On 10 Oct 2005 20:18:47 -0700, wrote:

They are balanced but the cast iron bases on a concrete floor just
don't provide enough coefficient of friction to make them stay put.

I need an interface material that grips concrete and cast iron.

The machinery is pretty good quality (ie mostly Baldor) but some of the
pedestals are Harbor Freight crap. (i don't mind buying Baldor buffers
and grinders but their pedestals seem too high priced for what you
get).

I filled some of the pedestals with concrete to give them more mass. It
helps but there's just not enough friction between cast iron and
concrete when motors are spinning at 3600 rpm.

I think you need a bigger footprint. I welded up a stand for my 2 HP
3450-RPM buffer with an H base that is 24" x 28". It doesn't dance at
all and it's easy to move. Include diagonal braces from the
extremities of the H footprint to somewhere on the column so it'll
be an acceptably stiff structure.

Underlying theory: the system must be balanced enough so it can't hop
clear off the ground. Beyond that, a dynamically unbalanced system
creates a rotating moment on the pedestal that can cause the base to
dance with rotating single-point contact to the floor. With a big
enough footprint and acceptable flexure in the pedestal and/or
compliance in the points of contact, there will always be at least
two contact points with the floor. When this condition exists, it
can't dance no matter how good the tunes on the radio. Larger
footprint means less motion at the work wheel for necessary
compliance between points of contact and dynamically-unbalanced
rotating mass if at least two points of contact are to be maintained.

A totally rigid structure supporting a dynamically unbalanced
rotating mass on a totally-noncompliant floor would dance with
single point contact and infinitesimal clearance elsewhere,
regardless of the mass of the base. Real world, no structure is
totally rigid and even a concrete floor has some compliance. My
point here is to show that pouring concrete (or lead or depleted
uranium) into the base isn't the only solution and perhaps not the
best solution if mobility is an issue. The trick is to get the
moments and compliances to where at least two points of contact are
always maintained by gravity (rather than bolts) with acceptable
motion at the work wheel. The compliance can come from rubber feet,
carpeting, flexure in the structure, whatever works. Larger
footprint means less motion at the wheel for given compliance.

Bottom line: bigger footprint works for me.