Table of Contents
Height gauge measurements allow us to mechanically determine the height of an object with respect to a flat reference surface. Please contact Matt McCulloch to get trained for using the height gauge. This page serves as a reference.
Dial gauge: Mitutoyo, No. 513-442-16 Range: 0-0.06 inches • Uncertainty: 0.0005 inches (0.5 mil) Remove the dial gauge from the height gauge arm when not in use. Pack it up carefully in its box and store it in the ATLAS cabinet in B33 large clean room. | Height gauge: Mitutoyo, Code No. 192-630 Range: 0-12 inches or 0-300 mm • Uncertainty: 0.005" This device uses a CR2032 battery. Remove the battery when not in use and store it with the dial gauge. |
https://www.mitutoyo.com/wp-content/uploads/2017/01/Dial-Test-Indicators-web.pdf | https://www.msi-viking.com/Mitutoyo-192-630-10-Height-Gage-0-12300mm-Range_p_794.html |
A good orientation for the dial gauge, showing how to mitigate the Cosine Effect |
Mitytoyo provides guidance on using height gauges and other measuring devices here.
For height gauges, see pages 22 and following. Our stand is a "digimatic height gauge" according to their nomenclature with a test indicator attachment. Page 24 warns about the consequences of using the height gauge without a flat, clean surface. Keep cleaning!
Page 32 describes the impact of the Cosine Effect on dial test indicators. The page includes comics illustrating why it's important to minimize the angle between the dial test indicator and its direction of use.
Working with the height gauge has several critical ground rules:
Zero the height gauge to the granite table surface.
Measure the height of the screen protector surface.
Zero the dial gauge; read the value on the digital gauge.
Note that the screen protector edge is rounded, so collect the height measurement at least 0.3 mm away from the edge.
Zero the height gauge again, so that the screen protector surface is the baseline for the measurement.
Repeat the measurement of the screen protector surface: It should be zero this time.
Measure points A-O (A-J) on the triplet (quad), following the markings in the images under the "Measurement positions" heading.
Check the zero, taken in steps 2-3, and record its value.
This value reflects whether the zero-point drifted during the measurement. Normal values are +0.0005, 0.0000 (often written down as "5 zeroes"), and -0.0005. The units are inches.
If step 6 reports an unusual value (> +0.0005 or < -0.0005), please repeat the measurement.
The labels and their order do not inscribe any special meaning. They do match how the height gauge measurement processing scripts handle the measurements, so please follow the order. It's important to collect height gauge measurements on top of the glue layer. The coverslips are thin so measuring at their edges without the glue underneath causes them to flex, producing inaccurate measurements.
Triplets | Quads |
---|---|
Notice that we use the center of the module and the centers of each star for quads. The label order also changes! |
Hannah and Rachel prepared some Python scripts to interpret CSV files of the height gauge measurements. The scripts are available at: https://gitlab.cern.ch/hherde/slacpixelsanalysis. The scripts produce "maps" of the samples, displaying the extracted glue layer height mapped to its approximate measurement position on the physical sample. The code marks each measurement point with an up or down arrow, indicating whether that particular point is above or below average compared to the rest of that module. The code also colors points outside the ATLAS thickness specification red.
The "sample map" plots are always uploaded to the gPhotos album associated with that sample collection run. You can access a complete list of datasets here.