Re: [avr-chat] Debuging spurious resets in Atmega16 application.

[Brian Dean]

On Mon, Sep 05, 2005 at 08:22:03PM -0700, David Shoemaker wrote:

> Does anyone have a good debug method for figuring out why my
> ATMega16 is resetting while my application is running (a robot).  I
> can get 20-30 seconds of run when it is driving but then I get a
> reset (I have radio communications over serial so I can see the
> startup banner go by again). I can then send another run command but
> it resets again shortly (repeat this cycle at will).

In addition to good suggestions by others, check that the power to
your MCU is seperated from that of your motors.  During times of heavy
current draw by the motors, the battery can sag which causes the MCU
to reset.

Also, check your grounds - feed each device from a common ground
point, i.e., don't daisy chain grounds.  Heavy current users
downstream or upstream can affect nearby devices by altering the
nearby ground potential and cause ground loops which are a bitch to
debug.  A star pattern from the battery terminal minimizes this
effect.  Along these lines, don't use the robot frame for ground.
Metal frames typically make poor conductors, and while it's tempting
to ground off at convenient points around the frame, spend the extra
few minutes and run a seperate ground return.  Be sure and use wire
that is large enough for the load.

-Brian
-- 
Brian Dean
ATmega128 based MAVRIC controllers
http://www.bdmicro.com/