a thread about janitoring individual bits and using oscilloscopes and stuff

The Something Awful Forums > Discussion > Serious Hardware/Software Crap > YOSPOS > a thread about janitoring individual bits and using oscilloscopes and stuff

«‹›65 »

hobbesmaster: Jan 28, 2008

gonadic io posted:

got myself some rectifier diodes to put across the motors and solenoid valves

perfect inductors will provide whatever voltage is necessary to prevent an instantaneous change in current so even that isn't fool proof

# ? Feb 23, 2017 16:03

Adbot: ADBOT LOVES YOU

# ? May 16, 2024 06:46

hobbesmaster: Jan 28, 2008

my car's infotainment system hung on my way home

my first thought was "wheres the debug port"

# ? Feb 24, 2017 01:34

Cocoa Crispies: Jul 20, 2001; Vehicular Manslaughter!; Pillbug

hobbesmaster posted:

my car's infotainment system hung on my way home

my first thought was "wheres the debug port"

last firmware update for mine added a menu option to reboot for when it fucks up too much

# ? Feb 24, 2017 02:10

Silver Alicorn: Mar 30, 2008; 𝓪 𝓻𝓮𝓭 𝓹𝓪𝓷𝓭𝓪 𝓲𝓼 𝓪 𝓬𝓾𝓻𝓲𝓸𝓾𝓼 𝓼𝓸𝓻𝓽 𝓸𝓯 𝓬𝓻𝓮𝓪𝓽𝓾𝓻𝓮

my infotainment system has never failed. can only be updated at the dealership, if ever

# ? Feb 24, 2017 02:18

hobbesmaster: Jan 28, 2008

Silver Alicorn posted:

my infotainment system has never failed. can only be updated at the dealership, if ever

more like your car has a good watchdog

# ? Feb 24, 2017 02:25

Stabby McDamage: Dec 11, 2005; Doctor Rope

gonadic io posted:

dumb noob question

if i only have a 24V battery, and i use some DC/DC regulator to get it to 3.3

can i have
pre:
+24V ----- motor --|
       |           |
     DC/DC         |
       | (+3.3V)   |
     arduino       |
       |           |
 +0v --------------|
i.e. i don't need to protect the arduino from the ground rail, 0 is 0? the motor will have a rectifier diode to prevent kickback.

That's fine and normal. Further, unless you have some crazy moon arduino, its onboard DC regulator will be able to handle the 24V, so no need for the extra parts. This is actually the standard architecture for basic arduino robots: 9V+ DC into the arduino, arduino regulates to 5 or 3.3 for logic, and a motor shield regulates the motor outputs using the higher raw voltage.

# ? Feb 24, 2017 04:47

Bloody: Mar 3, 2013

the hell regulators do arduinos have to take 24 vdc rails lol

# ? Feb 24, 2017 05:25

Zopotantor: Feb 24, 2013; ...und ist er drin dann lassen wir ihn niemals wieder raus...

hobbesmaster posted:

perfect inductors will provide whatever voltage is necessary to prevent an instantaneous change in current so even that isn't fool proof

just use a perfect diode, duh

# ? Feb 24, 2017 06:35

gonadic io: Feb 16, 2011; >>=

Zero:

quote:

The board can operate on an external supply of 6 to 20 volts. The recommended range is 7 to 12 volts.

MKRZero:

quote:

Vin
This pin can be used to power the board with a regulated 5V source. If the power is fed through this pin, the USB power source is disconnected. This is the only way you can supply 5v (range is 5V to maximum 6V) to the board not using USB. This pin is an INPUT.

# ? Feb 24, 2017 07:52

Harik: Sep 9, 2001; From the hard streets of Moscow
First dog to touch the stars; Plaster Town Cop

We use +60v regulators for vehicle operations, since it can be used on the usual 12v (really 13.8) but there's 2-battery (24v) and even 4-battery (48v!) systems out there.

gonadic io posted:

https://www.arduino.cc/en/Tutorial/Potentiometer

why this tutorial say to connect 1 pin with ground, one with the board, and then 1 to +V?

surely the same effect is achieved with enabling the internal pullup resistor, then connecting it to the middle pin then connecting ground to either of the side ones?

That would actually work, sort-of. The downside is your max voltage depends entirely on the exact resistance of your internal pullup - those are generally never specified or consistent.

Colonel Taint posted:

Pro tips from an actual IC designer:

If you don't have many registers, you can make I2C registers hold 9 bits! Just use the least significant of the register address as the 9th bit of the data, and let people multiply the listed address by 2. It's totally intuitive and people will love you for it!

Don't waste time with pesky logic that lets you write back onto the I2C bus. Users only want to set registers and won't ever want to read them back. Think of all the silicon you'll save!

You mean all I2C devices don't have a single write byte that decodes like:
b1xxxxxxx "Turn on"
b011xxxxY "Set foo flag to Y"
b010xxxxY "Set bar flag to Y"
b0010YYYY "Set Baz to YYYY"
....
b00000001 "Halt, catch fire"

?

Because that's the majority of the ones I've worked with!

Also fun story: I started using xmegas because some shittastic hardware we needed to drive used 9 bit 1-wire serial. That sucked, but at least the xmega USART supports 9bit TX/RX and could tristate the TX line without disabling RX so I just tied the lines together and ignored what I transmitted.

Storysmith posted:

look at how guitar foot pedals are done, you can either go geared potentiometer or led blocking wedge into the middle of a homemade vactrol in that size easily

you'll need to rig a return spring obvs and a kill switch on the dash

If I can't find a junker ~~guitar~~ keyboard pedal or similar that fits I'll 3d-print something like that. The blocking wedge sounds like the way to go but I'm not sure I can make the print clean enough to have it work properly. The ones I work with are pretty narrow-beam and the slightest defect in the print would gently caress them up. E: Keyboard pedals are nothing but pots, that's probably perfect for what I'm doing.

Since it's two-independent motors I'll probably end up putting another analog input on the steering wheel so I can run the wheels at slightly different speeds and help it turn less like a drunken cow.

Sadly, I've gotten jack and poo poo done on this so far aside from extracting all the electronics and setting it up in mockup form. Next step is just finding some cheap 20A ESC / brushless motor combos to test this with. I can use my normal RC receiver with a signal splitter to drive them in unison and see if the project is even going to work before I go all-in.

# ? Feb 24, 2017 13:17

Stabby McDamage: Dec 11, 2005; Doctor Rope

Bloody posted:

the hell regulators do arduinos have to take 24 vdc rails lol

I believe they have 7805 linear regulators, which can handle up to 35V. Now, whether dropping 30V as heat is feasible or not depends on how much 5V load you have, but if you aren't doing a bunch of LEDs or something, it may just work. Maybe just put a lil baby heatsink on it.

Some hobby motor controllers actually have 5V regulation on board, like this cheap L298 module, but I think that's still a linear regulator, so if you eat a bunch of 5V, it's gonna get hot.

If you want to actually do it right, you'd of course use a switching regulator. You can get cheap all-in-one modules for that like this one.

# ? Feb 25, 2017 00:23

Bloody: Mar 3, 2013

a pro tier redesign would have a buck-boost for maximally broad input range from like 1 to 48 volts

and it would probably cost more than an arduio

# ? Feb 25, 2017 00:25

Stabby McDamage: Dec 11, 2005; Doctor Rope

Bloody posted:

a pro tier redesign would have a buck-boost for maximally broad input range from like 1 to 48 volts

and it would probably cost more than an arduio

Not if China has anything to say about it.

# ? Feb 25, 2017 03:43

hobbesmaster: Jan 28, 2008

Stabby McDamage posted:

Not if China has anything to say about it.

hmmm a random Chinese board off eBay rated for 800mA at 76% efficiency

I'm going with it blowing up at 300mA

# ? Feb 25, 2017 06:55

atomicthumbs: Dec 26, 2010; We're in the business of extending man's senses.

i'm going with it having a switching frequency so low and producing so much noise the arduino just reboots forever

# ? Feb 25, 2017 07:12

Bloody: Mar 3, 2013

2-15 doesn't even get you to 24 which is an important level imo

# ? Feb 25, 2017 07:36

Jimmy Carter: Nov 3, 2005; THIS MOTHERDUCKER
FLIES IN STYLE

so is there a decent central resource for finding out about ICs are available on the market for a given task or is it just punching stuff into DigiKey and seeing what comes up? Someplace I can ask if I want, like, a non-poo poo boost converter that can handle like a third of an amp at 3.3v

# ? Feb 27, 2017 05:37

hobbesmaster: Jan 28, 2008

in general for prototyping just look on sparkfun or adafruit for a breakout

are you looking for something like this? https://www.adafruit.com/products/2190
or this: https://www.sparkfun.com/products/10967

like, 3.3V in or out

# ? Feb 27, 2017 05:43

Bloody: Mar 3, 2013

pmuch either digikey or $preferred-parts-manufacturer idk of any useful multi vendor trade study tools at a higher level. power supplies I mostly just check lt first in part because ltspice is good and free

# ? Feb 27, 2017 05:45

Hunter2 Thompson: Feb 3, 2005; Ramrod XTreme

The company where I work just has "contacts" aka sales engineers assigned to us by the various large manufacturers who frequently get emailed. If they don't have what we need at a price we want the EEs will search and send emails to rando Taiwanese companies and Chinese companies to get samples and dev boards.

# ? Feb 27, 2017 05:56

Jimmy Carter: Nov 3, 2005; THIS MOTHERDUCKER
FLIES IN STYLE

3.3v out but yeah it looks like Adafruit/Sparkfun are good starting points for parts/schematics for the stupid poo poo I want to do

# ? Feb 27, 2017 05:57

big shtick energy: May 27, 2004

the forums at realworldtech used to be real good, neckbeards having flamewars about poo poo like how wide an instruction decoder should be and poo poo like that. you could learn a lot

But recently they've been getting trolled by idiots and white noise posting and it's become almost unreadable. Are there other places where discussions like that happen?

# ? Mar 24, 2017 21:17

JawnV6: Jul 4, 2004; So hot ...

if you find one tell me

I almost had a good x86 argument on Dan kaminskys FB post

# ? Mar 24, 2017 21:38

BobHoward: Feb 13, 2012; The only thing white people deserve is a bullet to their empty skull

DuckConference posted:

the forums at realworldtech used to be real good, neckbeards having flamewars about poo poo like how wide an instruction decoder should be and poo poo like that. you could learn a lot

But recently they've been getting trolled by idiots and white noise posting and it's become almost unreadable.

it looks like maynard handley has finally figured out he should stop posting (for the wrong reasons of course) so maybe the snr will improve a bit. what a loving egotistical crank.

still too many other bad posters tho, wish david kanter would follow through on more of his warnings

tbh it's a field that tends to attract fringey thinking because every self declared genius is sure they can do a better job of designing a cpu than those dullards in industry. was going to suggest you should take a look at comp.arch on usenet but it always has its fair share of weirdo idiots too and there's no reason why that would have changed since i stopped reading it ages ago because lol usenet

# ? Mar 24, 2017 23:31

big shtick energy: May 27, 2004

Sure, there will always be the guys who decide that this one weird architecture developed at a university in the 70s is the one true way to do computing. Parts of their arguments, and refutations of their points, are often interesting.

But now there are a couple of posters that honestly may just be Markov chain bots.

# ? Mar 25, 2017 00:28

BobHoward: Feb 13, 2012; The only thing white people deserve is a bullet to their empty skull

DuckConference posted:

Sure, there will always be the guys who decide that this one weird architecture developed at a university in the 70s is the one true way to do computing. Parts of their arguments, and refutations of their points, are often interesting.

But now there are a couple of posters that honestly may just be Markov chain bots.

yeah I know what you mean. if "Ireland" isn't a bot, he's either trolling or he is the most dunning-kruger tech pundit there is. and boy does he like the sound of his own voice.

# ? Mar 25, 2017 00:53

Phobeste: Apr 9, 2006; never, like, count out Touchdown Tom, man

Jimmy Carter posted:

so is there a decent central resource for finding out about ICs are available on the market for a given task or is it just punching stuff into DigiKey and seeing what comes up? Someplace I can ask if I want, like, a non-poo poo boost converter that can handle like a third of an amp at 3.3v

https://octopart.com

unless you mean googling "stepper motor driver" and trying to find stepper motor drivers, in which case i recommend googling "stepper motor driver"

or, heck, post in this thread

# ? Mar 25, 2017 04:59

spankmeister: Jun 15, 2008

DuckConference posted:

But recently they've been getting trolled by idiots and white noise posting and it's become almost unreadable.

welcome to yospos

# ? Mar 25, 2017 07:45

big shtick energy: May 27, 2004

the dev kit for the chip didn't terminate their sdram bus so we figured it would be fine to not to terminate ours

the ringing is past the absolute max limits for the pin voltages

# ? Mar 27, 2017 18:47

hobbesmaster: Jan 28, 2008

DuckConference posted:

the dev kit for the chip didn't terminate their sdram bus so we figured it would be fine to not to terminate ours

the ringing is past the absolute max limits for the pin voltages

sounds like someone was too clever with their dev board layout

# ? Mar 28, 2017 05:18

JawnV6: Jul 4, 2004; So hot ...

hobbesmaster posted:

sounds like someone was too clever with their dev board layout

yeah that sounds like an analog wizard playing tricks on you

# ? Mar 28, 2017 05:22

longview: Dec 25, 2006; heh.

DuckConference posted:

the dev kit for the chip didn't terminate their sdram bus so we figured it would be fine to not to terminate ours

the ringing is past the absolute max limits for the pin voltages

that must have been a weird rear end dev board.
but you could get away with it for the CA bus if you had series terminations for a single chip design (at least for DDR3).

-

i got my stm32 boards that i asked about here a while back. took a couple of weeks before i could use them since i foolishly ordered an st-link to use with my st micros, but then i'd have had to use openocd for debugging which is apparently crap.

so i ordered a knockoff j-link v8, then had to do the old erase flash-reset-SAM bootloader-update dance to get a newer firmware into it, then update with the j-link software.

but then the most amazing thing happened:
i had installed the arm-eclipse project setup, made a new example project, selected my chip, changed the led blinky pin to the correct one for my board, made a debug profile with all defaults, and hit the debug button

and it worked!

gotta say the arm-eclipse folks have done a very good job with their documentation, i followed their instructions to the letter and everything works including debugging, named register views (something i never got to work in AVRs), breakpoints etc etc. worked on the first attempt.

haven't figured out how to get the trace output over the SWDIO pin yet (might have to just hook up the SWO pin?)

# ? Mar 28, 2017 18:11

hobbesmaster: Jan 28, 2008

openocd is fine though

# ? Mar 28, 2017 21:16

big shtick energy: May 27, 2004

yeah openocd has been surprisingly stable for me

the only annoying bit is that every so often the breakpoints will seem randomly jump around by a few lines. i assume whatever adjusts their position in eclipse when you add/remove lines just fucks up sometimes maybe

anyway, came here to post this (warning: reddit but it's a good post) : https://www.reddit.com/r/hardware/comments/621sd4/intels_10_nm_process_during_todays_intels/

intel is still real good at fabbing chips, at least their CPUs anyway

# ? Mar 29, 2017 02:05

atomicthumbs: Dec 26, 2010; We're in the business of extending man's senses.

someone dropped off a neat-looking but unidentifiable dev board at work for recycling and I snagged it. don't have it here but it's got an altera max fpga, some TI DSP, and an STM uc to go with em

# ? Mar 29, 2017 08:00

gonadic io: Feb 16, 2011; >>=

Got my Arduino's ADC working in Rust! i'm p happy not going to lie

code:

const VAR_RES: Port = A2;
const ADC: *mut ADC = 0x42004000 as *mut ADC;

pub fn main() {
    unsafe {
        VAR_RES.set_pin_dir(false);
        let mut digit = create_eight_segment!(D7, D6, D3, D4, D5, D8, D9, D2);

        loop {
            let s = (*ADC).analog_read(VAR_RES) as u32;        
            digit.display( s >> 7 );
            delay(100);
        }
    }
}

https://www.youtube.com/watch?v=rE2VUSBSRB8

gonadic io fucked around with this message at 14:06 on Apr 4, 2017

# ? Apr 4, 2017 13:54

big shtick energy: May 27, 2004

do people in this thread know about hotchipsvideos? you should know about hotchipsvideos

https://www.youtube.com/user/hotchipsvideos

also I enjoyed a couple of articles from everyone's favourite occasionally insightful raving lunatic:

http://semiaccurate.com/2017/03/19/intel-officially-introduces-xpoint-dc-p4800x-ssd/

http://semiaccurate.com/2017/04/04/intels-hyperscaling-is/

# ? Apr 6, 2017 07:02

big shtick energy: May 27, 2004

I may have found the problem causing the power supplies we're using to occasionally explode, so now I'm running a test to try and replicate it and knowing that it may explode at any time and I'll have to run over and kill the test is making me nervous.

# ? Apr 13, 2017 20:55

hobbesmaster: Jan 28, 2008

presenter: "our platform will run on any embedded Linux"
*mentions docker, node*
me: when you said "any" does that include armv5?
presenter: "a what?"

raspberry pis becoming the "lower end" standard for developers is driving me nuts. you can run the kernel on cortex m series devices ffs!

# ? Apr 21, 2017 21:24

Adbot: ADBOT LOVES YOU

# ? May 16, 2024 06:46

Sagebrush: Feb 26, 2012

yeah but why would you

hurr yes let's take a power-constrained device that can run a single raw executable and put a whole fuckin operating system on it and then write apps in java

# ? Apr 21, 2017 21:34

The Something Awful Forums > Discussion > Serious Hardware/Software Crap > YOSPOS > a thread about janitoring individual bits and using oscilloscopes and stuff

«‹›65 »