Sorry folks, I understand this question has been asked quite few times, but I wasn't able to find a definitive answer: my Mega 2560 R3 is powered by 7.7V from Vin, and runs at a steady 135mA (in theory it's 2.7 x 0.135 = 0.38W dissipation). But the regulator with a heat sink still gets hot. Not burning hot but definitely to a point where you'd question whether it's a good idea to leave it running for months.

Is this normal? Also is this a sustainable plan for a long running project?

The 7.7V (7.7 because I remember the regulator needs a voltage around that to work) is coming from a variable step down module that could deliver a 5V. However I've noticed although it does stabilize at 5V , when the power just comes on it would surge to ~9V for a fraction of a second. My understanding is the 5V input on arduino is assuming clean 5V, so I hesitated there.

Any suggestions? Thanks!

  • That is indeed odd. How you you measuring the power consumption? Is there anything else connected to the board? If you hold the processor in reset, does the regulator remain cool? How did you measure the 9v spike? It is entirely possible that it is real, but many meters have an "overshoot" in their response to step input changes, so it could also merely be an artifact of how your were measuring. A storage scope would be useful, but obviously even today costs a few hundred dollars. – Chris Stratton Jun 12 '15 at 18:28
  • 5V connects Vin, Gnd connects to multimeter, meter connects to negative. Current fluctuates a bit around 135mA but never goes over 140. For the voltage, the buck converter has a display that tells the output voltage, and it'd show ~9V and then quickly goes down to 5V. You could be right that it's overshooting but again it's a cheap ebay item.. – zee Jun 12 '15 at 18:37
  • @ChrisStratton The board is driving a GPS module and 10 basic logic chips of various kind, but the current meter should measure all these combined, and I think 140mA is still in the acceptable range of a mega, no? – zee Jun 12 '15 at 18:46
  • Using a multimeter to measure current directly is a bad idea. Get a very low value shunt resistor and measure the voltage across it instead, – Ignacio Vazquez-Abrams Jun 12 '15 at 19:36
  • If you have any communications cables connected which have their own grounds, trying to measure current in the power ground path will be inaccurate anyway, since those grounds may be lower impedance than the one going through the meter's burden voltage. You really need to unplug everything that unplugs in the way of peripherals and cables and test just the board by itself, both with the chip in reset and running. Then add things back one by one. – Chris Stratton Jun 12 '15 at 21:34

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Browse other questions tagged or ask your own question.