Switched polarity? arduino power trouble follow up

Hi Forum Experts, First THANKS to those who have been helping my son troubleshoot his Arduino. He's younger than you may have guessed, but well ahead of me in this realm, and the advice is really valuable.

As you may recall if you saw the thread from earlier today he was powering up a ArduMoto motor driver shield, attached to the Arduino, when the shield power trace fried. Apparently one (or both) of a pair of matching caps on the Arduino also got burned (one is obviously damaged). These are labeled C6 and C7 on the schematic. We have learned these were 22uf and should be 100uf , orange in color, and are ordering replacements which I can hopefully install. ( It looks like the board may be a Chinese made version or knock off -but otherwise seems to work fine.)

He was powering the Shield with a nominal 13.2v NiMh battery and thinks the polarity may have been backwards. QUESTION 1: can we confirm that bad polarity on the shield would cause the damage described?

QUESTION 2: He currently can not power the Arduino off the VIN pin (2.1mm barrel jack). It does power up fine thru the USB plug. Is this consistent with the above problem (damaged C6 and/or C7 caps) or is there something else going on?

QUESTION 3: He has also noticed that the 2 smds located between the barrel jack and the USB jack are getting quite hot when the board is powered over the USB. Is that normal, or perhaps also related to the one or more damaged caps?

We are hoping that we can replace the caps on the Arduino, jumper the burned trace on the ArduMoto board, and get him back to building his Roomba-based robot!

If you can offer any guidance (pls assume we are ignorant but learning fast) we would really be grateful.

Cheers, Andy & Tim

Question 1) yes, you pretty much have to look at the board, or use a meter to test resistance, if the + is wired to a ground pin on the arduino its backwards

Question 2) yes, USB voltage is not regulated by the on board voltage regulator cause its already a stable 5 volts, when the USB is plugged in its bypassing that part of the board

Question 3) could be, I dont have an offical design board so I dont know off hand, but if capacitors fail, it is possible they can fail from wide open and no connection to welded together shorted out, and anywhere between

can we confirm that bad polarity on the shield would cause the damage described?

You know, that's a tough one. I'm looking at those two caps and trying to figure out just how both could have been damaged. The best I can figure is the -13.2V on the motor shield board found its way to the 5V pin through the motor driver and pulled +5V on the Mega down to -13.2V or thereabouts. That would severely reverse-polarity the caps and quite possibly damage them (and destroy the voltage regulators too).

However...if this did happen, a lot more should be wrong with the board. As in, it should be totally dead (ATmega should have been destroyed). That it still works with USB plugged in is good for you but makes me feel like I don't really have a good explanation for the failure mode.

He currently can not power the Arduino off the VIN pin (2.1mm barrel jack). It does power up fine thru the USB plug. Is this consistent with the above problem (damaged C6 and/or C7 caps) or is there something else going on?

This is consistent with C6 and C7 being damaged AND the two voltage regulators being destroyed. Those caps can actually be removed and the circuit should work OK (they're just there to smooth things out and the circuit will often work just fine without them).

He has also noticed that the 2 smds located between the barrel jack and the USB jack are getting quite hot when the board is powered over the USB. Is that normal, or perhaps also related to the one or more damaged caps?

That's not normal. With only USB plugged in those 2 SMD's (the voltage regulators) should not be warm at all as they're not doing anything. I think you will have to replace those too along with the caps. Or just remove them and the caps and only use USB from now on :)

And final self-serving advice: consider buying a Gator :) We designed those things specifically for kids and students who don't realize that a simple moment of carelessness in the realm of electronics can set you back a long time in having fun.

-- Check out our new shield: http://www.ruggedcircuits.com/html/gadget_shield.html

Excellent analysis, hardly anything to add :-)

  • USB from Notebooks does generally not deliver 5V, but something like 4.75V and differs from model to model. This is fine but might sometimes confuse you.
  • On most boards there is a diode between the + of the barrel connector and the pin called "Vin" (and the capacitors..)
  • 13.2 V NiCd is marginal. Fully charged NiCd batteries have a voltage 10% above the nominal voltage which can be 15V in this case. This is too much for a capacitor rated 16V. The issue is not the capacity (22uF) but the voltage rating (should be 25V as i said yesterday)!