I am thinking of purchasing the Arduino, but I am somewhat confused by the disturbing lack of a power source attached to the circuit board :-/ . Does the circuit have a built in power supply, or is there a way to hook it up to a battery, or what?
Also, I noticed that you need a USB cable from the board to a PC... does the Arduino come with that, or is it separate?
It looks like an amazing product, perfect for my projects!
Power can be supplied either via USB or by virtually any plug pack (thats what the black socket is next to the USB).
The plug pack only has to be over 7v and under ~30v.
You should be able to get one cheaply or find a spare one around the place.
The USB cable is not included but you can also find them lying around or get them cheaply from any computer store.
You can hook it up to a battery as long as its over 7v.
You could technically use lower voltage supplies but it gets complicated then.
Just plug it in to the power socket and it'll work.
The plug pack only has to be over 7v and under ~30v.
...
You can hook it up to a battery as long as its over 7v.
You could technically use lower voltage supplies but it gets complicated then.
One minor correction, with the Diecimila the input voltage range is apparently this:
I thought the minimum was 7v due to the regulator.
there's a new low-dropout power regulator, so the board can run with a minimum input voltage of 6V instead of 7V, meaning that you can power it with 4 AA's
The maximum isnt 12v. It says that just to prevent people blowing stuff up.
Its really around 30v.
(don't give it more than 12V, though, unless, as Gianluca says, you want to cook eggs on it),
FWIW I've successfully run my Diecimilla from a 6V lantern battery. OTOH I have not verified giving it more than 12V blows it up.
In theory, the regulator can more than 12V (either 24 or 30, I forget which), but that assumes a much beefier heat sink than we have on the Diecimila. You might be able to get away with it for short periods of time, but it's probably not a good idea (and like Gianluca said, it will get very hot).
I just put a Fluke meter on a Diecimila powered by exactly 6.0V, and after the protection diode you get 5.3V, and the regulator output is 4.35V. To get 5V as advertised on the 5V pin, you need to put 6.7V minimum into the external power pin, which that is cutting it very close. The regulator has a dropout voltage of 1V, the diode drops about .7, so: 1.0V + .7V + 5.0V = 6.7V.
The Diecimila is still going to work ok at these lower regulator voltages, as the Atmega and the FTDI chip have minimum supply voltages of about 3V each. But... at 6V in, your regulator output will be 4.35V, which means you're not getting any supply regulation, as the regulator is just operating as a pass-through. This could lead to lots of problems as the power supply voltage varies, aside from the fact that it defeats the purpose of having a regulator in the first place.
Re the maximum input voltage, the regulator is an MC33269, which has a stated max input of 20V.With the drop across the polarity diode, that makes the max input voltage 20.7, after which your Diecimila automatically converts to toaster mode
So, I'd suggest that the input voltage range for the Diecimila specs should be changed to read: 6.7 to 20.7V, or just to round it out, 7-20V. You could add something like "above 12V we do not recommend driving loads over 250ma with the Diecimila."
he maximum input voltage, the regulator is an MC33269, which has a stated max input of 20V.With the drop across the polarity diode, that makes the max input voltage 20.7, after which your Diecimila automatically converts to toaster mode
Actually, it hits toaster mode comes from dissipating the difference between the input and output voltages, rather than just from being over-spec somewhere. If the diecimila draws about 100mA, and the input is 20V, the regulator has to dissipate (20-5)*.1, or 1.5 Watts, which is rather a lot for such a tiny package (compare to the heat produced by a 4W nightlight bulb, if you want... (and of course it gets worse linearly if you're powering "other stuff" that makes the board consume more than 100mA.) (and 100mA is only a rough guess/measurement from the local lab power supply...)
(although, looking at the datasheet more closely, it seems that allowable dissipation is CLOSE to 1.5W even for minimal heatsinking, so if you keep your current consumption down (no additional devices attached), you might be OK even with significantly higher input voltages. You can use that formula ( (Vin-5)*I ) to calculate your regulator power dissipation, and if it's under 1.5W, you are PROBABLY ok.)
so the 6V minimum input is obviously wrong... will the official spec be changed to reflect this? Many, many many a time we come up with useful things in here but nothing comes of it.
Well, except it works at 6V, right? What about adding a sentence that says something like "If supplied with less than 7V, the board's power may fluctuate. For maximum stability, apply greater than 7V."?
What about adding a sentence that says something like "If supplied with less than 7V, the board's power may fluctuate. ..."?
Quite a few people use the 5V from the Arduino to power their external hardware. For instance the MAX7221 is unforgiving when Vcc goes below 4.7 Volts. The Arduino-code on the AT is still running, but all Leds are off.
The KISS principle demands the sentence : ""If supplied with less than 7.5V, you are doomed!!!!",
(7.5V because its a common value you find at the wallwart powersupply store).
When operating on voltages less than 5V (on a 6V battery without an LDO regulator, or on a 4.5V battery) I have gotten very strange results when attempting to download a new program (e.g. the new program not loading). IIRC I have even trashed the bootloader this way.
The systems seems to run OK at a bit less than 5V, but programming at less than 5V is a bad idea, in me experience.
I have not experimented to see how this sub 5V operation affects writing to EEPROM.