I am having problems where I drill through to the interior of the
band to create the beginning of an azure, but find that my drill is
going in off-angle where is looks good from the top, but poor from
the bottom (all over the place).
Well, there are number of solutions to such problems. I will be
talking about it in details on my blog,
but that is coming later. Let's see what can be done here.
Perfectly perpendicular holes can only be done with drill press
(modern solution) and drillstock (old way). Drillstock requires a
lot of practice and it is almost impossible to find flat drills in
our sizes. They can be made, which I do if there are no other
options. Drill press requires coordinate table and dividing head, to
drill eternity type rings. I personally abhor this solution, but use
it nevertheless, when working with steel rings.
I suspect what you are asking are methods when some deviation is
allowed, especially if azure ( ajure ) is to be employed. If you
making hollow type rings, check out my DVD "Eternity Ring"
If you making solid type, with some practice you can get good
results too. The key is to learn to drill always using the same
position of the left and right hands and the same position on bench
pin. I again refer to the aforementioned DVD. If you do that, with
some practice, you will acquire muscle memory and will be able to
drill within 1 or 2 degrees tolerance. Azure will make deviations
unnoticeable. It is useful to have an idea of thickness where hand
drilling is acceptable. Assuming 2 degrees deviation and 5mm
thickness, distance from center to drill exit is computed as ring
thickness / tan(90-2) = 5 / 28.64 = 0.17mm. Even if you work with
half-pointers, the deviation is acceptable. Smaller thickness means
more dimensional precision and large thickness means increase in one.
So if staying within 2 degrees is too much to ask, reducing thickness
is another approach to think off.
There are few other methods of simulating drill press if you do not
have one, and others. The explanation requires drawings, which are
not possible here. So I will expand on the subject more fully in my
blog in some not so distant future.