Hello all. I'm wondering if it's possible to burn out something related to Vin on the diecimila? The reason I ask, is that of the two deci's I have, one of them gets SUPER HOT (the ATMega chip, nothing else) instantaneously when attached to my 12v/3a DC power supply via Vin/GND, but works fine when attached to USB. No programming works, but it doesn't get super hot when I attach a 9v battery to Vin/GND.
The other works A-OK when attached to the power supply in this fashion. In fact, I have been using this one for a few months hooked up this way. It just started recently, and the only thing I can think of would be that I might've, when wiring everything up for final packaging, accidentally reversed Vin/GND and had them swapped for a short period of time.
Is there something in that channel that I can replace, or is it shot? (It will have to run on external power, not USB for this project.)
Connecting power backwards is definitely something that can destroy a mega168. If something is getting hot, you have a short between power and ground. Try using a multimeter to look for shorts between VIN and GND. Since things work normally when powered via USB, it sounds like you destroyed the voltage regulator. If the regulator is next to the AVR, maybe it's the regulator that's getting hot and the AVR is just getting hot by due to its proximity? Plus it would make sense that the regulator would be the first component to go if you connect power backwards.
In short, I suggest you use a multimeter to evaluate the situation and then try replacing the regulator if the evidence indicates that's the problem. If the board uses any electrolytic capacitors to stablize the power lines you might need to replace those as well.
If the regulator is next to the AVR, maybe it's the regulator that's getting hot and the AVR is just getting hot by due to its proximity? Plus it would make sense that the regulator would be the first component to go if you connect power backwards.
I think the regulator is at the opposite end of the board from the MCU. But if it's damaged in such a way that it allows too high a voltage out, that would definately cause the MCU to get very hot very fast (unfortunatly, I know first hand thanks to a cheap "regulated" power supply and a couple of pic chips I killed very fast).
If the regulator is next to the AVR, maybe it's the regulator that's getting hot and the AVR is just getting hot by due to its proximity? Plus it would make sense that the regulator would be the first component to go if you connect power backwards.
I think the regulator is at the opposite end of the board from the MCU. But if it's damaged in such a way that it allows too high a voltage out, that would definately cause the MCU to get very hot very fast (unfortunatly, I know first hand thanks to a cheap "regulated" power supply and a couple of pic chips I killed very fast).
Yes, I think y'all are right that the regulator is blown. It was when the power didn't come on after wiring it up, I moved my hand over the board and the regulator was quite hot. I then dis-assembled everything and it worked on USB supply. Now, when I put it all back together, and try running it off of Vin, the mega chip (not the regulator) gets instantly hot. In fact, the regulator is pretty cool when this is happening. I haven't had a chance to take the multimeter to it yet. However, I understand that the regulator could be blown - but would that actually cause Vin and GND to short out? Do power regulators fail-deadly? =)
Yes. I had a 7805 on an arduino-derived project go, and the failure mode was passing high voltage to the regulated side. Smoked US$400 worth of sensors.
Strangely, the ATmega8 took over 8V on Vcc for an extended period of time and still seems to work. I replaced it, of course, but it's still around somewhere.
Yes. I had a 7805 on an arduino-derived project go, and the failure mode was passing high voltage to the regulated side. Smoked US$400 worth of sensors.
Strangely, the ATmega8 took over 8V on Vcc for an extended period of time and still seems to work. I replaced it, of course, but it's still around somewhere.
-j
Ah, sweet. Well, like I said, it still works, but only over USB. So I guess its time for me to break out the very fine soldering tip and replace that regulator. Thankfully those chips seem to be tough.
I don't think a reverse polarity on the power connector should burn out the reg. Diode D1 is directly in line with the input and would be reverse biased so no current will flow.
I would be interested to know what voltage you have on the Vcc line when you connect it up like that, it should be 5volt, but if the reg is shorted it will show the input voltage, for safety leave the header for power off and measure across C7.
I don't think a reverse polarity on the power connector should burn out the reg. Diode D1 is directly in line with the input and would be reverse biased so no current will flow.
I would be interested to know what voltage you have on the Vcc line when you connect it up like that, it should be 5volt, but if the reg is shorted it will show the input voltage, for safety leave the header for power off and measure across C7.
It's too late for that, as I've already replaced the regulator, and now it works fine. As I understand it, the diode is only on the DC jack, and not the Vin pins. I accidentally reversed the polarity on the Vin/GND pins, and not the power connector.
You were really lucky the rest of the ICs weren't blown. It looks like the regulator went short circuit and saved all the other chips from being fried. Do you do the lottery? ;D