Hmmm. I don't use a romer arm and in fact I think any portable-arm CMM is not a good inspection tool (for an inspector).
They are great tools for machinists, checking the work in-process with the stock still clamped up, but lousy to try to get efficient production inspection done.
That said, I don't see why a 2mm ball would not work on a portable-arm unit.
I would use a 0.5" ball, and try to use the sigle point circle.
On a cylindrical part Ø5, the diference of height between the highest point and the lowest point of a hole Ø0.25 is 0,04 mm (0,0015").
I believe that the uncertainty has approximately the same value on a Romer, I think you can do it by this way.
John Riggins : I agree with your point of view about using arms.
I use sometimes a T-probe with a tracker, and the 2mm ball is too fragile (the stem is too small) when the probing pressure is too high.
Yeah, that's my main issue with "arms".
They are very operator-sensitive.
When the operator gets a bit tired (after holding and manipulating an arm for hours) the accuracy suffers.