Since the motor is trying to take more current than the shield can provide, there is indeed going to be a problem. Most likely, the motor controlling chip(s) will break due to overheating trying (unsuccessfully) to provide the current demand of the motor. Your best bet is to find another motor controller than is rated for more than the 2.8A so that the controller chip is not running at its maximum output.
kylekozielski:
This still would cause problems even though the motor would be using 2 channels instead of just the one 2A channel?
Interesting idea. I believe this would allow for 4A to be provided (which is more than we need so that's good). However, the voltage at the positive terminal would be doubled, assuming you tell each channel to go to the same voltage. (Motor+ = (C1+) + (C2+))
The ancient, extremely inefficient L298 in that shield is overrated and can't provide anywhere near 2A/channel continuously, Furthermore it drops up to 4V internally, overheats and shuts down. It won't work at all with low impedance steppers.
You are much better off with a modern, current limiting stepper motor driver, such as those produced by Pololu.
A 2.8A stepper is a low impedance stepper (0.9 ohms in fact), so cannot be driven from any DC motor controller,
and definitely requires current drive. Note any voltage specification on stepper motors is usually completely
bogus and should be ignored - the current, resistance and inductance are the key figures.
Steppers of 5 ohms or less are current drive only, 30 ohms and above may be voltage driven (but wont
be capable of high speeds). Anything inbetween is fairly rare and frankly a bit weird!
2.8A is more than most single chip steper drivers can handle though - the TB6560 and similar (which
massive heatsink) is one option, or else a discrete design using separate MOSFETs - both of these are
a lot more expensive than a DRV8825 which can handle 1.5A or so.
If you can change to a motor with a lower current rating you’ll reduce the cost of the driver…