ss wrote:
Is there a consolidated opinion on the amount to nudge a piece of work
sitting in prussian blue in order to encourage marking? Procedurally
I have been using (in compass points) 1 inch north, then 1 inch south,
then 1 inch east, then 1 inch west. I use the rubber handle tips of a
pair of needle nose pliers to apply the force as much as possible to
either the x or y axis (e.g. with no pushing down on the piece). Is
this a good procedure, or should I try something different? (BTW -
when I do no nudging whatsoever, there appear some number of dots on
the surface, but so faint that I often need a magnifying glass in good
light to see them. The dots using the method outlined above are good
- but more smudges than dots. This fact is bothersome, I think. The
examples I have seen are clearing circular-ish dots).

It depends on the condition of the piece. But, you may have too
much dye on the surface plate. You want it VERY thin. I use my
hands to move the part, not tools. I sometimes lift up to reduce the
downforce, if the part is very heavy. Other times, especially when
the spin test shows a convex part, I will press down on one end and
lift up on the other, to show where the highest spot is. The spin
test is simply to see if the part will swivel on its center. If it
swivels, there is a pronounced hi spot in the middle of the part.
By moving the part in all 4 directions, you force both ends to
touch down as the part rocks, making you think it is roughly flat.
The press down/lift up trick will cause marking on only half the
part, if it is convex. The mark closest to the center is the
highest spot. Just scrape there, and repeat. I usually use a
circular or oval movement, and often make 3 circles of the part.
(That is an orbiting motion, not actually spinning the whole piece.)
Again, it depends on the condition, and what the marks have been
looking like.

I have stopped using the Prussian blue, it is just too messy (and
long lasting on the hands.) I use Canode dye, which is made for
marking & scraping purposes. It is not as good as the Prussian blue,
as it will ball up after being on a surface plate for a while.
(I guess that means there is latex in it.) But, it washes off,
and I find the performance of the stuff to be quite good, otherwise.
I spread it out with a rubber roller called a "Speedball Brayer"
apparently available at any art supply outfit.

One thing that may work just a little better with the Prussian blue is
the variations in the density of the mark. Modest high spots that
are just high enough to touch the dye get a dark mark. Higher
spots that really support the part by contact through the dye
to the surface plate leave a thinner mark in the center, with a dark
ring around it. Sometimes you also get a brown or gray color to it,
as some iron has rubbed off and mixed with the dye. These are the
highest spots, and more attention should be made to those marks.


Also, I have a light piece I am working on - 6 inches by 3 inches by
3/8 inch. I found that it does not mark well. Noticing the type of
surface plates and other things that are usually scraped, I think
usually the work is much heavier. Therefore I tried placing a couple
of matching pieces of the work on top after it is placed in the blue.
I seem to get a much more uniformly good marking. Is this procedure
acceptable? When the piece is not weighted, there are times that when
nudged, it will ride the surface like an air hockey puck with just the
lightest of nudges and I have to grab it to keep it from going off the
side. I think there is some sort of air pocket induced when this
light work is not weighted....

When you get it very flat, the air bearing effect can be pronounced.
Also, when it bites through the air, and sticks to the dye, you almost
need tools to pry it off!

If you are trying to get it absolutely flat, so that you get a
consitent blue color across the entire part, dream on! You will never
get there. This will achieve a flatness of about 10 uInches across
the part, which has no use unless you are trying to make your own
gauge blocks. Getting .0001" flatness is fairly easy, and it sounds like
you may be there, already.

Jon