[Author Prev][Author Next][Thread Prev][Thread Next][Author Index][Thread Index]
Re: [ATM] Commercially available spherometer with MitutoyoDigimatic Indicator
Dave,
> 1&2: once you enter the "tenths" range, you start heading for
> new sources of error. One of these is heat, such as the heat of your
> hand on the spherometer, which may cause the instrument to measurably
> warp. If you are seeing differences in readings with a thousandths
> dial gauge, and the dial gauge is in good repair, you will see differences
> with a tenths gauge. Ten times the differences, so to speak.
I've already taken that into account, and let the mirrors equilibrate
between grinding sessions, and then focus on getting good readings.
> In the olden days, we did not so much zero as simply write down what
> the zero reading was, and subtract or add it to subsequent readings.
Not an option with the instrument I'm using. The main issue here is the
bezel moves easily. You zero by aligning a line on the bezel with the
indicator hand manually, with the instrument on a reference flat. Problem
is that process alone adds a degree of imprecision. Also, while using the
spherometer, I have to avoid bumping the bezel. I zero it using the same
flat surface so that I'm at least consistent. But, given that it may not be
truly flat (right now I'm using the backside of one of the mirrors) then at
least I'm consistently offset in my lack of precision. In other words, I
may be reading .015, but in may reality be .0146 (GIGO).
> If you find your zero changing markedly during a series of measurements,
> something is wrong; it means your non-zero measurements are probably
> also changing.
I think it's more an issue with the indicator. Hence the desire for a more
accurate instrument. I can pick it up, and set it down in the same
location, and consistently get varied readings +/- .002. The needle just
doesn't always stop in the same place, even when measuring the same
location.
> You aren't going to measure half a tenth without controlled conditions,
> chiefly temperature, but also cleanliness (dust). As I remarked
> above, at the tenths level in normal machine shop practice, one
> waits for the workpiece and mike to achieve thermal equilibrium and
> to be at the calibration temperature before measuring.
I thoroughly rinse the mirror and let it dry and equilibrate prior to
obtaining the readings.
> Exercise: "A micrometer reads 0.5000 inch and is verified against
> a perfect standard. The micrometer is made of carbon steel. If
> the temperature rises 10 deg Fahrenheit, what is the new distance
> between the mike anvils? Part (b) What is the temperature of
> your lab, and what is the temperature of your hand?"
(a) Don't know. Obviously, it would be less than .5000 inch, as the metal
would ever so slightly expand.
(b) Delta would be about 30 deg F. Of course, it is not a linear model, so
3 times the temp does not equate to 3 times the expansion.
Terin
_______________________________________________
ATM mailing list http://www.atmlist.net/