Arduino Mini boot time

Hi,

I recently bought an official Arduino Mini. It works perfectly but when i power the device it takes close to 8 seconds before my sketch begins to run. If I run a sketch from the IDE it runs immediately after the programming is completed!

For comparison, my teensy 2.0 and my cheap Chinese Pro Mini clone both boot up almost instantly (in less than a second). Is my Arduino Mini broken or do I need to flash another bootloader?

Thanks.

It should be the same.

You have a Mini ? http://arduino.cc/en/Main/ArduinoBoardMini

Not a Micro, not a Mini Pro ?

Could you try a minimal sketch ? Blink a led or send a message to the serial terminal.

Yes, it is definitely a mini.

And as i said, it works fine. I tried with the Blink Example on all 3 boards: teensy 2.0, chinese clone arduino pro mini, and official mini. Only the official mini has 8 seconds.

As far as I know, only the ATmega8 NG has a bootloader with a delay of 6-8 seconds.
That is to be able to press the reset button to upload a sketch.
http://arduino.cc/en/Hacking/Bootloader

But then I read that very old Arduino boards also have a delay.
http://arduino.cc/en/Hacking/MiniBootloader

I assume that your board is with an ATmega328. So that should work the same at the other boards with an ATmega328(p).
Did you try to rewrite the official Arduino bootloader again ?
That might sound unnecessary, but some problems (only a small percentage) have gone after I rewrote the bootloader.
I would not know what else to do.
You could upload your minimal sketch between
[code][/code] tags.

Is something else connected to the board (to the reset line) when you test it ?
Could you measure the shortcut current of the reset line (should be 0.5 to 0.7mA) ? To check if the 10k pullup resistor is soldered properly.

Do you know which version you haven. Each version seems to have a slightly different reset circuit.

Thank you. The board is indeed an atmega328.

I was reading about bootloading... Not yet up for it. I've ordered a chinese avr programmer, but i might try burning it with the teensy as programmer.

There's no reason to upload the sketch here. It's the basic unmodified Blink example.

I can't read the board version, see the picture, the board was unreadable when i got it. Perhaps old stock?

Finally, i'm not sure what you mean by the resistor, there's only the reset line capacitor on the usb2serial reset line. If i remove everything except the led+resistor and power it with 5v, it takes 8 seconds to boot. I changed nothing.

The reset line has a 10k resistor from reset to 5V. If you shortcut the reset to ground that would be 0.5mA (or a little more). Could you check that ?

Is the usb2serial connected when you test this ? What happens if you remove that ?

According to the diagram, the reset line has a 10k pull-up resistor connected, so i do not see why an additional would help.

Yes, the usbserial is connected (with the reset line and a 10uF cap as instructed):

We have a little misunderstanding.

I took a good look at your photo of the backside. I'm pretty sure that it is model R5 (version 5). There is a 10k pullup resistor for the reset on that board. But if that smd component is badly soldered, the reset might do strange things. When nothing else is connected to the reset, you could measure the current from reset to GND. If that is 0.5 to 0.7mA the 10k resistor is okay.

The usbserial should be connected with a 100nF capacitor. The DTR line is used to reset the board to be able to upload a sketch. If you have 10uF connected in the wrong way, that could explain the long delay.

I assume that you connected everything right. And with only 5V power and the led, (nothing connected to reset line) you still have problems ? Then the bootloader is perhaps the cause of all the trouble.

Thank you for your quick response. I will investigate, and let you know.

RobvdVeer: Thank you for your quick response. I will investigate, and let you know.

As it turned out, i was using the wrong capacitor (10uF).Thank you for your time!

I'm glad you found the problem. The guide talkes about "0.1uF capacitor", but that is normally called a "100nF capacitor".

I had no 0.1uF laying about so i thought a 10 would do. Apparently not.