In the Arduino Micro schematic there is a power selector that seems to cut off the ability for current to flow from the 5V to VUSB if VIN is present, which seems fine. However, if external regulated 5V is applied directly through the 5V pin, VIN is at 0V. Will this not mean that VUSB is powered from the 5V even when no USB cable is attached? Since VUSB is connected to the VBUS sense pin of the atmega32u4, this will trick the USB controller inside the chip into thinking that a USB cable is attached. I don't know if this has any consequences for the Arduino framework? Will it start wasting CPU cycles trying to communicate with a PC that is not there?
Worse yet, if I actually connect a USB cable to a weak USB port for example on a laptop, providing a slightly lower voltage than my externally supplied 5V, will this not mean that power is fed from my 5V regulator to my laptop USB port?
I want to supply the Micro with external 5V because I need more current, but at the same time I want to be able to plug in the USB to a computer for programming/debugging without risk of frying the computer. I can't find anything stating that the Arduino Micro is actually designed to be supplied with external 5V, but I can't find any docs or posts saying it shouldn't be done either.
Better to be safe and supply my Arduino Micro through VIN and let the external 5V power the other stuff?
In my opinion and that of many many others there is nothing to worry about.
The Arduino only tries to communicate with the USB connector when you tell it to.
However one member will worry about this and quote some Arduino beginners advice not to do it.
We will see if he comes along later.
Whether there is something to worry about must depend on the amount of voltage difference and the kind of circuit in the computer USB port. There is a 500 mA resettable fuse on the Micro that should prevent disaster. However I wouldn't be too happy if a few hundred mA flow continuously from my 5V regulator to the computer USB port producing unnecessary heat in my 5V regulator.
These fuses are, electrically speaking very slow to blow. In fact the speed of blowing is dependent on the voltage difference between the fuse's rating and the excess current. This could take many seconds in marginal cases.
And how is that going to happen? You can only get current flow when there is a voltage difference, so you are only talking here of a fraction of a volt difference.
The only things resisting current flow are the resistance of the traces/cables (could be very low), the resistance of the transistor in the power selector and the resistance of the polyfuse. The last two are unknown to me. If we have 0.3V of voltage difference (4.8V in the laptop USB port when battery operated is what I have measured, 5.1V maximum from my external regulator according to the datasheet) I need at least 3 ohms to get below 100 mA.
A FET has a variable resistance between drain and source called Rds. When the FET is on this resistance is called Rds(on) in most datasheets. The MOSFET in the power selector is in an on-state in this case.
You are getting confused about what will happen here.
If anything gets past the switch then you will be driving the internal regulator backwards. That is applying a voltage to the output of the regulator. Anybody who has ever done a non trivial amount of electronics has wired a linear regulator up back to front as it is only a three pin package. So when you wire say a 12V to 5V regulator with 12V on the input and measure the output then you know that you get nothing on the output.
First, I am not talking about driving the regulator backwards. What I am referring to is transistor T1 in the Micro schematic. We have 5V on the +5V-terminal and 0V on the VIN-terminal which means T1 will have a low resistance from +5V to VUSB. I have verified this with measurements. When applying 5V to the 5V-pin the same 5V show up on the VUSB net and in the unconnected USB-cable. Thus, since the path through T1 is open only the Rds of T1, the resistance of the polyfuse and the resistance of the traces/wires will limit the current. In fact, I just measured this current to 150 mA on my breadboard when plugging in the USB cable to the laptop, so there is everything to worry about. And this is including the internal resistance of the multimeter, which helps to limit the current. If I bypass the multimeter, I see the analogue current scale on the lab bench supply go up even more.
Second, it is not true that a linear regulator will not pass current backwards. The internal 5V regulator of the Arduino Micro will pass current backwards because it has an internal diode from Vout to Vin, as all regulators in the NCP1117 family do. T2 in the Micro schematics will prevent this and make sure VIN stay at 0V even when applying external 5V. So we can disregard this entirely.
@brydling you came here to ask a question. Now I know that you are not open to receiving an answer welcome to my ignore list. I would rather spend my time helping people who want to listen to my advice.
Goodby.
@Grumpy_Mike Yes, I came here to ask a question because I thought I could avoid having to analyze the details myself. I thought such a fundamental question as "Can this device be powered by external 5V?" must have been documented in some official documentation that I had just missed. I noticed your answers had errors in them so I ended up analyzing it myself anyway. And as far as I can see the results proved you wrong. If that is a reason to be added to your ignore list then I guess that's just who you are. Given your chosen nickname I think you are already aware of that.
You are welcome to point out any errors in my analysis if you have found any. I did not come here to fight. You are mad because I did not listen to your advice. If I had listened to your advice I would have ended up in trouble. Your advice was wrong this time.
