First, I would like to state that I do not have near the experience
rebuilding player stacks that others have. If Paul thinks that
a few thousandths of an inch makes a difference in stack performance
(over time) this is based on knowledge and experience well beyond mine.
When I said that I did not think that a few thousandths made
a difference, what I should have said is that I don't think that
the time it takes for a valve to travel over the distance of a few
thousandths of an inch is of much consequence. The difference in flow
through the area exposed by valves with different gaps is relevant.
This is part of my reason for wanting to measure flow through the
When I rebuilt stacks for myself and a few others, my interest was
in getting the valve performance as even as possible. In choosing
a number, I deferred to the experts. In fact, I think I used the
numbers that Paul recommended. I set a number of valves to the
recommended clearance and then measured the flow to get a value to
apply to the rest of the valves.
I used the flow measurement method only because I had noticed that
accuracy of measurement with a dial gauge was not as repeatable as
I hoped. It seemed that things like crooked valve lifter buttons
wires, guides, deformed valve plates, etc., might be causing the valve
to settle differently from valve to valve and from measurement to
measurement. For example, if the top face of the valve body is not
parallel to the bottom, the valve will tilt in its travel and can
disturb the measurement. The amount of measurement error is likely
to be small but it is just one more uncertainly to be added to the rest
and any method that eliminated it was of interest to me.
My method simply measured the flow through the top seat with the valve
at rest. It might be good to apply vacuum to the stack to insure equal
compression of the leather but I did not do that. It was not my
intention to measure leakage.
Flow through a small valve gap is influenced by turbulence and other
factors so the flow through a crooked valve is going to be different
from that through one that is parallel to its seat. Since a valve
in its open or closed position will be resting on one of the seats,
any difference in valve gap caused by non parallel faces of the valve
body will be the same in the open or closed position.
For this reason I thought it reasonable to measure the flow through
the top valve plate since it is much easier to do. Since the top and
bottom valve plates might be distorted differently, it would be best
to measure the flow through both the top and bottom and to make any
necessary corrections to the valve geometry. This is much more effort
than I was willing to invest.
Measuring flow through the top seat is easy and produces repeatable
measurements. This is probably more true of the microscope method than
the dial gauge but I never thought of using my microscope in that way.
In summary, either measuring valve travel with a dial gauge or
a microscope is probably just as good as flow measurement if done
carefully and if the geometry of all valves is consistent. To the
extent that these dimensions vary, flow measurement might be better.
There is probably a reason that Ampico used the flow method for
adjusting unit valves. Whether it was to achieve greater accuracy,
as stated in their promotional literature, or just because it was
easier to build a machine to automatically set the valves, is not
clear. As far as my work is concerned, I prefer methods that reduce
the chance of errors in tedious repeated tasks.