 PM Emil: enh: what does TOTAL_CAPACITIES mean?
12:00 PM enh: Is is a refernce for the size of the capacity list
12:01 PM Emil: ummm
12:01 PM enh: When I loop through them on a for loop, I use that number to know where the end is
12:01 PM Emil: enh: enums are enums
12:01 PM Emil: ah
12:01 PM Emil: that's umm
12:01 PM Emil: well it works :D
12:01 PM enh: it does
12:01 PM Emil: But it's kluuuudge as fuck
12:02 PM enh: what does kludge mean? Sorry for the ignorance
12:02 PM Emil: https://en.wikipedia.org/wiki/Kludge
12:02 PM Emil: workaround or quick-and-dirty solution that is clumsy, inelegant, inefficient, difficult to extend and hard to maintain.
12:02 PM Emil: hacky
12:02 PM enh: I like it.
12:02 PM enh: Clear and simple. Easy to maintain.
12:03 PM Emil: People often like themselves
12:03 PM enh: You do.
12:03 PM enh: I do not like myself too much.
12:03 PM enh: I'm sure you could make this much simpler.
12:05 PM enh: But what I like in this solution is that it is coherent though all the code.
12:05 PM enh: A few enums and a data array holds all the system data and transfer it between modules without the need for protocol translations
12:06 PM Emil: If you like what you do/did you like yourself by proxy
12:07 PM enh: This statement can be wrong. I'm sure you know it can.
12:07 PM enh: But ok. I'm happy because the thing is moving forward.
12:08 PM enh: The nice thing is that by using that matrix the system knows which slots have the data it has to fuse.
12:08 PM enh: By using a simple iteration it takes care of module redundancy
12:11 PM enh: What would you use instead of enums, Emil?
12:35 PM enh: https://hackaday.io/project/11724-yauvec-yet-another-unmanned-vehicle-controller/log/66902-telemetry-and-module-capacity-matrix
12:45 PM polprog: People from that channel are very nice and have contributed with solutions to programming problems and hardware related ones.
12:45 PM polprog: <3
12:45 PM polprog: im not sure if i ever helped but that's nice :)
12:47 PM Emil: polprog: which channel?
12:47 PM polprog: also, https://hackaday.io/project/11724/gallery#14e328078c594e1efd2384b01632491a is this an old fluke?
12:47 PM polprog: Emil: it's taken from enh's post
12:52 PM enh: It is true.
12:52 PM enh: You guys help a lot
12:55 PM Emil: polprog: it's referencing #sparkfun ;)
12:55 PM Emil: polprog: no glory for us
12:55 PM Emil: ;9
12:55 PM Emil: ;)
12:56 PM enh: and #avr
03:09 PM polprog: not good.. nobody on the market i go to get quick small resupplies sells atmegas, today the guy only had one m8 in smd...
03:10 PM HighInBC: this is why I have a small bag of them
03:10 PM HighInBC: and a bag of attiny85s too
03:10 PM polprog: i try to keep some spares all the time
03:10 PM polprog: i have a ton of m16
03:11 PM HighInBC: there is never enough it seems
03:11 PM polprog: but not that much m8...
03:11 PM polprog: that's true :D
03:53 PM polprog: can female pin header connectors be crimped without special tools? like those: https://www.buildyourcnc.com/images/3position2.54pitchfemaleheader-300.JPG
03:59 PM Tom_itx: i use one i got from RS
03:59 PM Tom_itx: for like 10 bux
03:59 PM thardin: it won't look pretty if you use the wrong tool
04:38 PM HighInBC: polprog: it can be but it is a pain, and it is hard to tell if it is a good crimp or not so I always touch it up with a little solder
05:34 PM Tom_itx is now known as Tom_L
05:50 PM Lambda_Aurigae: polprog, I do it with needle nose pliers all the time.
07:03 PM enh: I will *never more* put a via below a sensor.
07:04 PM enh: Never ever
07:06 PM cehteh: heh
07:07 PM cehteh: groundplane ftw
07:15 PM Lambda_Aurigae: enh, cause issues did it?
07:24 PM enh: I live near the sea. sometimes bad solders between smd parts and PCB work because there is a residual mechanical contact between them. But near the sea, moisture and salt get in the middle of that and create an insulating layer. This only appears with time here.
07:25 PM Lambda_Aurigae: been there, done that.
07:25 PM enh: I tried reflowing the sensor, but the via below it started to push the IC up. Only three sides of the sensor were effectively touching the PCB.
07:25 PM Lambda_Aurigae: used to live in florida.
07:25 PM Lambda_Aurigae: flippin giant sandbar.
07:25 PM enh: I had to put a lot of solder paste to reflow that thing
07:25 PM enh: Now I have all the sensors online again.
07:27 PM enh: https://imagebin.ca/v/3ZxcLHiItTzb
07:27 PM enh: Florida must be even worse than here. Almost surrounded by salt water
07:28 PM enh: I 'll eat something and come back soon
07:28 PM Lambda_Aurigae: there's salt in the ground in florida...plants have evolved to live with the salt it seems.
07:48 PM Jartza: well. I had to implement fix into firmware to prevent customer's products from rusting :)
07:48 PM Jartza: so it happens
08:02 PM enh: Jartza: Sorry. How can you prevent rusting by implementing something on the firmware?
08:06 PM Tom_L: dc current
08:06 PM enh: Lambda_Aurigae: I live in the extreme east of south america. A city named João Pessoa, in the nearest point from Africa. There is sea in half the horizon here. Amazing how bad solders show up months after assembling the PCB. It must be the salt eating the contacts, or just oxides forming between almost good contacts.
08:07 PM enh: Tom_L: really? You mean keeping the tracks polarized?
08:07 PM Tom_L: was more of a guess
08:08 PM enh: My father used to say that keeping the equipments on and warm prevented rust
08:09 PM Tom_L: i run my furnace pilot year round
08:09 PM enh: good practice
08:10 PM enh: at least you prevent mechanical stresses on the heaters
08:15 PM Jartza: enh: well, it's a long story and can't really release the specifics, but eventually ended up doing some fancy PWM:ing
08:17 PM Tom_L: enh just like they put sacrificial metal on ships to prevent rust
08:17 PM Jartza: it had to do with batteries charging and such
08:18 PM Tom_L: sp/just like/similar to
08:18 PM enh: interesting ideas...
08:19 PM Tom_L: https://en.wikipedia.org/wiki/Galvanic_anode
08:20 PM enh: I read about that some time ago, Tom_L. Ions prefer a lot those anodes than anything around.
08:20 PM enh: But PWMing for rust prevention is interesting...
08:22 PM enh: oxygen os a symmetical dipole, but water is not.
08:23 PM enh: I worked on an experiment long ago that could reduce the etching rate of hydrofluoric acid on glass by recording a high electric field inside the glass.
08:25 PM enh: that electric field, repelled the fluor and attracted hydrogen. It increased the probability of the glass being hit by the hydrogen side of the molecule, leaving the active part far away from glass and reducing the reaction probability.
08:27 PM enh: Whatever the specifics are, you really should not release without filing a patent, Jartza
08:29 PM enh: Anybody here ever heard of Royal Raymond Rife?
08:29 PM enh: That is a subject worth studying.
08:30 PM enh: Destroying patogens with oscillating electric fields, or whatever.
08:31 PM enh: Detecting them too.
08:33 PM enh: Based on his work, you could make a person walk through a corridor where his pathogens would be detected in the beginning of the corridor and eliminated in the end. The person would leave the corridor without bad viruses or bacteria.
08:33 PM enh: And that includes cancer
08:34 PM enh: http://rife.org
10:30 PM enh: This issue is sooo annoying. I reflowed everything in my AVR PCB until I desisted and searched google for the AVRDUDE complain. Found this: http://www.avrfreaks.net/forum/avrdude-problems-stk500v2command-error
10:30 PM enh: Changed 3 ICS, all capacitors, the LED resistor and the crystal, to discover it was a speed problem.
10:31 PM enh: Learning is, sometimes, so painful...
10:31 PM Casper: enh: liar! learning is always joyfull and pleasant!
10:38 PM Thrashbarg: heh
10:39 PM enh: :/
11:19 PM NAVIL: Hey