help-with-arduino

[pine bramble]

No with like a piece of plastic.

Mechanical pressure.

Enough to move something 1/64" of an inch. ;)

[somber burrow]

o u mean like literally just touching it to see if anything is lose

[pine bramble]

'flexing' the circuit.

[somber burrow]

ic ic

[pine bramble]

Yeah you're trying to find invisible cracks and like that. Stuff that isn't consistently conducting electricity.

I'm just guessing.

Some people used to use Freon to make the joints super cold!

[somber burrow]

why am i testing the circuit when the problem is with the mcu and maybe the power, the board works when its hooked up through usb

[pine bramble]

You can also try moving everything to the other side of the house, using a different electrical outlet, or just turn the symmetrical plugs upside down.

[somber burrow]

tried all 3 of those

nothngs working

[pine bramble]

I have to take a break from this conversation now. ;)

[somber burrow]

no worries, ping me if u ever come back

[reef ravine]

@somber burrow I'm very late to the convo so apologies if this has been covered, do you have a multimeter?

[somber burrow]

yes

[reef ravine]

does the 5v output change as you connect / disconnect the USB?

you are using something like

[somber burrow]

yes, when the usb is connected i get around 5v (4.9ish) when it is disconnected i get 2.5 votls

yes

but where it says 9v, i have 12v

[reef ravine]

maybe a bad solder joint on the barrel jack?

[somber burrow]

cause Vin is just the supply that is connected to it

[reef ravine]

2,5 is why the arduino goes weird

[somber burrow]

i cant see anything weird about the solder joint

Lol is this a problem?

The hole in it

[reef ravine]

maybe the regulator is bad, the USB is supplying 5v ok, but the 12v into the jack (which should regulate to +5) isn't

ohh just saw that - there's yer problem

[somber burrow]

i only have 1 arduino 😭

[reef ravine]

the reg isn't used with USB

[somber burrow]

so it blew up and i never noticed

[reef ravine]

likely drew too much current at one point

[somber burrow]

so when i get a new arduino how should i avoid this from happning again

[reef ravine]

never draw more than 200 or 300 mA from the board itself

[somber burrow]

oh, my supply was 300 ma

o no

it was 3 A

so 3000 mA

[reef ravine]

it's ok to draw <whatever> from your supply, just not through the onboard regulator

[somber burrow]

but for that schematic u sent earlier i have to go through the board dont i

[reef ravine]

that is what the transistors (or MOSFETs) do, use a low current control signal from the arduino to control a much larger current in the load

[somber burrow]

would this work?

[somber burrow]

but the arduino doesnt control the 12v power

would that work, sorry for t he crappy drawing

[reef ravine]

the concept is correct

the 12v supply into the barrel jack provides 5v after the reg to run the arduino (basically) only

the arduino provides a low current 5v signal to the base or gate to control a much larger current (your LEDs) via the collector / emitter or drain / source

if the LEDs are connected directly to arduino pins there is far too much current flowing in the arduino

[somber burrow]

they arnt tho, are they?

https://cdn.discordapp.com/attachments/537365760008257569/792543312958717982/ledstripbjt.png

[reef ravine]

on the arduino in that drawing "9v" is actually Vin, it's OK to connect 12v there

then the 12v goes to the LEDs

the ground connection is made through the transistors, not through the arduino itself

when the transistors are on it's like connecting R G or B to ground with a nice thick wire

at some point if you experimented without the transistors the large current was too much for the onboard regulator

[somber burrow]

i never experimented with r, g, and b without the transistors, but i never connected +12v to a transistor

[reef ravine]

in the circuit above you indirectly are (which is fine), +12v through the LED, through the transistor to ground

at some point something drew too much current through the onboard reg

[somber burrow]

so now instead of indirectly powering it i have to directly power it, because the problem was that it went through the arduino and broke the regulator, so if i cut out the arduino part then there is no regulator to break

o wait thats dumb, cause then i cant power the arduino lol

[reef ravine]

your circuit may be just fine, even a momentary accidental short can cause an issue

[somber burrow]

oh

ok

so do i just have to get a new arduino?

[reef ravine]

i assume your circuit works fine if both supplies are connected?

[somber burrow]

yes

[reef ravine]

because the USB is providing the 5v (the reg isn't used on USB)

[somber burrow]

right

[reef ravine]

if there was a more permanent wiring issue i don't think it would work at all

[somber burrow]

so what do i do about this broken regulator

[reef ravine]

well it could be replaced but honestly it's probably easier (and considering shipping) cheaper to get a new UNO

[somber burrow]

yeah im not good at smd soldering yet

so that really isnt an option

so if i get a new uno, will it break again?

with this circuit

or do i just have to be careful about shorting it, and touching it

[reef ravine]

so long as there is no short on +5v no

[somber burrow]

oooohhhhh

i remember

[reef ravine]

for it to bubble like tht it must have gotten hot

[somber burrow]

i broke it when i accidently had a connection between ground and 5v

that was a while ago

[reef ravine]

that'll do it 🙂

[somber burrow]

thanks man, this took so long to debug

[reef ravine]

no problem, mystery solved

[somber burrow]

would ya look at that

its the same problem that happened to my old board too

this is a different one that i broke a while ago, but on the same project

[reef ravine]

that one is cracked? (hard to see)

[somber burrow]

yup

[reef ravine]

to be on the safe side you can always measure +5v to ground with the ohms setting before applying power

should be hundreds or thousands of ohms

10 ohms - the magic blue smoke escapes and it stops working 🙂

[somber burrow]

23k ohms on the new one

600 ohms on the old one

[reef ravine]

600 ohms on a 5v bus is only 8 mA, perfectly acceptable

[somber burrow]

yeah but also....

that was the 5v connection on the old arduino

[reef ravine]

never measure ohms on a powered circuit

[somber burrow]

that was volts 😳

lol

[reef ravine]

man my eyes are old 😦

if 5v was reading that the regulator shorted and likely blew up everything on the board

[somber burrow]

aight well i gotta go steal an arduino from my neighbor, hopefully i dont break his or else he'll notice

[reef ravine]

lol

nice neighbors, mine would have no idea what an "arduino" was...

[somber burrow]

lol, thnx for your help

[reef ravine]

very welcome, good luck !

[torn trench]

New to electronics, trying to understand how to connect a pull-up resistor to the SQW/INT breakout here. Do I connect a resistor just between the GPIO pin I want to use and the SQW/INT breakout? Or do I need to connect a resistor between +3v, SQW/INT and the GPIO pin? https://learn.adafruit.com/ds3231-precision-rtc-featherwing/pinouts

[pulsar junco]

should be just from SQW to 3v3

[torn trench]

How do I know what GPIO pin the interrupt will trigger?

[pulsar junco]

I'm not sure I'm understanding your question, the way I read it, it will trigger on the INT/SQW pin, which you can connect to a GPIO pin

[torn trench]

I'm probably confused. 🙂 Ah, so I connect a resistor between SQW/INT and 3v3 and then a jumper/wire between SQW/INT and whatever GPIO breakout I want to trigger, if I understand correctly?

[pulsar junco]

that's how I read it 🙂 I've used this chip but not the interrupt function

[torn trench]

Thanks, I'll give it a try. Hoping to use the interrupt function + an alarm every second to update a clock/display via a MAX7219. 🤞

[pulsar junco]

nice!

[karmic cape]

So I tried to make an IR sensor thing

I'm not sure what I did wrong but I have a wire running from the "Digital 2" bit on the arduino to the right pin on the IR sensor, the middle pin on the sensor goes to "Power GND", and the left pin on the sensor goes to "Power 3V"

#define RECEIVER_PIN 2 // define the IR receiver pin
IRrecv receiver(RECEIVER_PIN); // create a receiver object of the IRrecv class
decode_results results; // create a results object of the decode_results class

void setup() {
  Serial.begin(9600); // begin serial communication with a baud rate of 9600
  receiver.enableIRIn(); // enable the receiver
  receiver.blink13(true); // enable blinking of the built-in LED when an IR signal is received
  Serial.println("Startup done!");
}

void loop() {
  if (receiver.decode(&results)) { // decode the received signal and store it in results
    Serial.println(results.value, HEX); // print the values in the Serial Monitor
    receiver.resume(); // reset the receiver for the next code
  }
}```

And then I just have this code

I tried this code from another site and still nothing

#include <IRremote.h>

int IRPIN = 2;

IRrecv irrecv(IRPIN);

decode_results result;

void setup()
{
  Serial.begin(9600);
  Serial.println("Enabling IRin");
  irrecv.enableIRIn();
  Serial.println("Enabled IRin");
}

void loop() 
{
  if (irrecv.decode(&result)) 
  {
    Serial.println(result.value, HEX);
    irrecv.resume(); 
  }
  delay(500);
}

[reef ravine]

@karmic cape you need to know the pinout of your sensor, it might be +, sense, -. or -, sense, + or...

[karmic cape]

How can I find that?

[reef ravine]

part number?

[karmic cape]

Darn it isn't listed on this box

one second

AX-1838HS

Is this strapped to a small circuit board

https://www.banggood.com/KY-022-Infrared-IR-Sensor-Receiver-Module-p-916178.html?cur_warehouse=CN&rmmds=buy

Looks like this

[reef ravine]

ok, ground, Vcc, sense

[karmic cape]

So swap the left and middle wires that connect to the pins?

[reef ravine]

• to ground, middle to 3.3v, "S" to pin 2

[karmic cape]

• is the left one, right?

Sorry, I'm really new to this

[reef ravine]

it looks to be marked "-"

[karmic cape]

Hmmm, I don't see that on mine

There's a - on the top right

And an s on the bottom left

Oh, I did get it working though

Cool

[reef ravine]

very good!

[karmic cape]

Yay

Now I need to make something that turns the HEX into a string

Thank you :D

[reef ravine]

you could use Serial.println(result.value); instead of Serial.println(result.value, HEX);, the ,HEX just shows you the hexadecimal value of the result

[karmic cape]

Oh, ok

That prints random numbers 🤔

But some buttons in the remote give the same output

If I put it back into HEX mode, holding a button gives FFFFFFFF for all buttons

Weird

Apparently it's a "repeat" code

[gray holly]

Does pulling the power on an arduino cause corruption?

[cedar mountain]

You mean like shutting off a hard drive while it's writing something? No, not unless you're doing similar stuff like saving data to an SD Card at the time.

[pine bramble]

Using a Pro Mini clone powered by a 3S LiPo for a peak voltage of 12.4V.
Will I murder the board if I supply that to the Raw-pin?

[cedar mountain]

Depends a little on the details of the clone. The voltage regulator used in the real Pro Mini can accept up to 16V, but clones might use a different part.

[pine bramble]

I'll give it a try. If it dies, I'll just get the real deal.

[full condor]

Hello! I have the Feather M0 Bluefruit LE device and trying to get a COM serial connection working between it and Windows. I have it paired but it doesn't seem to create a COM port. I tried adding one but it doesn't let me select the device. I saw back in 2015 articles that windows didn't have great BTLE support, is this still the case?

[pine bramble]

It's very likely identical to the Arduino M0 Pro, and all SAMD21 boards, in terms of recognition.

The BLE stuff is not visible to the OS, unless I have something wrong.

ATSAMD21G18A is the usual MCU for a SAMD21 board; it supports USB directly.

@full condor So the problem you have devolves to getting any SAMD21 board recognized in Windows.

[full condor]

Dang ok, any idea if the Arduino Nano 33 will have the same issue?

also ordered an HC-05 bluetooth module that supposedly works better as well

Thanks @pine bramble

[pine bramble]

@full condor All I can tell you is that variants of your question are asked here every single day of the year, no exceptions.

If you scroll back through the message channels here, eventually you'll find a conversation particular to your needs.

As I only run Linux I have no knowledge of troubleshooting com port assignments in Windows.

Certainly in #help-with-arduino and very likely in #help-with-circuitpython

Also in #general-tech and #help-with-projects

[stuck coral]

What are you using to go from a BLE connection to a serial device

Do you have a guide you are following?

[full condor]

I used the example projects which allow me to create an HID keyboard over bluetooth which works fine, and I was able to communicate over serial (UART) between my phone.

Theres just no COM port showing up on windows which seems to be an inherent issue with this chip from everything i can find, was hoping that may have changed since 2015

[stuck coral]

I feel like there is a bit of a misunderstanding, because with the phone, you used an app to open the serial connection over ble correct?

[full condor]

Correct

So some software has to initiate the serial connection on the PC side?

or can arduino initiate it?

[stuck coral]

The OS handles the actual BT connection, and there are OS services for a few of the popular use cases such as a HID device or headphones, but from my limited knowledge of windows, I do not know of a built in bluetooth serial service, I have only seen serial over bluetooth on windows using a commend line utility

Of which I do not know the name of, but you would think there is a windows service that uses a BT serial connection and just maps it to a COM device

[full condor]

Hm ok, i had assumed that a successful connection over bluetooth inherently created a serial/COM connection

[stuck coral]

Ah, no, the serial connection is on top of bluetooth, not a part of it

[full condor]

interesting, so like what is beneath that? I thought serial was basically the lowest form of communication

how do the 2 communicate otherwise? is it just like straight up bluetooth stack?

[stuck coral]

Im not sure what you mean by "otherwise". And doing a little googling, turns out there is supposed to be a OS service for serial over BT to expose a COM port, but every post Im finding explains how it just doesnt work, lol. Thats windows for ya

[full condor]

yea i think i was going through that thread earlier haha

I mean like 1 of them has to send commands to the other to initiate this serial connection i guess, but how do they do that, to me thats what serial is haha

[pine bramble]

Silabs employee says can't do it - BLE unsupported.
https://www.silabs.com/community/wireless/bluetooth/forum.topic.html/bgm13p22_windowsbluetoothvirtualcomport-tz1W

[stuck coral]

Its a L2CAP application, so a socket is opened

[stuck coral]

There you go, multi million dollar software folks

[full condor]

heh would love to use linux but its for input for windows only vidya games

[woven mica]

They could run under wine

[full condor]

a separate solution would be emulating an HID joystick over bluetooth

[pine bramble]

Forget everything else I said; I was referring to the USB physical connector on any SAMD21 board, and was entirely ignoring Bluetooth in any form. ;)

[full condor]

well, yea but the idea is that you get it and it works out of the box 🙂

[stuck coral]

Proton has made huge strides thanks to wine as well, quite impressive some games run better

[stuck coral]

In 2012 this was true

[pine bramble]

The Silabs person said 'Bluetooth Classic' is what you want for serial port stuff in Windows.

[full condor]

that should be fine im not stuck on BLE, just what i happened to have on me

HC-05 should work I believe

https://components101.com/wireless/hc-05-bluetooth-module

[pine bramble]

I think you can get a common Adafruit bluetooth module with TX and RX pins.

So, connect those to any Arduino MCU board, and write bridge software.

[stuck coral]

If you are controlling video games, why serial?

[pine bramble]

Let the Arduino present as a COM port to Windows.

[full condor]

Another software im interfacing with reads from the COM port

[stuck coral]

So why not write that software to use GATT?

[full condor]

Its not my software

[stuck coral]

Ah

[full condor]

i mean its open source but the idea was to do this in a weekend 🙂

with 20 lines of arduino code and call it a day haha

[stuck coral]

with 20 lines of arduino code and call it a day haha
Famous last words

[full condor]

Im not even going to attempt to learn how this software works i just know how to spit out the proper serial packet to make it work haha

https://github.com/opentrack/opentrack

Generating random values working yesterday over USB serial

[stuck coral]

Cool

[full condor]

Seems like it should work fine with the proper hardware coming tomorrow, the all in 1 device would have been nice but rip BLE on windows

[stuck coral]

To RIP, it needs to exist in the first place, lol

[stuck coral]

The hardware you have can technically do the job, but the software support you need isnt there

That chipset on your feather should be able to make a classical BT connection

[full condor]

Oh? Thats what i was thinking

Something else that might help anyway, I haven't been able to run the update for it

Its on version 0.6.7 from 2015, in the mobile app for controlling it theres an update to 0.8.1

[stuck coral]

In fact, I am amazed Adafruit has been lazy with it, that feather should be able to support BT 4, BLE, Gazell, thread, xbee, zigbee, etc

[full condor]

I just get a GATT ERROR after it switches to DFU mode, seems like it doesn't turn the bt back on or something

[stuck coral]

That is common, usually you need a reset, my pinetime has the same issue

[full condor]

like just turn it off and back on?

I flashed a blank sketch so nothing was happening while it did it if tha thelps

[stuck coral]

Not really if the ble chipset is what is performing the actual DFU process, I would have thought the main processor would do that but for makers maybe that makes sense

[full condor]

Yea there may or maynot be a way to jump a pin to manually cause DFU but the BLE chip is the one initiating it in this case

I think this firmware is for the BLE specifically, it doesn't effect the sketch as far as im aware

[stuck coral]

Now we're kinda in the board specific knowledge and I am not familiar

[full condor]

yea who knows, il just get a new board lol this ones been in my drawer for years, maybe something newer will just work out of the box

We're doing a custom board at the end of the day im just trying to get a proof of concept working with what i have on hand

this one would be dope cause it has everything i need built in https://store.arduino.cc/usa/nano-33-ble

if that BLE chip can also work in classic mode

oh perfect

The communications chipset on the Nano 33 BLE can be both a BLE and Bluetooth® client and host device.

[stuck coral]

Well so can your hardware

I think what you want is a nRF528xx or similar board

[full condor]

cool yea thats what that has

How do i know if this is an Arduino AVR board or SAMD?

Assuming it'd just be the Arduino Nano board profile

[stuck coral]

Well, a SAMD is labelled as such, the comparison you want to make is AVR vs ARM

SAMD is a series of micros from Microchip that use a ARM processor, vs older atmel parts commonly use a AVR processor

[full condor]

ah so its ARM Cortex M4

[stuck coral]

Correct

ATMEGA and ATTINY series micros for instance, use AVR CPUs

[full condor]

cool found it in IDE

What exactly is the difference between arduino and normal embedded hardware dev? Does the IDE just handle some of the heavy lifting behind the scenes? Can I install an Arduino HEX file into any other device running a Cortex M4 chip?

[stuck coral]

Arduino is just a framework, so it provides a HAL, some basic interfaces, and gives you a minimal environment. And sort of, the code will execute but since the chip has different peripherals and clock systems it will run into an issue but the byte code will technically be executed by the CPU

Also assuming the memory map is similar

Arduino compiles firmware for a specific target, it is not an OS

So not in the way you imagine

[full condor]

so as long as you had a device that matched a board profile it'd work

[stuck coral]

So for instance, if I load a SAMD51 compiled application and tried to run it on a nrf52, the target will hang because the clock peripheral access is totally different. So no not really, technically the CPU will run, but the system as a whole is unusable

So arduino has base package for the CPU for low level processor related stuff like delay(), which is then used in a board variant package which will map out peripherals and set up clocks

Which then you can write a complete application on

[pine bramble]

Usually, embedded development takes place via JTAG, C2, SWD or the like .. or someone writes a serial bootloader for it.

[full condor]

Interesting, I had another device at one point that just used visual studio and I could get it flashing an LED but thats about it haha

[stuck coral]

Bit out of scope though, with certain arduino targets you can use any of those methods

[pine bramble]

Either way, program execution starts at a specific base address.

[full condor]

downside of having basically no documentation, or well documentation meant for industry consumption not consumers

[stuck coral]

Yes, I mentioned entry address, but luckily ARM CPUs share a vector table at 0x0 which makes that nice and easy

Then you just need to worry about memory mapping

[pine bramble]

What Arduino adds, primarily, is an audience: a captive market of people who will buy and use add-on gadgets to the Arduino.

[stuck coral]

It depends on what you mean by Arduino, do you mean wiring, the ide, governance, company, or hardware

Arduino is a weird term

[pine bramble]

So hundreds of libraries for fun add-on gadgets exist, and many of the libraries (and gadgets) work on many different Arduino mainboards.

This provides an entry point for someone with basically no commitment to 'embedded development' to do simple experiments that succeed, without a lot of help or reading or study.

[full condor]

Yea thats been really nice, ive been meaning to get more into it but just haven't had a project til now, i have all the stuff sitting around to do it haha

[pine bramble]

I've forgotten what got Hansel and Gretel into the gingerbread house, but eventually they wind up in a more complicated situation than they bargained for.

[full condor]

Luckily I got a friend in defense/aerospace electronics that can make a custom board (though maybe not necessary except for a charging circuit with the Nano 33)

[pine bramble]

String message = "please enter your number: ";
String message2 = "your number is:";

void setup() {
  // put your setup code here, to run once:
  Serial.begin(9600);
}

void loop() {
  // put your main code here, to run repeatedly:
  Serial.println(message);
  while (Serial.available()== 0){
    
  }

  myNumber = Serial.parseInt();
  Serial.print(message2);
  Serial.println(myNumber);
}```
Anyone know why it's printing message twice, when I haven't inputed anything?

[full condor]

Before or after it outputs your number?

[pine bramble]

before, like right as the program runs

[full condor]

Cause it will output once, wait, and then come back around on the next loop and say it again

[pine bramble]

https://gyazo.com/eb4d31e155be5759d56c9e9ff8ec8a0d

[full condor]

try this in your setup after Serial.begin:

while (!Serial)
delay(10);

only guess is that its bypassing your wait for entry because serial is not intialized yet

[stuck coral]

If on a micro with integrated USB, that will not work on boards with external USB-serial

[pine bramble]

Still

[full condor]

works for me

[stuck coral]

Which board are you using?

[pine bramble]

Mega 2560

[full condor]

i changed the baud to 115200, dont think thats why tho

[pine bramble]

Lemme try that

[stuck coral]

Ah, I dont have a mega to test, do you have a genuine one or a clone?

[pine bramble]

Oh there we go, now it works

[stuck coral]

Lmao, was the baud the issue?

[pine bramble]

I guess so

[full condor]

oh weird haha

[stuck coral]

Thats hilarious, so the USB-serial reset probably caused that

[pine bramble]

lol

[full condor]

i just changed it so i didn't have to change my viewer lol

[pine bramble]

awh crap, now we got another issue

https://gyazo.com/8065fd751b80d03093945b146e7c421e

I didn't input that 0

Also @stuck coral I'm not using an offical board

[stuck coral]

What did you input?

[pine bramble]

Only thing I inputed was 5

[full condor]

Did you enter a space at the end?

5space does that for me

[pine bramble]

Nope

[stuck coral]

What is your line ending?

No line ending should be used with that sketch

[pine bramble]

Wdym?

    char t;
    t = Serial.peek();

[stuck coral]

^ That will tell you, but the serial console will append a newline by default

[full condor]

Newline and No line ending give the same thing

[stuck coral]

There is a drop down to change it

[stuck coral]

What about newline and CR?

[pine bramble]

🤔

[full condor]

Bottom right dropdown in Console viewer

should say Newline for you

change to none

[stuck coral]

Or new line and carriage return, which is it on now?

[full condor]

Thats what happens with a trailing space

[stuck coral]

Run nis's code, it will tell you what is being sent over serial so we dont need to guess

[pine bramble]

Where should I be putting it?

[stuck coral]

This should work

int myNumber;
String message = "please enter your number: ";
String message2 = "your number is:";

void setup() {
  // put your setup code here, to run once:
  Serial.begin(9600);
}

void loop() {
  // put your main code here, to run repeatedly:
  Serial.println(message);
  while (Serial.available()== 0){
    
  }
  char t = Serial.peek();
  Serial.println(t, HEX);
  myNumber = Serial.parseInt();
  Serial.print(message2);
  Serial.println(myNumber);
}

Now every character sent will have its hex value printed to the console

[pine bramble]

https://gyazo.com/df27a3631d00f0dd08b928476a0ff2b6?token=1db0620bd33e2bb9c8bf4b8d236b7fb3

[stuck coral]

You sure you have the right baud selected?

9600 is what the sketch wants

[pine bramble]

oh ye oops, changed it again

[stuck coral]

Oh, to clear that IDE error, close the console and retry uploading

[pine bramble]

Now I get the same result

Im so confused

[stuck coral]

"now" being after what action?

[pine bramble]

After using the code you sent

[stuck coral]

From the screenshot you sent, it wasnt uploaded

Unless you are having mad serial issues

[pine bramble]

https://gyazo.com/a47cee9880e9cefea423f8b379cac42d

[stuck coral]

The code was not uploaded, you jsut compiled it

[pine bramble]

Ctrl + U

Lol, I did upload tho

[stuck coral]

Not according to your screenshot

[pine bramble]

Resize the window to show more of the black box.

There's usually an orange message there. You can make things more verbose in the File menu, I think it's called.

https://gyazo.com/801f4a67872195f9548191166eb494dc

I have definitely (and recently!) assumed code was uploaded when it was not. Occupational hazard. ;)

[stuck coral]

Looks like you only compiled it to me

[pine bramble]

I clicked upload lol

[stuck coral]

Well the computer does what you tell it to

Maybe there is a verbose error?

[pine bramble]

Ill even send you a gif

Resize the window, please.

[stuck coral]

Nis wants to see the entire IDE window including the bottom edge

[pine bramble]

The black sliver at the bottom is insufficient and we already know what's in the top (white, editor) half.

I also want peace on Earth but right now I'll settle for an enlarged window there. /bargain

https://gyazo.com/db35a6f2b13fcff1b56d11000893564f

This is all I got

[stuck coral]

Hm, I am actually starting to wonder if we are actually having a serial issue, because even though your IDE says done uploading, the process is not starting

File>Preferences please click show verbose logging when uploading

[pine bramble]

Multiple client windows for the Arduino IDE should be closed.

[stuck coral]

There would be a error about not opening the port

[pine bramble]

In Linux I keep a dozen of them open, and drag one over to a new desktop, so that when it compiles and uploads, it knows which session is which.

[pine bramble]

Now I get orange text

Let's see it. ;)

https://gyazo.com/39806edbe362ca493338684c175c1ea6

I'll be back in 10 min or so, I think. Got something to attend to.

Yeah the flash was overwritten with a firmware image (in .hex). afk

pro tip: change your first print message every 5 edits, so you KNOW it's changed on the target's flashrom.

How exactly would I fix it then?

[pine bramble]

https://termbin.com/joih

@pine bramble That's fairly verbose but it does illustrate what happening.

[elder hare]

@obtuse spruce can you take a look at this :S i've never encountered this befor

Qt c++ end

void MainWindow::on_S_Brightness_sliderMoved(int position)
{
    // Debugging
    qDebug() << "[Moved][Setting][Brightness] : " << position;

    socket.sendCommandStrip(QString("brightness"), position);
}

void Socket::sendCommandStrip(const QString &bName, const QJsonValue bValue)
{
    qDebug() << "[SOCKET][INFO][VALUE] : " << bValue;

    m_webSocket.sendTextMessage(QJsonDocument(QJsonObject{{bName, bValue}}).toJson(QJsonDocument::Compact));

    qDebug() << QJsonDocument(QJsonObject{{bName, bValue}}).toJson(QJsonDocument::Compact);
}

when i move the slider all the way down i get this

[Moved][Setting][Brightness] :  0
[SOCKET][INFO][VALUE] :  QJsonValue(double, 0)
"{\"brightness\":0}"

that is correct!

now on the receiving

    if (doc["brightness"])
    {
        Serial.printf("Socket Brightness : %d \n", doc["brightness"].as<int>());

        Strips[m_selectedStrip]->setBrightness(doc["brightness"]);
        
        Serial.printf("Get Brightness  : %d \n", Strips[m_selectedStrip]->getBrightness());
    }

this gets me

Socket Brightness : 1
Get Brightness  : 1 

when on the other end i send 0 :S i dont understand why! i've been at this for almost 2 hours now!!

the slider has the value range from 0 to 255 and i can send 1 to 255 without problem but 0 gets sendt over socket but at the ESP32 end it always fetches/stops at 1 :S

[cedar mountain]

My eyes are drawn to if (doc["brightness"]). Possibly you need a more specific condition, since a value of 0 might fail there.

[errant hound]

@elder hare at a guess... are you sure you actually want doc["brightness"] i.e. have you double checked the bName value being sent? and also I assume you're using some json library for arduino? and that it is indeed parsing correctly? I would dump the whole doc as a string and make sure its structure is as expected.

looks to me that is might be that your value sits under a brightness prop.. so when doing the as<Int> it might simply be doing a logical check for the property (which exists => true) and converting 'true' to int - which might be '1'? 🤷‍♂️ just a random guess heh

actually good spot @cedar mountain you're probs spot on there... should be if (doc["brightness"] >= 0) or something 🤔

i clearly need ☕ lol

[vivid rock]

arrrgh... got this when testing the new board design

Guru Meditation Error: Core  1 panic'ed (IllegalInstruction). Exception was unhandled.

back to debugging
Who wrote these error messages? guru meditation error

[gilded swift]

Lmao

[safe shell]

I wondered the same thing a while back, and looked it up. There's a history to that error message.

[obtuse spruce]

@elder hare - no idea ... but... I'd suggest printing out the string received at the ESP32 before interpreting it as JSON - to be certain about what it is receiving -- and to correlate each message with other debugging output.

Now - why does setBrightness() take a JSON value? That seems odd - shouldn't it take an integer?

I think that @cedar mountain is onto something... I don't know what that library does for JSON values converted to bool - as you've used in your if statement.

[vivid rock]

btw, i figured what was causing the kernel panic: I switched feathers from m4 to esp32 and forgot to change pin definitions. Turns out that declaring pin 6 as INPUT_PULLUP on esp32 leads to kernel panic...

[cedar mountain]

Ah, yeah, that'd probably do it. GPIO6 is the SPI CLK for the internal flash memory, so turning it off while you're executing code from those addresses would be... bad. 😅

[misty spruce]

Hi! I'm unable to launch Arduino IDE (1.8.13). I have also tried different varients like windows app, hourly builds, Beta Builds. None of them worked in my system. When I try to launch the application it shows its interface having "preparing boards" after that it disappears. I'm using a windows 10 system which is updated. Please help me with the issue 😀

[north stream]

Try running it from a CMD window to see if it provides any additional output.

[pine bramble]

@misty spruce Might be worth a reinstall. You may have missed a step.

[misty spruce]

I have done that several times also I tried with older versions as well none of them worked

[vivid rock]

@misty spruce can you check logs?
at C:\Users\<username>\AppData\Local\Arduino15\logs

Note that by default, AppData folder is hidden from view on Windows

you might need to enable hidden items checkbox in View tab of the Windows file explorer

[misty spruce]

Do I need to delete that file?

[vivid rock]

no, you can read it and see if it contains some error messages that might tell you what the problem is

if nothing helps, you can try to delete the whole C:\Users\<username>\AppData\Local\Arduino15 folder - it may contain some corrupted cached data

and then reinstall

[misty spruce]

Inside logs there is a text document but when I opened it it is empty 😭

[pine bramble]

In Linux, I think you need some basic java stuff present for the Arduino IDE's 'processing' sub-modules.

JRE or JDK maybe.

[vivid rock]

which version of Arduino are you currently using? is it Windows store one?

[misty spruce]

I was using the Windows app and yesterday it was not opening so I tried using the software downloaded from the site version 1.8.13 ( I tried all the builds ). Then I tried with an older version 1.8.12. none worked

[vivid rock]

Windows store app stores all logs and preferences in a different location

[pine bramble]

Maybe ask on the Arduino discord.

[vivid rock]

let me try and find where

[pine bramble]

The aduino.cc site didn't seem to suggest you need to know anything to install it.

[misty spruce]

Okay

[misty spruce]

Okay

[pine bramble]

They offer an .exe and a .zip and the .zip has caveats for a portable installation; the .exe is player-piano.

[misty spruce]

Yeah

[vivid rock]

I'd suggest:

• uninstalling all versions of arduino

[pine bramble]

dyno says no invite links so go through the web. There may be an invite link on this Adafruit server somewhere. Maybe in a pinned note, in this chan.

[vivid rock]

• removing folder C:\Users\<username>\AppData\Local\Arduino15

• reinstalling arduino (not from widnows store but downloaded from Arduino website)

[pine bramble]

Yeah I always uninstall the old version of the Arduino IDE (just by moving two directories) before installing the upgraded one.

[vivid rock]

but it is also important to remove config directory - uninstall doesn't do it, iirc

of course, manually moving would also work - I was thinking of uninstall using windows app management

see this as well (where they suggest less radical solution): https://arduino.stackexchange.com/questions/55881/arduino-ide-1-8-6-not-starting

[pine bramble]

In Linux you own everything - I don't think there's a root privilege needed to reinstall. Maybe just for udev stuff under /etc, if that.

(But I manually install so I'm not sure)

[vivid rock]

sure, but Mercury is using Windows...

[misty spruce]

Yeah I love linux but my college don't allow linux

[tawdry flint]

Hey could anyone help me with the adafruit mpl 3115A2 (its an altimeter) sensor? It is not showing up with the example code or the IC2 test. I soldered the pins today and I think I have all the wiring right. I currently have another ic2 sensor connected (the tsl2591 a light sensor) and that one is found by the ic2 test. If I try to interchange the alt sensor with the light sensor the alt sensor still does not show up 😦 I am using the esp8266. Does anyone maybe have any idea what I could do?

[woven mica]

How cant it allow linux?

[misty spruce]

Their software is built only for Windows and Mac

[vivid rock]

@misty spruce so did "remove everyithing including config directory and reinstall" work?

[misty spruce]

I'm checking on it

[tawdry flint]

It is a bit messy maybe

[pine bramble]

I usually bother to route power and ground using cut wires (wire strippers, solid hookup wire).

That removes some of the complexity, visually, as it's very clear where power and ground are.

[tawdry flint]

Yeah I only have these long wire for the time being this is my first project but I am for sure going to do that

[pine bramble]

Solderless breadboards sometimes split the two rails in half, horizontally, and those need a jumper to connect the rails together. Top and bottom rails are often jumpered together to form two rails in total for the entire breadboard.

In this case you used the top rails for SDA and SCL (interesting choice!)

[misty spruce]

Sry, that also didn't work

[tawdry flint]

I tried to move the second sensor up but it did not work sadly the ic2 scan can still only see the light sensor

[tawdry flint]

Thanks

[pine bramble]

@tawdry flint carry SDA and SCL with jumpers across the void in the middle of the board. If not needed, doesn't harm anything at all.

Those two red lines you drew in - make them real. ;)

[tawdry flint]

Oh ok

So maybe that is also something I have to do for power on the other side

[pine bramble]

Right. Depends on the breadboard.

The large one I have with the metal plate (from Adafruit) needs the jumpers I talked about.

It's handy to have them separate, at times, but when you need one long rail, it's important to bond them yourself.

[vivid rock]

@misty spruce ok... did it produce a log file?

[tawdry flint]

OMG it worked!!

Thanks

The fourth sensor now has lights! Thank you so much @pine bramble I would not have found this anytime soon!

But yeah I really need to get propper short wires to get rid of this spagetti mess 🍝

[pine bramble]

LOL lucky guess on my part. ;)

Still that's pretty cool when it works out to be something simple.

[tawdry flint]

Yeah exactly

Amazing thanks again @pine bramble

[misty spruce]

Yeah but it's empty 😭

[vivid rock]

can you try and start it from the command line?
in windows search field ("type here to search" on the panel), enter "run"

it shoudl give you command prompt

and then enter

"C:\Program Files (x86)\Arduino\arduino.exe"

(with quotes)

[misty spruce]

Okay

[misty spruce]

Still the same

[vivid rock]

but does it say anything? any error messages?

[gilded swift]

So i'm running into an issue

and here's where the issue lies:

  if (&i2c_dev) {
    delete i2c_dev;
  }
  i2c_dev = new Adafruit_I2CDevice(i2caddr);
  // Serial.println(i2c_dev->address(), HEX);
  if (!i2c_dev->begin()) {
    return false;
  }
  writeRegister8(AT42QT107_RESET,
                 (0x01)); // simply write a nonzero number to reset
  delay(1);
  writeRegister8(AT42QT107_RESET, 0x00);
  for (uint8_t i = 0; i < 0x7f; i++) {
  }

  writeRegister8(AT42QT107_CAL, (0x01)); // Calibrate the device by writing a non
                                       // zero value to calibration register
  // may not need this curr variable...
  // uint8_t curr = readRegister16(AT42QT107_KEY_0_1, AT42QT107_KEY_0_2);
  return true;
}```

In this method, the proper i2c address returns on the Serial print call

but when I called the same i2c_dev->address() in another method, the i2c address comes back 0xE0

  uint16_t read16;
  Serial.print("i2c address from i2c_dev: ");
  Serial.println(i2c_dev->address(), HEX);
  Adafruit_BusIO_Register read_reg1 =
      Adafruit_BusIO_Register(i2c_dev, regMSB, 2);
  read_reg1.read(&read16);
  //Serial.print("Value of 16 bit register being read... : ");
  //Serial.println(read16, HEX);
  return (read16); // change to value returned from register
}```

the serial print in this method returns the wrong i2c address which I've narrowed down as to why my library isn't operating as intended when I call other helper methods that use the readRegister16 and readRegister8

[pine bramble]

The empty for loop needs to be volatile.

(it's optimized out otherwise since it does nothing)

    for (volatile int i = 1776; i>0; i--) { foo(); }

[gilded swift]

that didn't change the output

somewhere along the line the value of the i2c_dev address variable is being overwritten

[elder hare]

what am i doing wrong with this pattern, the pattern is correct the only thing is that it starts from 1 end and doesn't continue it goes from 0 to 84 (strip length) and then restarts

void StripSettings::TheaterChase()
{
    fadeToBlackBy(m_Leds.data(), m_NumLeds, 255);
    for (uint16_t i = 0; i < m_NumLeds; i = i + m_Spacing)
    {
        m_Leds.data()[(i + m_Index)] = CHSV(m_Hue, m_Saturation, m_Brightness);
    }
}

void StripSettings::Increment()
{
    if (m_Direction == 1)
    {
        m_Index++;
        if (m_Index >= m_NumLeds)
        {
            m_Index = 0;
        }

    }
    else if (m_Direction == -1)
    {
        m_Index--;
        if (m_Index <= 0)
        {
            m_Index = m_NumLeds - 1;
        }
    }
}

how do i make it go continuously ?

[raven plank]

I'm working on a Bluetooth MIDI mod for the Nintendo Power Glove using an Adafruit Feather nRF52832. For the keypad, I set up a matrix and had prototype circuit boards printed. Every PWM output row is working, except for one row which is connected to GPIO Pin 5, which is also the TX pin.

I'm using BLEMidi to transmit MIDI via bluetooth if that is of relevance.

[elder hare]

anyone?

[safe shell]

@raven plank pin 6?

[raven plank]

Yes, sorry! 6.

[topaz compass]

where's the code that calls these methods?

[errant hound]

@elder hare sometimes it takes a while for people to get to your questions... best to try and be patient.

Hard to say with just that snippet... you are assuming your issue does not lie outside of that code but at first glance, how are you changing direction? perhaps something like:

if (m_Direction == 1)
{
    m_Index++;
    if (m_Index >= m_NumLeds)
        {
            m_Index = m_NumLeds; // might be m_NumLeds-1 depending on indexing of arrays etc
            m_Direction = m_Direction * -1;
        }
    
}
else if (m_Direction == -1)
{
    m_Index--;
    if (m_Index <= 0)
        {
            m_Index = 0;
            m_Direction = m_Direction * -1;
        }
}

🤷‍♂️

LED output might look something like:

1-0-0-0-0-0-0-0-0-0
0-1-0-0-0-0-0-0-0-0
0-0-1-0-0-0-0-0-0-0
0-0-0-1-0-0-0-0-0-0
0-0-0-0-1-0-0-0-0-0
0-0-0-0-0-1-0-0-0-0
0-0-0-0-0-0-1-0-0-0
0-0-0-0-0-0-0-1-0-0
0-0-0-0-0-0-0-0-1-0
0-0-0-0-0-0-0-0-0-1
0-0-0-0-0-0-0-0-0-1
0-0-0-0-0-0-0-0-1-0
0-0-0-0-0-0-0-1-0-0
0-0-0-0-0-0-1-0-0-0
0-0-0-0-0-1-0-0-0-0
0-0-0-0-1-0-0-0-0-0
0-0-0-1-0-0-0-0-0-0
0-0-1-0-0-0-0-0-0-0
0-1-0-0-0-0-0-0-0-0
1-0-0-0-0-0-0-0-0-0

[raven plank]

https://devzone.nordicsemi.com/f/nordic-q-a/29932/pwm-doesn-t-work-on-pin-4-and-pin-6 Seems like this person was having a similar issue and posted a support request to the Nordic DevZone site, except with a Nordic development kit nRF52832_DK instead of the Adafruit Feather nRF52832 which I'm using. Setting NRF_LOG_BACKEND_UART_ENABLED to 0 in the sdk_config.h seemed to work for them. I wasn't able to find a library named sdk_config.h in the files I had downloaded for the board through Arduino's board manager. Could this be the setting I'm looking to disable as well? If so, I'm confused on how to go about doing that.

[gilded swift]

I should add that not only is the i2c address not returning properly, i'm also not getting any value returned from the registers when I attempt to read from them.

If I do standalone code that doesn't utilize my driver, I can read from the registers fine.

[eager sundial]

@gilded swift so the issue is that your i2c is getting changed somehow?

[gilded swift]

Yeah, it appears it's changing sometime after exiting the ODT_AT42QT1070::begin() method

[eager sundial]

What address do you expect it to be?

It ends up being 0xE0?

[gilded swift]

the i2c address is permanently set by microchip to 0x1b

right after making the Adafruit_I2CDevice instance, the i2c_dev->address() call returns 0x1b

  if (i2c_dev) {
    delete i2c_dev;
  }
  i2c_dev = new Adafruit_I2CDevice(i2caddr);

  if (!i2c_dev->begin()) {
    return false;
  }
  writeRegister8(AT42QT107_RESET,
                 0x01); // simply write a nonzero number to reset
  delay(1);
  writeRegister8(AT42QT107_RESET, 0x00);
  for (uint8_t i = 0; i < 0x7f; i++) {
  }

  writeRegister8(AT42QT107_CAL, 0x01); // Calibrate the device by writing a non
                                       // zero value to calibration register
  // may not need this curr variable...
  // uint8_t curr = readRegister16(AT42QT107_KEY_0_1, AT42QT107_KEY_0_2);
  return true;
}```

[north stream]

@elder hare Maybe something like this? ```arduino
void StripSettings::Increment()
{
if (m_Direction) {
if (++m_Index >= (m_NumLeds - 1)) {
m_Direction = 0;
}
} else {
if (--m_Index <= 0) {
m_Direction = 1;
}
}
}

[gilded swift]

I've tried commenting out the writeRegister methods to make sure those were not the issue. I'm not sure if maybe my BusIO library is just sucky

maybe I need to redownload it

[eager sundial]

So after begin() all is good, it's only getting weird when you try to read/write?

[gilded swift]

yeah, when calling read functions the i2c address comes out 0xE0

[eager sundial]

okay, can I see you sketch's code?

[gilded swift]

#include <Adafruit_I2CDevice.h>
#include <Adafruit_I2CRegister.h>
#include <Adafruit_BusIO_Register.h>
#include <Wire.h>

#include <ODT_AT42QT1070.h>

//Adafruit_I2CDevice i2c_dev = Adafruit_I2CDevice(0x1B);

void setup() {
  // put your setup code here, to run once:
  while (!Serial) { delay(10); }
  Serial.begin(115200);
  Serial.println("Qt Py using ODT AT42QT1070 Acorn: i2c detection test...");
  ODT_AT42QT1070();
  //ODT_AT42QT1070().begin();
  //delay(1000);
  //Serial.print("printing value of ODT_AT42QT1070.begin() : ");
  //Serial.println(ODT_AT42QT1070().begin());
  Serial.print("Attempting to read register 0x02: ");
  Serial.println(ODT_AT42QT1070().readRegister8((0x02)));
  //Serial.println(ODT_AT42QT1070().touched());
  /*if (!i2c_dev.begin()) {
    Serial.print("Did not find device at 0x");
    Serial.println(i2c_dev.address(), HEX);
    while (1);
  }
  Serial.print("Device found on address 0x");
  Serial.println(i2c_dev.address(), HEX);*/

  /*Adafruit_BusIO_Register id_reg = Adafruit_BusIO_Register(&i2c_dev, 0x00, 1);
  uint8_t id;
  id_reg.read(&id);
  Serial.print("ID register = 0x"); Serial.println(id, HEX);
  
  Adafruit_BusIO_Register fw_reg = Adafruit_BusIO_Register(&i2c_dev, 0x01, 1);
  uint8_t fw;
  fw_reg.read(&fw);
  Serial.print("Firmware register = 0x"); Serial.println(fw, HEX);*/
  Serial.println(ODT_AT42QT1070().touched());
  delay(1000);
   
}

void loop() {
  // put your main code here, to run repeatedly:
  
  //Serial.print("value of register detect register: ");
  
  //Serial.println("what's going on here...");
  //Serial.println(ODT_AT42QT1070().readRegister8(0x0));
  /*Adafruit_BusIO_Register detect_reg = Adafruit_BusIO_Register(&i2c_dev, 0x02, 1);
  uint8_t dt;
  detect_reg.read(&dt);
  Serial.print("Detect register = 0x"); Serial.println(dt, HEX);*/
}```

[lost nest]

Do you have your library somewhere accessible?

[gilded swift]

https://github.com/skerr92/at42qt-acorn

[eager sundial]

My first guess was something about the Adafruit Bus Device is getting cleaned up, but you're creating it on the heap during begin()

[gilded swift]

don't mind the CI status, it's mad because i haven't updated Doxygen tags 😛

[lost nest]

So in the sketch you currently posted, it seems like you don't explicitly do the begin (it's commented out and your constructor doesn't do it by default).

[eager sundial]

yeah just caught that too

[gilded swift]

even uncommented, it doesn't. I mostly had it commented because I was checking an individual call of BusIO to read the registers

[lost nest]

But without the begin, i2c_dev is never initialized

(since again, not initialized in constructor)

[gilded swift]

This is with begin() uncommented

the "im being called" are in a print statement for writeRegister to make sure it's being called

register 0x02 is the detect_status register

[eager sundial]

also, the commented out code in your loop creates a new ODT_AT42QT1070 every time on the stack, which is mostly fine, but if you call begin with that it'll leak the Adafruit Bus Device that begin allocates on the heap unless you free it in the destructor for ODT_AT42QT1070.

[gilded swift]

should be 0x1

[eager sundial]

okay, are you sure this is succeeding? https://github.com/skerr92/at42qt-acorn/blob/28d671f45328acbaded7ec3e4096b15078cb4b9e/ODT_AT42QT1070.cpp#L55-L57

[gilded swift]

yes, begin returns 1 on exit

[eager sundial]

hrm

At this point I'm not sure. There's a lot of layers of abstraction here, but there isn't anything that super stands out as memory corruption to me.

Does this code work if you inline it in your setup() instead of having it in this class?

[lost nest]

Err it looks like you're making a new ODT_AT42QT1070() every time you call it

Wouldn't you make one global?

and then instantiate it once?

[gilded swift]

wait I did something

[lost nest]

I don't even want to think about what happens if that gets out of sync with the underlying pointers

[eager sundial]

mentioned that, unless you caught a different instance

[lost nest]

It's more so that even in begin

e.g. line 15/16

[eager sundial]

ah I see

[lost nest]

ODT_AT42QT1070();
ODT_AT42QT1070().begin();

[eager sundial]

oh gosh I totally missed that

[gilded swift]

I was doing a Java thing when I should have been doing a c++ thing

[eager sundial]

yeah you want something more like:

ODT_AT42QT1070 sensor = ODT_AT42QT1070();
sensor.begin();
...

[gilded swift]

I hate how heavily they focused on java in my undergrad. Not that it's bad, it's just not productive for writing C

[lost nest]

Or even more cleanly put the I2C address in the constructor and pass it through that way

[eager sundial]

@lost nest it's a default argument, which is fine?

[lost nest]

At the moment it's a default argument of begin()

[gilded swift]

@lost nest the i2c is defined in the header and passed explicitly in the begin header parameters

I had changed it in the cpp file just in case I was messing something up

[eager sundial]

I think that makes sense, since if a device can have multiple addresses the program may not know which address to use during initialization of global variables.

[gilded swift]

but now that I realized I am the goof (as I guess I was) I can change it back

[lost nest]

I would expect the programmer to know though, and make it explicit in that initialization. But that's just a different-codebases-different-approach nit, I think.

[eager sundial]

yeah

I was gonna look at what the adafruit drivers do

[gilded swift]

benefit to this chip is i2c address is fixed and can't be manipulated. not necessarily the case for all i2c devices though

[eager sundial]

yeah adafruit does that, @lost nest: https://github.com/adafruit/Adafruit_SCD30/blob/master/Adafruit_SCD30.cpp#L68

[lost nest]

If it's truly fixed, why have the extra complexity of the modularity?

[gilded swift]

I followed a template and I will be modifying it to reflected the fixed nature

it's my first driver code so plenty to learn 🙂

[eager sundial]

yeah

[lost nest]

Yeah, benefit of this configuration being in the constructor is that you can chain initialization to have everything defined in constructors from the start.

(IMO)

[gilded swift]

well, first standalone driver code that is

[lost nest]

But if that's not the standard of the codebase

I'm turbo-not-rocking-the-boat

[eager sundial]

@lost nest one use case I can think of for leaving the address in as a parameter is i2c multiplexers that give devices different addresses.

& yeah, also to just conform to existing style.

[gilded swift]

yeah, I believe the other variants of the AT42QT chip have different i2c addresses so it might help for modularity in the future

[eager sundial]

Did you get your stuff working, @gilded swift?

[gilded swift]

yeah 🙂

thank you ❤️

and for your help too @lost nest

[eager sundial]

Harry spotted it!

[gilded swift]

i've been pounding away at this all day and I just needed new eyes

[eager sundial]

I couldn't see the forest for the trees. I paid too much attention to the driver code and not enough to the sketch. 🙂

[gilded swift]

🙂

[eager sundial]

Cool, I'm signing off for a bit. Good luck, have fun!

[gilded swift]

thanks again!

[green thunder]

Hello! I have a couple of questions about using the DMA NeoPixel functionality, on 2 boards specifically. Firstly, for the Itsy Bitsy nRF52840 Expess, according to https://learn.adafruit.com/adafruit-itsybitsy-nrf52840-express/pinouts I should be able to use DMA on pin 5. However, the Adafruit_NeoPixel_ZeroDMA library will not compile; it complains that the nRF52840 is not a SAMD. Is there an alternate version of the library for nRF52840, or is there something else I'm missing?

Secondly, the QT PY. I know in the guide at https://learn.adafruit.com/adafruit-qt-py it doesn't specifically mention that DMA is possible (and indeed in my tests, it does not work), however it seems like it should be possible since it uses a SAMD21 M0. Is there a technical reason why this doesn't work, or has it just not been implemented on the QT PY (yet)? Any help shedding light on this would be deeply appreciated 😄

[vivid rock]

@green thunder neopixel_zeroDMA library only works on m0 and m4 based boards, so it will not work on nRF52840. The statement in learn guide is plain wrong - my guess isthat it is result of copy and paste from ItsyBitsy m4 guide; these two boards are very similar, but

[green thunder]

@vivid rock I was worried that was the case 😦 Thank you!

[vivid rock]

as for QT Py, one probably can make ZeroDMA library work with it - one needs to check carefully which samd21 pins are broken out on qt py and which peripherals (sercom, DMAC) are available on these pins. Maybe someone more knowledgeable than me knows the answer.

[green thunder]

Okay thanks! That is also what I suspected. I'd love to hear from someone with specific knowledge on this, but in the meantime I'll look further into if/how it would be possible to enable DMA support for QT PY.

[lost nest]

@green thunder It does appear that some amount of DMA/PWM functionality is supported for nrf52 series chips, at least from my reading of https://github.com/adafruit/Adafruit_NeoPixel/blob/master/Adafruit_NeoPixel.cpp#L1345

It's just in the mainline Adafruit_NeoPixel library instead of the specialized specifically ZeroDMA library

[green thunder]

@lost nest Oh that's interesting! It's unclear to me at first glance how DMA is enabled (e.g. is it just automatically enabled if we assign the neopixels to pin 5?) but I will play with this tomorrow and see what I can come up with. Thanks for the help!

[lost nest]

I'm not sure if it even needs to specifically be pin 5, but it needs to be a PWM pin

[green thunder]

hmm okay good point

[lost nest]

(for which pin 5 may just be the easy default one, or have some other mapping that's more complicated e.g. other pwm pins are used for other things). Otherwise it looooks like it should be just done by default assuming resources are available

but I also made the classic blunder of digging directly into the cpp files

[green thunder]

Ah well, pin 5 is of particular interest because for the itsy bitsy it's output only and level shifted to 5V, making it great for driving neopixels. DMA is icing on the cake lol

[lost nest]

Ah perfect.

I suspect then it will be a PWM pin

slash pin that works for this application

[green thunder]

I suspect as much too. My brain is mushy tonight but I'll play with it in the morning and confirm

[elder hare]

@north stream funny, i tried your code and it's the same the only difference is that yours when hitting the end it moves backup to start 😛

[elder hare]

this is how far i've gotten

void StripSettings::TheaterChase()
{
    fadeToBlackBy(m_Leds.data(), m_NumLeds, 255);
    for (uint16_t i = 0; i < m_NumLeds; i++)
    {
        if ((i + m_Index) % m_Spacing == 0)
        {
            m_Leds.data()[i] = CHSV(m_Hue, m_Saturation, m_Brightness);
        }
    }
}

void StripSettings::Increment()
{
    if (m_Direction == 1)
    {
        ++m_Index;
        if (m_Index >= m_NumLeds)
        {
            m_Index = 0;
        }

    }
    else if (m_Direction == -1)
    {
        --m_Index;
        if (m_Index <= 0)
        {
            m_Index = m_NumLeds - 1;
        }
    }
}

This works with the m_Spacing and m_Direction the only thing is WHEN m_Index hits 84 (total length of strip) / maybe it is when it reaches arround to 0 it gets a hickup (lag/jump) in the pattern but then continues! this goes for both direction Reverse and Forward :/

[elder hare]

sooo

void StripSettings::TheaterChase()
{
    fadeToBlackBy(m_Leds.data(), m_NumLeds, 255);
    for (uint16_t i = 0; i < m_NumLeds; i++)
    {
        if ((i + m_Index) % m_Spacing == 0)
        {
          m_Leds.data()[i] = CHSV(m_Hue, m_Saturation, m_Brightness);
        }
    }
}

void StripSettings::Increment()
{
    if (m_Direction == FORWARD)
    {
        m_Index = (m_Index + -1);
    }
    else if (m_Direction == REVERSE)
    {         
        m_Index = (m_Index + 1);
        // if totalsteps are greater than or equal to numleds reset and start from 0
        if (m_Index >= m_NumLeds)
        {
            m_Index = 0;
        }
    }
}

focusing on the Direction = Reverse while it works when it comes to the wrap arround the pattern JUMPS each time it reaches the end (strip length) and resets to 0 :S what am i doing wrong?

[errant hound]

thats because you literally tell it to go back to zero m_Index = 0;- then you add one and it continues from first pattern

the only thing you need to do on each side is check if over limit (0 or numleds) and if so, reset back to either min/max and change direction

as per my earlier code snippet which worked

also im assuming you want 'knightrider' back/forth lights... if not, please describe the pattern you are actually seeking 😬

should have asked that earlier LoL but im stuffing my face with food so kinda distracted hehe

[woven mica]

bon apetit 👌

[elder hare]

@errant hound is there anything wrong with doing this

void StripSettings::TheaterChase()
{
    fadeToBlackBy(m_Leds.data(), m_NumLeds, 255);
    for (uint16_t i = 0; i < m_NumLeds; i++)
    {
        if ((i + m_Index) % m_Spacing == 0)
        {
        ColorMode(i, m_Brightness);
        }
    }
    m_Index = (m_Index + m_Direction) % m_Spacing;
}

reason im asking is it totaly works! but on strip2 im getting green flickering LEDs :S

[errant hound]

well, if you log m_Index after the assignment you will see that depending on direction var it will either be -N to 0 (when m_Direction = -1) and 0 to +N when m_Direction = 1 🤔 are you sure thats what you want?

as I don't know what on earth TheaterChase is or fadeToBlackBy or m_Leds or what hardware you have its a little tricky to know exactly what should be happening - unless you can share all your hardware/code (in a bin/image host/repo) or articulate it very clearly 😬

[frozen idol]

is it common for AMG8833 sensors to have uneven heat sensitivity?

the pixels on the right side of the camera seem to be reading hottest even when a hot object travels across the screen

when i move my hand across the screen the hottest pixels occour at x coordinate 2 (y coordinates vary more)

or really x coordinates 0-3 but never 4-7

i mightve swapped the x and y coordinates on accident... lemme swap them and double check again

okay that was the issue the sensor reads y columns first, not x rows

[ivory fulcrum]

Hi there! I'm looking for easy to use library for a membrane switch module (keypad) which consists of "*, #, 1, 2, 3, 4, 5, 6, 7, 8, 9, A, B, C, D" Anyone got a good one?

[thick stag]

How about this? https://playground.arduino.cc/Code/Keypad/

You specify the input pins and rows, and the corresponding keys you want it to output.

[meager flax]

Hi I am using nRF52840 itsybitsy and keep getting this error when compiling in arduino.... Flow control is disabled, Single bank, Touch disabled
Timed out waiting for acknowledgement from device.

[stuck coral]

Compiling or uploading? Lol

[meager flax]

uploading

[stuck coral]

Over what?

I assume USB?

[meager flax]

Yes sorry over USB

[stuck coral]

Which nrf52 device are you flashing?

[meager flax]

Upgrading target on COM1 with DFU package C:\Users...\AppData\Local\Temp\arduino_build_806481\Combined_NRF52_Ble_Usb_Joystick.ino.zip. Flow control is disabled, Single bank, Touch disabled
Timed out waiting for acknowledgement from device.

Failed to upgrade target. Error is: No data received on serial port. Not able to proceed.
Traceback (most recent call last):
File "nordicsemi_main_.py", line 294, in serial
File "nordicsemi\dfu\dfu.py", line 235, in dfu_send_images
File "nordicsemi\dfu\dfu.py", line 200, in _dfu_send_image
File "nordicsemi\dfu\dfu_transport_serial.py", line 179, in send_start_dfu
File "nordicsemi\dfu\dfu_transport_serial.py", line 243, in send_packet
File "nordicsemi\dfu\dfu_transport_serial.py", line 282, in get_ack_nr
nordicsemi.exceptions.NordicSemiException: No data received on serial port. Not able to proceed.

Possible causes:

• Selected Bootloader version does not match the one on Bluefruit device.
  Please upgrade the Bootloader or select correct version in Tools->Bootloader.
• Baud rate must be 115200, Flow control must be off.
• Target is not in DFU mode. Ground DFU pin and RESET and release both to enter DFU mode.

nRF52840 itsybitsy by adafruit

[stuck coral]

Ah this is specific to the Adafruit board, they have their own bootloader, neat.

Have you tried following the updating the bootloader guide? https://learn.adafruit.com/bluefruit-nrf52-feather-learning-guide/updating-the-bootloader

[meager flax]

I looked at it but then the device status light wasn't turning on so I unplugged it for a few minutes and then it came on. I will have to try again.

So I wasn't sure if computer was even detecting it

[stuck coral]

Got it, Im not really sure, maybe someone more familiar with the Adafruit bootloader for the nrf52 can chime in, have you gone through all the possible causes listed?

[meager flax]

It seems like the computer is having trouble communicating with it now. I have to figure out how to reload the bootloader I think

[queen fjord]

Could anyone help me understand some things related to the PCA9685 library and it's use?

[cedar mountain]

The rule of thumb is just to ask your question rather than waiting for a volunteer to step forward. In many cases people can answer things that they haven't previously worked with just because they're good at quickly scanning datasheets, source code, or schematics.

[queen fjord]

Oh alright, thank you

1. What does this mean? The amount of servos I'll be using?

[cedar mountain]

It looks like it's intended as an index to multiple servos, which might be incremented to loop over them or something.

[queen fjord]

2. Is this really necessary? I can't afford an oscilloscope.

[cedar mountain]

I can't comment on how necessary it is (probably not very, though it'll affect the accuracy of the servo commands), but I've gotten a surprising amount of usage from one of those $20 8-channel "24 MHz" USB logic analyzers. Highly recommended for people who can't afford an oscilloscope.

[queen fjord]

Alright

One more thing, how would I come about making the pwm.setPWM(channel,on,off) function work with degrees? Like, how would I map/configure it?

Because right now I'm trying to use angles coming from a gyroscope to move servos, but I don't know how I would pull it off with the on and off parameters

[cedar mountain]

It depends on how your servo works. Typically a servo will move to its minimum angle with an on pulse of 1000 microseconds, and its maximum angle with 2000 microseconds.

[queen fjord]

it's an SG90

[cedar mountain]

The off pulse would be calculated based on the balance of the 50 Hz update period.

[queen fjord]

Using the Arduino servo library, I did tests and the min and max values are 540 and 2500 for my SG90s

540 is 0 and 2500 is 180

[cedar mountain]

Cool. So you'd calculate microseconds likemicrosec = 540 + (2500 - 540) * degrees / 180;and then the on/off periods are in terms of a 4096-unit-long 50 Hz period, so it'd be likeon_pulse = 4096 * microsec / 20000; off_pulse = 4096 - on_pulse;

(I think, anyway. Not tested.)

[queen fjord]

Thank you very much, I'll test it

[cedar mountain]

Actually, that's wrong. The function wants time values, not durations, so it'd be like setPWM(num, 0, on_pulse), where you want the pulse to start at the beginning of the period, and then stop after on_pulse units.

It's be easier to just use the writeMicroseconds() function.

[queen fjord]

Oh, alright

then I would just map degrees to a min output of 540 and a max output of 2500?

[cedar mountain]

Yup!

[queen fjord]

Perfect, thank you very much

[loud horizon]

Hello, I've connected two 10 segment leds to an arduino nano and it started blinking for some reason when I hooked it up. What code should I use to control the led's? I'm going to use this for racing using simhub.

[queen fjord]

@cedar mountain

#include <Wire.h>
#include <Adafruit_PWMServoDriver.h>
#include "I2Cdev.h"
#include "MPU6050_6Axis_MotionApps20.h"

Adafruit_PWMServoDriver pwm = Adafruit_PWMServoDriver();
MPU6050 mpu;

#define SERVO_FREQ 50 // Analog servos run at ~50 Hz updates
#define OUTPUT_READABLE_YAWPITCHROLL

void setup() {
  Serial.begin(9600);

  pwm.begin();
  pwm.setPWMFreq(SERVO_FREQ);  // Analog servos run at ~50 Hz updates

  delay(10);
}

double mapdouble(double x, double in_min, double in_max, double out_min, double out_max) //Map function
{
 return (x - in_min) * (out_max - out_min) / (in_max - in_min) + out_min;
}

void loop() {


  double testdegree;

  if(Serial.available() > 0) { //Check for serial entries
      testdegree = Serial.parseInt(); //Save entry
      }
  
  
  double degree =  mapdouble(testdegree, 0, 180, 600, 2400);

  Serial.print(degree);
  Serial.println();
  
  pwm.writeMicroseconds(6,degree);

   }

So, I was making the code and I ended up with this. Tho the results are not as intended

The servo I have on the first channel of the PCA9685 flickers when I upload the sketch or enter the serial monitor

Regardless of the writeMicroseconds being assigned to channel 6

I checked the mapping calculations through the serial monitor and they are correct

Any idea of what I could be doing wrong?

[cedar mountain]

You say you have the servo on the first channel of the PCA9685, but you're trying to control channel 6?

[queen fjord]

I initially had it on 1, couldn't control it, changed it to 6 to see if it would still flicker

the number I put on the code that is

the servo remained on channel 1 all the time

[cedar mountain]

Possibly you need channel 0?

[queen fjord]

could be the case, let's see

Setting it to 0 stopped the flickering, but still can't control it

nvm it didn't stop

[cedar mountain]

Having a glitch at power-up isn't too concerning... sort of expected if you don't take care with the state of the pins.

[queen fjord]

Opening the serial monitor does it too

[cedar mountain]

I'm not sure why the control isn't working. You say it's printing reasonable values for the microseconds you're commanding?

[queen fjord]

yep

0 gives 600

90 gives 1500

180 2400

as it should

[cedar mountain]

Did you try running the PCA9685 library's example code previously?

[queen fjord]

Adafruit's example code didn't do anything, but I found a simpler version in the internet that did

Let me test it one more time to make sure

[cedar mountain]

That's a little weird. It might be worth digging into the differences to see what's causing one to fail.

[queen fjord]

I tried to no avail

Ok so I tried it

It made the first channel servo flicker

nothing else

I have another servo connected to the 5th channel that didn't move

I am 100% sure that both servos work perfectly

[cedar mountain]

There's also that setOscillatorFrequency() function. It's possibly the servo is sensitive enough to care about the exact period.

[queen fjord]

oh boy that's gonna be fun to set up

[cedar mountain]

I'd just try a few values near the nominal value... 23, 24, 25, 26, 27 MHz, etc.

Or see if that other example code had a value which worked.

[queen fjord]

this is the example that I found worked with all servos

#include <Wire.h>
#include <Adafruit_PWMServoDriver.h>
 
// called this way, it uses the default address 0x40
Adafruit_PWMServoDriver pwm = Adafruit_PWMServoDriver();
 
void setup() {
  Serial.begin(9600);
  Serial.println("16 channel PWM test!");
  pwm.begin();
  pwm.setPWMFreq(1600);  // This is the maximum PWM frequency
 
  // save I2C bitrate
  uint8_t twbrbackup = TWBR;
  // must be changed after calling Wire.begin() (inside pwm.begin())
  TWBR = 12; // upgrade to 400KHz!
 
}
 
void loop() {
  // Drive each PWM in a 'wave'
  for (uint16_t i=0; i<4096; i += 8) 
  {
    for (uint8_t pwmnum=0; pwmnum < 16; pwmnum++) 
    {
      pwm.setPWM(pwmnum, 0, (i + (4096/16)*pwmnum) % 4096 );
    }
  }
}

[cedar mountain]

That's unexpected, since it's using 1600 Hz as a frequency versus a 50 Hz servo standard.

[queen fjord]

yup

maybe I just have to use setPWM?

[cedar mountain]

Stupid question... are your servos expecting an analog signal instead of a 50 Hz digital servo signal?

[queen fjord]

I have tested them using digital pins, so I assume digital

What would the correct calculation for the pwm.setPWM parameters be? I guess I should try using that instead of microseconds

[cedar mountain]

It'd be the setPWM(num, 0, on_pulse) formula from above.

You can also use the map() function to calculate it, with on_pulse = map(microsec, 0, 20000, 0, 4096).

[queen fjord]

So like that?

[cedar mountain]

Yes, looks correct to me.

Oh, hang on...

[gray holly]

Hey ed, thanks for your help, my clock works great now

and lewstherin

[cedar mountain]

The loop() function is wiping testdegree on every iteration and using the undefined value to set the PWM parameters. You should rearrange things so that it only runs the setPWM code inside the if clause.

Glad to hear it @gray holly !

[queen fjord]

So like this?

[cedar mountain]

Yes. Was it previously continually printing (garbage?) values continually even if you didn't type anything?

[queen fjord]

It printed the value for like a second then it returned to 0

oh and one thing

[cedar mountain]

That might explain the glitch, then, since those 0's would have been sent to the servo.

[queen fjord]

the on_pulse variable has a value of 122 at input 0

should that be the case?

[cedar mountain]

Seems about right, yeah. 122/4096 is about 600/20000.

[queen fjord]

yeah it's not doing anything

servos not moving at all

and this is the serial monitor output after opening the serial monitor, typing in 100, then typing in 150

After printing the mapped value, it prints a 122 aka 0

[cedar mountain]

I'm running out of ideas, I'm afraid. I don't understand why the 1600 Hz servo test code works at all, versus the 600-2400us pulses you measured elsewhere which worked.

I think you might be seeing a one-second default timeout on serial.available(), so it's returning 0 in that case. You might need to set a different timeout for the serial library.

[queen fjord]

hm

should I try setting it up with a potentiometer?

[cedar mountain]

Possibly. Or maybe retry the setup with the 600-2400usec pulses that worked previously. Was that just with the Arduino Servo library?

[queen fjord]

Yes

Used the default arduino servo library to test both servos

both of them worked

[cedar mountain]

Weird. Yeah, I'm running out of ideas, sorry...

[queen fjord]

No need to worry about it

you don't owe me any help

I should probably talk to someone who's used this thing before

[cedar mountain]

The only other thing to double check is whether you have the correct channel numbers on the PCA9685 pins. The other example code tries all 16 channels, whereas you've been picking particular individual ones. So if things were just on the wrong pin...

[queen fjord]

I tried putting the servo on the first channel

then tried using 0, 1, and 2 in the code

neither worked

[cedar mountain]

It might be worth just having the code set every channel, just in case.

Could be something incredibly silly like the channels actually being counted right to left instead of left to right on the board or whatever.

[queen fjord]

If it's that I'm going to hang myself

let me try it

Seems like I live another day

[cedar mountain]

Heh heh

[queen fjord]

wait

the servo that wasn't moving moved a bit

I may not live another day

it's twitching every so often if the serial monitor is often

man today is not my day

hold on

hold on why are calculations off

[cedar mountain]

How are they off?

[queen fjord]

0 is 600

180 is somehow 500 something

ok it's working normally now

Alright I give up for today

it's 4AM

I've been trying to figure this out for hours now

[cedar mountain]

Ouch, yeah, get some sleep...

[queen fjord]

one thing

am I allowed to swear?

[cedar mountain]

Only mildly. The bot will censor most swear words. 😅

[queen fjord]

let's try it

Oh come on

fine

It works!

And I will have to go hang myself too now

The channels were indeed counted from the opposite side

and on top of that I forgot to set the frequency back to 50

Alright one more thing before I go hang myself to sleep, how do I know when a servo has reached it's maximum and minimum pulse length?

[cedar mountain]

Whew! And the universe realigns to make sense again...

[queen fjord]

Because if I read correctly it was possible to make them go past 180 degrees but it could damage them

And yes, the universe realigned itself

thank you very much for your help

[cedar mountain]

It shouldn't damage them per se, I think, but sometimes you can get a scenario where the servo controller tries to drive the motor past a hard stop, so it'll just burn power or vibrate instead of staying calmly in place at a particular angle.

[queen fjord]

I see

so I should just try to get it to reach 180 and that's all?

[cedar mountain]

Yep. Many servos have an internal range limit in their feedback anyway, so they'll just go to their max and stop there even if the input signal is commanding 200 degrees or whatever.

[queen fjord]

Alrighty

Thank you so much for all

You saved me from being tormented in my sleep lol

[cedar mountain]

Heh heh, sure thing. Nice debugging!

[queen fjord]

Yup, have a great day/night

[proven mauve]

Okay, so I'm trying to wrap my head about something. Arduinos output 0-5v and can be used as sound output with the right software, like Auduino or Mozzi. So, to hook it to other audio equipment you want to change the voltage levels of the output to a range around -1.7v to +1.7v. So, I've learned you can use a voltage divider to lower the 0-5v output to 0-3.4v output, and then put an electrolytic capacitor in series with the +v audio output, connecting the audio gnd to 0v. Then your audio will swing +1.7v to -1.7v. I've tested this, and it works well, and feeding it 0-5v it will output -2.5 to +2.5v. What I don't understand is why this happens, and googling I can't even find anyone talking about it. I just see people using opamps to do things like controlling the swing like this. Does anyone know the name of this principle or somewhere to do reading on it?

[lost nest]

I've been trying to find a good entry-level tutorial on this and haven't had much luck. http://www.ee.ic.ac.uk/pcheung/teaching/de1_ee/Lectures/Lecture 7 - Capacitors and Inductors (x1).pdf could be of use.

The short answer: You can think of your "0-5V" signal as "2.5V DC +/- 2.5V AC"

And when you feed it through a DC-block cap, it takes away the 2.5V DC, and you're left with just the +/- 2.5V AC.

[lusty mountain]

Anyone know how I can isolate two circuits running off the same battery?

I have an arduino and a relay running off the same battery, and when the arduino trips the relay, it causes the arduino to reset

[lost nest]

I'd recommend just adding a hefty capacitor to the battery.

[lusty mountain]

When I power the relay with a separate power source, everything works fine

That makes sense. I have a couple of beefy caps that might reduce the voltage spike

[lost nest]

Yeah, I strongly suspect you're just browning out the input supply to the Arduino with the current spike into the relay.

[lusty mountain]

Any suggestions on what rating cap I should use?

[lost nest]

At least your battery voltage, ideally at least 2x your battery voltage

[lusty mountain]

I have a couple 2700uF caps, but those seem overkill lmao

[lost nest]

Beefy electrolytics are great for this

[lusty mountain]

I haven't dealt with EMI before so i'm a bit out of the loop

[lost nest]

2.7mF is definitely a bit overkill, but also you can always walk it back later

[lusty mountain]

But it wouldn't hurt, right?

[lost nest]

It's unlikely to

especially if they're electrolytics (which they almost certainly are, you'd know if you had a supercap/giant ceramic array)

[lusty mountain]

Yeah they're electrolytics

[lost nest]

They'll inrush your battery a bit when first plugged in and charging

[lusty mountain]

That's mainly what i'm worried about

[lost nest]

so if you have a mechanical switch for turning the battery on and off, that's the most likely place to have a problem.

[lusty mountain]

When I charge them, they definitely spark a bit

I'll see what caps I have and tinker with it tomorrow

[lost nest]

for great hacks you could also put two in series

to halve the capacitance

[lusty mountain]

I didn't know that

[lost nest]

It's not generally advised, especially for polar caps

Honestly, you're not going to do tremendously more damage with just one cap than you've already done if you've connected them and seen sparks. So you may as well just try the current setup and see if it solves your problem. If it does you can look to drop. I don't know how big your relay is but I'm guessing ~220uF-470uF of cap is probably a sweet spot

[lusty mountain]

Alright, thanks

[proven mauve]

Thank you much, it looks like it's called DC Blocking. Gonna have to look more into it.

[gray holly]

So, my neopixel strips work on my arduino nano just fine, but they don't work if I put my rtc module on it

I have a ds3231 module

my schematics are like this

I don't get anything on the setserial code for the ds3231 library

[gilded swift]

does that RTC have pullups on it for the SDA/SCL lines?

[gray holly]

what are pullups

[gilded swift]

pullups are resistors that ensure voltage levels for SDA/SCL lines

[gray holly]

Can you point it out on a pic of it?

Like this one

Could I use these holes instead?

[gilded swift]

it appears that it has pullup resistors on it. I just wnated to make sure

[gray holly]

Other sketches that concern only with leds work

but any sketch with rtc and leds won't work if the rtc is connected

[gilded swift]

does the rtc work when the LEDs are not present?

[gray holly]

no

and the leds don't work if the rtc is present

[gilded swift]

so the RTC doesn't work at all?

[gray holly]

no

[gilded swift]

can you share the code you are using?

[gray holly]

brb

[fallen jetty]

hello everyone.
I've got the following board -> pinouts:
https://learn.adafruit.com/adafruit-huzzah32-esp32-feather/pinouts

question. If I have a Lithium Battery connected to the proper connector, can I use the USB Pinout (solder usb breakout board to it) to load this battery?

so I would solder a usb breakout board to usb and gnd to have an separate usb for loading, if possible

[gray holly]

Back

[gilded swift]

you should be able to use usb and battery at the same time

[fallen jetty]

yeah that's possible with the normal usb slot and battery, but if I add a second usb connector to the usb pin, will this charge the battery as well?

[gilded swift]

why would you need a second USB adapter?

[gray holly]

https://content.instructables.com/ORIG/F1I/L1NY/K6ZAL3M4/F1IL1NYK6ZAL3M4.ino

[fallen jetty]

I 3d printed a casing, but this casing is to small to use the normal usb. So I will add one on a different spot and connect it to the board

[gilded swift]

@gray holly so far I can see that you never call the DS3231_set(setupTime; in your code

so it never actually sets the RTC

and it appears you only ever call the setup on a button press, it would make sense that you would configure you RTC in

void setup() {
 //your RTC method here
}```

[queen fjord]

Hey guys, I've been having some issues with coding a turret that use an MPU6050 and a PCA9685

The idea is that my code takes the Yaw Pitch Roll input from the MPU6050, maps it into pulse length, and sends it to the respective servos using:

pwm.writeMicroseconds(0,yaw);
pwm.writeMicroseconds(1,pitch);
pwm.writeMicroseconds(2,roll);

When I don't include the piece of code above, the MPU6050 readings are correct

But when I do, the readings from the MPU6050 go crazy

Seems like when I send data to the PCA9685, it interferes with the MPU6050 readings

This is how the whole calculation process goes:

  // MPU6050 READINGS

  Serial.print("ypr\t");
  Serial.print(ypr[0] * 180/M_PI);
  Serial.print("\t");
  Serial.print(ypr[1] * (180/M_PI) * 2);
  Serial.print("\t");
  Serial.println(ypr[2] * (180/M_PI) * 2);
  
  // CALCULATIONS
  
  int rawyaw = ypr[0] * (180/M_PI);
  int rawpitch = ypr[1] * (180/M_PI) * 2;
  int rawroll = ypr[2] * (180/M_PI) * 2;
  int yaw =  map(rawyaw, -90, 90, 600, 2400);
  int pitch =  map(rawpitch, -90, 90, 600, 2400);
  int roll =  map(rawroll, -90, 90, 600, 2400);
  
  pwm.writeMicroseconds(0,yaw);
  pwm.writeMicroseconds(1,pitch);
  pwm.writeMicroseconds(2,roll);
   }

}

Again, if I delete the piece of code that I showed, the readings work just fine. Anyone has any idea of how I could fix the issue?

[north stream]

Could be a pin conflict, electrical interference, or power supply interference.

[queen fjord]

Hmmmm

The MPU6050 and PCA9685 are connected to the Arduino's 3V3 and GND

And also connected to it through SDA and SCL which both have 4.7k resistors going to the 3V3

the servos are powered externally

[north stream]

Powering the servos externally is a good idea, and will probably avoid power supply interference

Can you try putting longer wires on the MPU6050 and seeing if it operates better if it's farther away? That's one way to narrow it down.

[queen fjord]

Sure, I'll try right now

Nope, no changes

Ok so someone on another discord told me to try with a single servo in the code

it worked

But what do I do with the other 2 lol

[north stream]

Hmm, that's useful information. So it's probably not electrical interference. Do the other two work, one at a time?

[queen fjord]

Yes

All of them work one at a time

Another piece of useful information

It will go crazy even if the servos aren't connected

As long as I try to control 2 or more servos in the code, it will go crazy

[loud horizon]

Hello, I've connected two 10 segment leds to an arduino nano and it started blinking for some reason when I hooked it up. What code should I use to control the led's? I'm going to use this for racing using simhub.

I've hooked it up to a purple/pink hc595 driver

https://youtu.be/Q2DdccKuso0

[elder hare]

too many initializer values

class StripSettings
{
    private:
        uint8_t m_Pattern_Color_Gradient_Palette_Buffer[256];
    public:
        StripSettings()
        {
            m_Pattern_Color_Gradient_Palette_Buffer[256] = { 0, 255, 0, 0, 255, 0, 255, 0 };
        }
};

Why?

[north stream]

Drop the [256] from the initialization line

[green thunder]

So it does in fact seem that DMA is possible on the QT PY. Though my understanding is still quite limited, I noticed that many other boards allow the use of the MOSI pin for NeoPixel DMA, with the caveat that you couldn't use SPI anymore. After digging around in the ZeroDMA and the NeoPixel_ZeroDMA code for a bit, as well as the 'variant.h' file for QT PY in the core, I figured out that I could modify the 'pins.h' file in the NeoPixel_ZeroDMA library to add a definition for the QT PY. So, I added:

#if defined(ADAFRUIT_QTPY_M0) &sercom2, SERCOM2, SERCOM2_DMAC_ID_TX, 10, SPI_PAD_2_SCK_3, PIO_SERCOM_ALT, #endif

and it worked!

There may be other pin/sercom combinations possible that would not require giving up SPI, but at the moment I still haven't quite pieced together how to figure that out. I'm not entirely confident that I've done the correct thing here, but it does seem very promising.

[gilded swift]

@green thunder as far as I am aware, SPI can be assigned on a few different pins for the samd21e18a

Which is what is on the qt py

So technically you could change the pins assigned for SPI as needed for DMA

I believe the UART pins also should support DMA as well

[green thunder]

@gilded swift I know that other pins could probably be used, but due to my limited knowledge of samd21 processors and after reading the comments at the top of https://github.com/adafruit/Adafruit_NeoPixel_ZeroDMA/blob/master/pins.h, I'm not entirely sure how to figure out which pins should be used lol. It seems like there is probably some "ideal" combination of pins/sercoms per board, but at the moment I have no idea what the best way to figure out what that would be.

[gilded swift]

Well, if ya ever need help making a custom variant, I can probably help 🙂

[green thunder]

That would be most appreciated! 🙂

I'm still digging around trying to understand how all of this fits together, but ultimately I'd like to add (or help add) official support for DMA NeoPixels on QT PY. I'm going to continue just kinda poking around for a bit, but if you have any advice on where I could find more information specifically about why the choices were made in 'pins.h' that would be really helpful. Also, would doing this require modifying the variant.h/cpp files in the core for QT PY?

[green thunder]

Upon further exploration, it seems there are a couple of other pins that can be used by NeoPixel_ZeroDMA on the QT PY, for a total of 3: A6(TX), SDA(D4), and A10(MOSI). Of course, because the NeoPixels will be taking up a critical output pin, that means in each case you'd have to sacrifice UART, I2C, or SPI capable pins, respectively. I think this would be an acceptable tradeoff to make personally for the ability to have zero-cpu NeoPixel driving.

[loud horizon]

No, I uploaded the original code

two 10 segment led's, with ground pin on each side to a nano

one ledkeeps flashing

[pine bramble]

@green thunder Yeah but you're pushing the qt py paradigm way out of its niche.

You can squeeze a few more pins out of the SeeSaw and save a buck or two.

(ATSAM09D iirc)

[green thunder]

It's not so much that I'm trying to drive a lot of NeoPixels, or do anything like what NeoPXL8 is trying to do. Mainly I just want to e.g. be able to use NeoPixels and use an IR Receiver, or a servo

Neither one of those things seem outside of it's niche to me

[pine bramble]

What she gave you eleven pins on that qt py? (I had no idea; I overlooked it).

[green thunder]

yeah 11 pins

[pine bramble]

Thats' pretty good especially in pins vs dollars to buy one. ;)

[green thunder]

Indeed! 😄

I have become pretty enamored with it lol

[pine bramble]

Generally Limor is pretty good at figuring out how to optimize for a given design pinout.

variant.cpp is what encapsulates that decision tree, iirc.

[green thunder]

I think she was here too! I can understand that perhaps she decided maybe it would simply just not be a good idea to allow DMA NeoPixels given the tradeoffs involved

[pine bramble]

well they gotta pay the rent, so it's germane to sell a lot of $6 boards in volume, which means pleasing the needs of the widest variety of likely buyers.

[green thunder]

ha that's a good point

[stuck coral]

Sorry Im trying to understand the issue, ZeroDMA should be using a SERCOM from the SAMD21

You should be able to use any sercom

[pine bramble]

So modify the lib not the variant.cpp for the core. ;)

[stuck coral]

Oh okay, I just had to read a little more, duh

[green thunder]

well, for my part, I might make the case that a) since I rarely need SPI on the QT PY, and b) since the tradeoff to use the MOSI pin on other boards already exists, that it might be nice to at least have the option for one DMA NeoPixel pin on the QT PY 😄

[stuck coral]

Yep, should be fairly simple to add in

[green thunder]

yeah @pine bramble that's what I did, it works great!

Added this to 'pins.h':

#if defined(ADAFRUIT_QTPY_M0) &sercom0, SERCOM0, SERCOM0_DMAC_ID_TX, A6, SPI_PAD_2_SCK_3, PIO_SERCOM_ALT, &sercom1, SERCOM1, SERCOM1_DMAC_ID_TX, SDA, SPI_PAD_0_SCK_1, PIO_SERCOM, &sercom2, SERCOM2, SERCOM2_DMAC_ID_TX, MOSI, SPI_PAD_2_SCK_3, PIO_SERCOM_ALT, #endif

[pine bramble]

I was able to easily upload a functional program to the original CPX crickit, treating it as an ItsyBitsyM0.

(ignoring its design use as an adjunct board; I made it the primary board).

[green thunder]

Ha that's awesome

[pine bramble]

It wasn't too bad, and I learned a lot about how the core code was written/designed, in doing it.

[stuck coral]

Could you make a pull request to the core fix it for the next guy?

[pine bramble]

I was pretty close to writing my own board support package for the crickit but I didn't follow through - was proof of concept.

[loud horizon]

Could anyone help me with this error message?

[green thunder]

@stuck coral I was actually in the process of looking into doing just that 😄

[stuck coral]

Oh sorry

Lol, you want the lib not the core

[green thunder]

right lol

I forked it, I'm gonna make the change in my fork and make a PR with a comment explaining the rationale

at least for pin A10

[pine bramble]

@loud horizon What do you got there, moose.

[stuck coral]

Yep, no other way on github 😜

[loud horizon]

@loud horizon What do you got there, moose.
@pine bramble

One blinking light and lost hope

[pine bramble]

That is awesome dense colorful wiring!

I love it.

What's the target board? It looks pretty small.

[rough torrent]

It's really hard reading the text, (next time, please copy the error messages and paste it directly in Discord. Also inclose it in backticks) but I think it's an upload error. Make sure that the Arduino is connected, appears as a COM port, and you can read the Serial.println stuff coming out of it.

[stuck coral]

AVRDude cannot connect to target

[loud horizon]

Arduino: 1.8.13 (Windows 10), Board: "Arduino Nano, ATmega328P"

Sketch uses 3024 bytes (9%) of program storage space. Maximum is 30720 bytes.

Global variables use 210 bytes (10%) of dynamic memory, leaving 1838 bytes for local variables. Maximum is 2048 bytes.

avrdude: stk500_recv(): programmer is not responding

avrdude: stk500_getsync() attempt 1 of 10: not in sync: resp=0x61

avrdude: stk500_recv(): programmer is not responding

avrdude: stk500_getsync() attempt 2 of 10: not in sync: resp=0x61

avrdude: stk500_recv(): programmer is not responding

avrdude: stk500_getsync() attempt 3 of 10: not in sync: resp=0x61

avrdude: stk500_recv(): programmer is not responding

avrdude: stk500_getsync() attempt 4 of 10: not in sync: resp=0x61

avrdude: stk500_recv(): programmer is not responding

avrdude: stk500_getsync() attempt 5 of 10: not in sync: resp=0x61

avrdude: stk500_recv(): programmer is not responding

avrdude: stk500_getsync() attempt 6 of 10: not in sync: resp=0x61

avrdude: stk500_recv(): programmer is not responding

avrdude: stk500_getsync() attempt 7 of 10: not in sync: resp=0x61

avrdude: stk500_recv(): programmer is not responding

avrdude: stk500_getsync() attempt 8 of 10: not in sync: resp=0x61

avrdude: stk500_recv(): programmer is not responding

avrdude: stk500_getsync() attempt 9 of 10: not in sync: resp=0x61

avrdude: stk500_recv(): programmer is not responding

avrdude: stk500_getsync() attempt 10 of 10: not in sync: resp=0x61

Problem uploading to board. See http://www.arduino.cc/en/Guide/Troubleshooting#upload for suggestions.

This report would have more information with
"Show verbose output during compilation"
option enabled in File -> Preferences.

[pine bramble]

alright so that's a 328p which does not have built-in USB.
It's talking to a bridge chip like CP2104.

Bodger said there's been at least three solutions to the bridge problem over the years.

[loud horizon]

what would that be?

[pine bramble]

They're all various chips used to do the same job.

What's the operating system?

[stuck coral]

Looks like windows 10 to me

[loud horizon]

windows 10

[pine bramble]

What's the exact target board called? Nano?

[loud horizon]

arduino nano (com8)

com8 is the port

[pine bramble]

"I have an Arduino Nano (ATMega 328p-based) and am in Windows 10. Can't upload to the target; the IDE is using AVRdude but I get an error."

Something like that is your question, I think.

[loud horizon]

i cannot get the code to fully upload onto my nano

that's all I know

[pine bramble]

Yeah I am not familiar with Windows at all.

[reef ravine]

did you try "Nano | Old bootloader"?

[loud horizon]

ill try it

same message

Arduino: 1.8.13 (Windows 10), Board: "Arduino Nano, ATmega328P (Old Bootloader)"

Sketch uses 3024 bytes (9%) of program storage space. Maximum is 30720 bytes.

Global variables use 210 bytes (10%) of dynamic memory, leaving 1838 bytes for local variables. Maximum is 2048 bytes.

avrdude: stk500_recv(): programmer is not responding

avrdude: stk500_getsync() attempt 1 of 10: not in sync: resp=0x0b

avrdude: stk500_recv(): programmer is not responding

avrdude: stk500_getsync() attempt 2 of 10: not in sync: resp=0x0b

avrdude: stk500_recv(): programmer is not responding

avrdude: stk500_getsync() attempt 3 of 10: not in sync: resp=0x0b

avrdude: stk500_recv(): programmer is not responding

avrdude: stk500_getsync() attempt 4 of 10: not in sync: resp=0x0b

avrdude: stk500_recv(): programmer is not responding

avrdude: stk500_getsync() attempt 5 of 10: not in sync: resp=0x0b

avrdude: stk500_recv(): programmer is not responding

avrdude: stk500_getsync() attempt 6 of 10: not in sync: resp=0x0b

avrdude: stk500_recv(): programmer is not responding

avrdude: stk500_getsync() attempt 7 of 10: not in sync: resp=0x0b

avrdude: stk500_recv(): programmer is not responding

avrdude: stk500_getsync() attempt 8 of 10: not in sync: resp=0x0b

avrdude: stk500_recv(): programmer is not responding

avrdude: stk500_getsync() attempt 9 of 10: not in sync: resp=0x0b

avrdude: stk500_recv(): programmer is not responding

avrdude: stk500_getsync() attempt 10 of 10: not in sync: resp=0x0b

Problem uploading to board. See http://www.arduino.cc/en/Guide/Troubleshooting#upload for suggestions.

This report would have more information with
"Show verbose output during compilation"
option enabled in File -> Preferences.

[reef ravine]

did it upload before you wired it?

[rough torrent]

Well, not exactly the same message. The response is different.

[loud horizon]

upload before wiring it?

I can try another nano of the same type

[reef ravine]

if it's easy hook up the bare board to make sure you don't have a PC side issue

you may or may not need to flip off ATMEGA Old Bootloader

[loud horizon]

I'll try another nano and upload on the regular atmega

[reef ravine]

some need it - some don't

[pine bramble]

Yeah a known good board is a great differential diagnosis

[loud horizon]

Arduino: 1.8.13 (Windows 10), Board: "Arduino Nano, ATmega328P"

Sketch uses 3024 bytes (9%) of program storage space. Maximum is 30720 bytes.

Global variables use 210 bytes (10%) of dynamic memory, leaving 1838 bytes for local variables. Maximum is 2048 bytes.

An error occurred while uploading the sketch

This report would have more information with
"Show verbose output during compilation"
option enabled in File -> Preferences.

^from using a different nano

[reef ravine]

flip the boot loader

[loud horizon]

old one now?

[reef ravine]

yes

[loud horizon]

Arduino: 1.8.13 (Windows 10), Board: "Arduino Nano, ATmega328P (Old Bootloader)"

Sketch uses 3024 bytes (9%) of program storage space. Maximum is 30720 bytes.

Global variables use 210 bytes (10%) of dynamic memory, leaving 1838 bytes for local variables. Maximum is 2048 bytes.

An error occurred while uploading the sketch

This report would have more information with
"Show verbose output during compilation"
option enabled in File -> Preferences.

[reef ravine]

is it a power only USB cable?

[loud horizon]

im using my ti84 color edition cord

[reef ravine]

ok so long as it passes data from the TI to the PC ok

does it "bing" when you plug the new one in?

[pine bramble]

nice verb

[loud horizon]

the l led is blinking on the naked arduino

*L

[reef ravine]

probably pre-loaded with "Blink"

so it has power

[pine bramble]

So then this has been going on for hours, as that's the same result the OP had many hours ago.

Sounds like the target has not once received new, locally-authored firmware. Ever.

[reef ravine]

ohhh hours eh?

didn't scroll up....

[pine bramble]

Unless I'm conflating two people, seems it's been a while.

[reef ravine]

common issue 🙂

[loud horizon]

man I wanted this work so bad

any ideas as to why its giving me errors?

[pine bramble]

I really am going to buy an AVR within 30 days from now. You read about it here, first.
sniper's got me convinced to go with 32u4 rather than 328p

[reef ravine]

does it make a noise when you plug it in?

yeah get with the 2020's and go 32u4

[pine bramble]

I'm also gonna get one of those fancy new Adafruit SPI debugging doo-dads with the TFT display.

[loud horizon]

I got a ws2812 led to work on my simhub display for rev lights on the same type of nano

[pine bramble]

I think I decided on ItsyBitsy, 5V logic, 16 MHz 32u4 to be compatible (the other Itsy option is 3.3v at 8 MHz).

[loud horizon]

it makes the usb cable noise when I plug it in

[reef ravine]

ok, open Device Manager

if you don't have a bunch of 3.3v stuff...

[pine bramble]

SPIDriver by Excamera Labs
http://adafru.it/4268

[loud horizon]

there's com5 and com8 to choose from

[reef ravine]

which shows up as an ATMEGA?

[pine bramble]

I want to be maximally compatible with the people I'm trying to help (people I have not met as of 31 Dec 2020, in the Arduino Discord).

[pine bramble]

I'm tired of saying "I don't know a thing about AVR sorry" lol