Re: [CAD_CAM_EDM_DRO] Bar stock

Ian Wright wrote:

> Hi Jon,
>
> I have replied to the list admitting that, if production nowadays is as
> Ballendo suggested, then the centreless ground stock is likely to be very
> true. However, the problem I mentioned earlier about lobing of the stock
> with the older 3-wheel method is valid and the biggest problem with it was
> that it didn't show up using 'normal' measuring methods. That is, you could
> check the diameter at as many places as you wanted to and with as accurate a
> micrometer as you could find and the answer would always be the same - but
> the stock would still have 3 lobes. The only true way to check it was to
> mount the bar between centres on the lathe and use a fixed micrometer to
> measure any eccentricity (not a dial indicator as these are subject to
> 'stiction' - stick-slip and are not particularly accurate as they are only
> intended as comparators). By the time you had done all this is was quicker
> to turn a bar true anyway.

There are several ways to detect these sorts of defects. One way is to
put the bar in a V block, and mike the top of the bar. This will definitely
detect 3-lobe shapes, but may not detect other variations from perfect
cylinders. Also, with centerless grinding, these 3-lobe patterns will not
be in line, but will likely have some twist to the pattern, making it not only
easier to detect, but actually visible to the naked eye.

These hardened and ground bars are usually consistent on the diameter
to better than .0001", have marvelous surface finish, and are free of
circularity
defects. It is a substantial job to setup the toolpost grinder, dress the
wheel,
and then turn something to that level of accuracy. It is vastly easier to pick

up a hunk of stock that is already to that level of quality, if the available
diameters will suit the task.

As for dial indicators, the good ones have jeweled bearings, and I have
one that directly reads out in .0001" marks. I can pretty much resolve
variations down to .00001" on that indicator, without any backlash or
slip/stick effects. If you want to know if a diameter is consistent or if
it is out of round, this will clearly show it, even if the exact reading is
not valid. (If the axis of the indicator is not perfectly normal to the work,
then the calibration is off.)

Jon