8 micro inch reference specimen required - I have a Mitutoyo SV-622 profilometer

I have a Mitutoyo SV-622 profilometer that we currently calibrate with a standard 3 patch precision reference specimen. The lowest patch range is 16 micro inches. I’ve just been asked to acquire an 8 micro inch specimen for calibration purposes due to a blueprint callout on some manufactured parts. It is my understanding that it is not wise to use a specimen of such small nominal value due to the higher percentage of uncertainty it would generate.

The question I’m being asked by the end user is how would I know that the profilometer is giving accurate readings below 16 micro inches if it is not calibrated below that range.

Any help would be greatly appreciated.

-SAM

SAM135 said:

I have a Mitutoyo SV-622 profilometer that we currently calibrate with a standard 3 patch precision reference specimen. The lowest patch range is 16 micro inches. I’ve just been asked to acquire an 8 micro inch specimen for calibration purposes due to a blueprint callout on some manufactured parts. It is my understanding that it is not wise to use a specimen of such small nominal value due to the higher percentage of uncertainty it would generate.

The question I’m being asked by the end user is how would I know that the profilometer is giving accurate readings below 16 micro inches if it is not calibrated below that range.

Any help would be greatly appreciated.

-SAM

Click to expand...

Sam,
I also use a Mitutoyo profilometer, though not the same model as yours. The user manual includes a calibration procedure. In the procedure, only one reference standard is required for calibration. That would mean if there is no discrepancy between the reading and the known value of the roughness specimen, then the device will be accurate over the entire range of measurement.

Though this truly isn't adequate, it does say something for the confidence the manufacturer has in its equipment. What you will need is at least two roughness specimens with different roughness values (which you have) that have been calibrated by an accredited laboratory. Take a measurement on one of the specimens, adjust the gain if necessary. Then check the second specimen. If the profilometer returns the correct values on both specimens, then it is accurate over its operating range. You will be safe measuring well below an 8 micro.

On a related note: I once could not get our profilometer to accurately read both specimens, turned out the stylus on the detector had worn out and needed to be replaced.

SAM135,

Measuring surface roughness measurements of very smooth surfaces is difficult. In order to get accurate results for surface finishes below 10 micro inches or 0.25 microns the stylus tip has to be in nearly brand new condition. If the stylus is slightly worn the profilometer measurements will underestimate the roughness of very smooth surfaces.

Many years ago I requested a deviation approval on behalf of a prototype supplier for the surface roughness of centerless ground pinion shafts. They measured 8 micro inches where 6 was required in the specification. The product engineer refused the SREA but allowed the supplier to polish to achieve the specified finish. I asked if current production shafts were polished and he said no. He stated that the production supplier was able to achieve the specified finish via grind processing only. He claimed he had data that confirmed that function deteriorated when initial finish exceeded 8.

As it happened I reviewed prototypes ordered from the production supplier not more than week later and armed with my brand new little Taylor Hobson portable Surftronic analyzer I discovered that regular production measuring 6 micro inches on the production profilometers actually measured 10 on mine. Needless to say my ah-ha caused a ruckus. We went to their lab to find calibration test blocks to settle the difference. While there I noticed a very high-end proficorder, they claimed was seldom used and that no one present had ever operated it. I assured them that I was familiar with it, we set it up, measured, and corroborated the 10 micro inch findings. All three devices including the production instrument with the worn stylus checked OK for value and linearity on the calibration test blocks.

The test blocks that come with the equipment that are used to calibrate gain and linearity have a lower value that is insensitive to slight stylus wear (I think approximately 25 micro inches) and an upper value sufficiently greater to set linearity without exceeding typically measured finishes (I think approximately 100 micro inches). There is standard that you can buy to check the condition of the stylus (I think approximately 10 micro inches) but to distinguish a 5 from a 7 the stylus has to be nearly new.

Paul

Thanks guys for the useful information.

I’m one of the guys that have no actual certified metrology training and must learn from books or asking questions as I go along. My employer won’t send me out for any training. Most of the calibration is cut and dry as I’m following written calibration instructions but some of it is vague and or leaves too many open ended questions.

I’m glad I found this site. I’ll be back for sure…