Rural road traffic study using infrequently sampled data.

I'd like to perform a traffic study to determine how many cars per day travel down my road. I don't have access to one of those car counting machines that you sometimes see laying across the road. However, I do have a mechanical counter and can count the cars that pass. I can't spend the entire day counting cars, but can count the number of cars that pass in some period of time. And I can do this over a period of days.

Will I be able to an accurate idea of how many cars pass per day? How many samples would I need?