What's more likely to cause an overload? High voltage or lower temps?
Does it matter? A 10% higher voltage will result in a 21% higher current, but once the coil is hot, the higher resistance will compensate this to a degree.
A 33W - 12V coil resistance is typically around 4.5 Ohm at room temperature, and this value changes about .4% per degree centigrade.
What matters here, is how the controller handles the current limit, ans as far as I know, the PWM output is overload-triggered sooner than the DIG-Out.