13
Why is there a clock in my arduino?
Because that is how computers and microcontrollers, etc, work.
With a 16mhz clock, each line of my program will take 1/16000000 second, right?
No.
even a very very big line will only take 1/16000000 of a second ?
No.
The clock defines at what speed the machine code instructions are fetched from memory and ...
13
The ceramic resonator in yellow is connected to the main MCU, and the crystal in red is connected to the other MCU. USB requires tighter tolerances than is available from a resonator therefore a crystal must be used instead.
answered Nov 7 '16 at 23:20
Ignacio Vazquez-Abrams
17.1k1 gold badge23 silver badges32 bronze badges
10
Not answered in @Majenko's post: Yes, with a 8MHz clock, each line will take the double of the time. Unless the line waits for something that is not clock driven - e.g. external input.
In addition to @Majenko's answer, a CPU has a clock to ensure the instructions are complete, before the next step starts. A CPU is made out of lots of transistors (I found a ...
10
The clock speed selected will affect both delay*() and millis()/micros() as well as the functions in <util/delay.h>, therefore you must use the target system's correct clock speed if you are using any of these functions. This can be easily done by editing boards.txt and copying an existing entry for the Uno and changing the value of the f_cpu parameter ...
answered Mar 17 '14 at 11:27
Ignacio Vazquez-Abrams
17.1k1 gold badge23 silver badges32 bronze badges
7
This outputs 8 MHz on pin 9:
#ifdef __AVR_ATmega2560__
const byte CLOCKOUT = 11; // Mega 2560
#else
const byte CLOCKOUT = 9; // Uno, Duemilanove, etc.
#endif
void setup ()
{
// set up 8 MHz timer on CLOCKOUT (OC1A)
pinMode (CLOCKOUT, OUTPUT);
// set up Timer 1
TCCR1A = bit (COM1A0); // toggle OC1A on Compare Match
TCCR1B = bit (WGM12) |...
7
Officially, no. Practically, often but not always.
If I follow the datasheet to the the letter, this configuration should not work. BUT... After some test, it did work.
The data sheet does not say that it "should not" work, it says that it is that the part is not qualified for operation at 16 MHz below some voltage higher than you want to use. That ...
6
There's no hardware reason (as far as I'm aware) that you couldn't run a suitable Arduino at 20 MHz. You'd run into minor timing inaccuracies in software though, as the standard Arduino configuration apparently works best if the clock speed is either 8 or 16MHz.
I suspect this is largely for historical reasons. Early Arduino boards were based on chips which ...
5
Do I need an oscillator installed just so I can wake it it up and tell it to not use an oscillator?
Yes, yes you do.
Pick up a 8-16MHz crystal (not an oscillator) and two 15pF capacitors and use very, very short wires to connect them to the chip as shown in the "System Clock and Clock Options" section of the '328P datasheet. You will then be able to ...
answered Oct 27 '14 at 5:15
Ignacio Vazquez-Abrams
17.1k1 gold badge23 silver badges32 bronze badges
5
The default configuration of these chips, as they come from the factory,
is to use their internal 8 MHz RC oscillator downscaled at 1 MHz. So you
do not need any extra oscillator to program them.
Once you program the chip the first time, if you configure it to use
an external resonator/oscillator, then you do need to have that
attached in order to reprogram ...
5
No, you won't be able to build a 16 MHz clock signal using a 555 timer. That's far beyond the 555 capabilities (see N.B.).
If you could squeeze out of the 555 a mere 1 MHz (which I seriously doubt), the clock stability and jitter would be so bad that it won't be usable but for simple sketches not requiring any timing accuracy. At that point, you would be ...
4
Sharing a crystal between multiple devices is a very bad idea. The crystal is only a piece of the oscillator, the rest of it being a small bit of circuitry within the IC itself. Having multiple ICs connected to the same crystal will cause them to fight over who is driving it.
The datasheet for the CH340T/R, which I was able to find in English, says the ...
answered Dec 23 '14 at 2:57
Ignacio Vazquez-Abrams
17.1k1 gold badge23 silver badges32 bronze badges
4
You can, and I do.
It's not technically an Arduino anymore, I guess. I made my own board, replacing a resonator on a board will require some fine soldering, but it is definitely possible.
There are some gotchas: Anything time-related (e.g. millis() ) will be run faster than is should - 25% more millis() per second. Also, the bootloader expects 16mhz, so ...
4
I doubt if that would be fast enough. See my thread about connecting to a VGA monitor.
Borrowing from that page, so as to not make a link-only answer ...
For 640x480 pixels of active video you have something like this:
Vertical Sync
Let's start with the vertical sync pulses. In fact, we'll show how all the timing data can be derived from three figures:
...
4
If you are using an LED to indicate that your sketch is running, then that may be the problem.
For example, if you have a fairly average red LED, it might have a forward voltage of 2.0v @ 20ma (refer to your spec sheet for your particular LED). Putting these values into the LED calculator, it would tell you that you need a 150 ohm resistor. You might have ...
4
It was a brownout fuse problem.
Changing boards.txt from:
atmega328bb.bootloader.extended_fuses=0x05
to
atmega328bb.bootloader.extended_fuses=0x02
solved it for me (*).
The chip and red LED now work at 3.01V (1xAA alkaline and 1xAA rechargeable) and 3.19V (2xAA alkaline).
(*) not sure why I had to invert the bits (0000 0010=0x02 versus 0000 0101=0x05)...
4
You can set the clock prescaler for that:
void setup() {
noInterrupts();
CLKPR = _BV(CLKPCE); // enable change of the clock prescaler
CLKPR = _BV(CLKPS0); // divide frequency by 2
interrupts();
}
This is explained in section 10.12 and 10.13 of the
ATmega2560
datasheet.
Of course, changing the clock frequency will mess with the time-...
4
A little test, not very scientific, but shows it runs fine on 3.3V - https://www.youtube.com/watch?v=If0eOa0Uj5I.
Again, if you would like to stay withing the specification just use a "clock divider/2" and run your MCU at 8MHz if they have a 16MHz Crystal:
clock_prescale_set(clock_div_2)
http://www.nongnu.org/avr-libc/user-manual/group__avr__power.html
4
According to the datasheet you need 3.78V to run at 16 MHz, and 3.3V isn't that far off. I would be surprised, for example, if it stopped working at 3.77V, because that would mean that the operating at 3.78V was marginal.
I have operated these chips out of spec, for example under voltage, or with a 24 MHz crystal. They seem to work. However I would guess ...
4
You can set the clock prescaler at run time. For example:
#include <avr/power.h>
void setup()
{
clock_prescale_set(clock_div_8);
// etc...
}
More details in the documentation from
avr-libc.
Note that with this method your Digispark will still be overclocked when
booting, so it might not be 100% reliable. But it will be overclocked
only for ...
4
Yes, you could use a 555. You can also go even simpler and use an RC network. Even simpler still is to use the 1MHz RC network built in​ to the chip.
The main drawback of these methods, and the 555 particularly, is that of stability. Using a crystal and capacitors you can get very accurate timing at high frequencies. Using a 555 you can get ballpark timing ...
4
If it requires a crystal, as the ones in question clearly state, then obviously you need to connect a crystal (and load capacitors). Or use a ceramic resonator.
Only those labelled as internal are internal. Everything else is external.
3
For your external source to be an SPI bus master, it has to be the only one generating the clock signal. Yet, calling SPI.transfer() also generates a clock signal from the Arduino; both devices are attempting to drive the clock line simultaneously, hence your problems.
You have to bit-bang this or manipulate the SAM registers yourself to configure it as a ...
3
You can use one of the PWM pins on Arduino to output a PWM signal.
If you want a constant clock, you need to set the duty cycle of the PWM to be 0.5, i.e. 50%.
Syntax: analogWrite(pin, value) where the parameter "value" is the duty cycle ranges from 0 (always off) to 255 (always on) since it is a 8-bit PWM generator inside Arduino.
If you need a PWM wave ...
3
I've built a few 3.3v home-brew-duinos with 16Mhz crystals (because I have a bag of 50 xtals!) - they run fine in the conditions I've put them in so far (on my desk). Some of them may be closer to their margins than others, and might fail at at some otherwise in-spec temperature, for example. If these were intended to go outside (enclosed, of course), I'd ...
3
The easiest way is to use the settings:
Tools > Board > Arduino Pro or Pro Mini
Tools > Processor > ATmega328 (3.3V, 8MHz)
The problem with that is the Pro Mini bootloader is 2kB and has a bug that causes an endless reset loop if you do a watchdog reset.
The better option is to use https://github.com/MCUdude/MiniCore, which includes the Optiboot ...
3
As I said in my comment, the idea of using the Arduino's own PWM as a
clock source is completely misguided. It will be no more accurate than
using millis() or micros(): both these functions and the PWM rely on
the same primary clock source, namely the ceramic resonator clocking the
chip.
Also, you should avoid using delay() for timekeeping. The time it
...
3
It depends on how you're programming it.
If you're using a bootloader to accept programming commands via a serial connection then you will need to use the clock method specified by the fuses currently programmed, since the chip at that point will actually be running code programmed on it.
If you're using ISP to program the chip out-of-band then you can ...
answered Apr 9 '17 at 14:00
Ignacio Vazquez-Abrams
17.1k1 gold badge23 silver badges32 bronze badges
3
Delay uses millis. It just waits examining millis until a certain time has passed.
Millis is driven by an interrupt. The only thing that stops it working is another interrupt (you can only have one interrupt executing at once), which is why you can't use delay in an interrupt.
3
Yes, things will break. The main culpret is this:
#define clockCyclesPerMicrosecond() ( F_CPU / 1000000L )
With any value below 1000000 the result of that calculation will be 0.
That means code like micros() will break, since it uses:
return ((m << 8) + t) * (64 / clockCyclesPerMicrosecond());
64/0 is ... um ... yes... And whatever that ends up ...
3
However, despite ESP8266 running at 160MHz, 10 times faster than nano, it's taking 3.5 ms to complete the loop, compare to 2.5 ms than that of nano.
There's a number of reasons for that. Chiefly:
The ESP8266 has no I2C hardware, so has to use software "bit-banging" to emulate I2C. That is slower to execute.
The ESP8266 also has to manage the WiFi hardware ...
Only top voted, non community-wiki answers of a minimum length are eligible
