Home | Principle | Specifications | Measurements | Contact |

Light source Sample Sample Holder Tilting Sample Housing with camera connected with computer |
||

Schematic drawing of SAMS Side View |
Measurement of slice done by SAMS |

The SAMS is used for automatic measurement of stress in glass. It measures the light intensity of polarised light going through the glass, which is dependent of the stress. From this intensity, the phase-shift is calculated. SAMS has automatic compensation for changes in light intensity of the light source or transmission of the sample. Also for some common used samples (slices, strain slabs) the SAMS analyses the measurement and gives the result.

For larger sample types, the viewing area can be enlarged, as is shown in the given set-up for In-line SAMS:
Set-up for In-line measuring of stress and temperature, see also the Measurements.

For Parrallel polarisers the intensity due to the stress is Ip = 1/2 Io ( 1 + cos( f ) )

Elimination of Io which is determined by the intensity of the light source and the transmission of the sample gives Icn = ( Ip - Ic ) / ( Ip + Ic ) = cos( f )

From this, the phase-shift can be calculated f = arccos( Icn )

SAMS uses the intensities from the images with crossed polarisation and parallel polarisation to compute the phase-shift.

To the top of the page

The accuracy is determined by the number of bits of the ADC (Analog to Digital Converter) which converts the analog camera signal to digital values for the computer, and by the level of the phase-shift because the conversion f = arccos( Icn ) is not linear.

For very low stresses, the accuracy is improved to

When the stress is given in nm / cm thickness, the accuracy of the stress value changes with thickness: for a thickness of 0.5 cm, the accuracy of the stress value is 6 nm/cm (plus additional inaccuracy because of the thickness).

The measured value of retardation can be converted to MPa with the SOC (Strain Optical Coefficient). With SOC = 25, the accuracy is 0.1 MPa.cm. For stress values, this has to be devided by the thickness in cm.

Normally the range of f = arccos( Icn ) is 0 to +p, in SAMS this is doubled. For stress values the thickness is very important: for high stresses, we can take a small sample thickness and by this we can measure higher stresses. For a thickness of 0.5 cm, the range is -600 nm/cm to +600 nm/cm or - 24 MPa to 24 MPa.

Giving the thickness in mm automatically converts the measured values in nm to nm/cm.

With a SOC which can be set, you can choose nm/cm or MPa for stress values.

For well defined samples the measured values ( graph over the width ) is worked up, so

for strain slabs: correction for small distortions like cords, and the values at the seal are extrapolated.

for slices cut out of glass: correction for small distortions like cords, and the values to the edge are extrapolated.

The maximum values and the graph can be written to file.

To the top of the page

Example 1 of measured stress: Given is the stress profile given in MPa along the thickness of a slice given in mm. This slice is cut out from the front of a TV screen. The higher compression stress is at the outside of the screen, the positive tension stress is the bulk stress in the middle of the screen. |

Example 2 of measured stress: Given is the stress profile given in nm along the length of a strain slab given in mm. A strain slab consists of 2 pieces of glass sticked together. One piece is the reference glass, and the other piece is the glass from which the expansion coefficient has to be determined. The stress build up at the interface is an accurate indication for the difference in expansion between the two glasses |

Measured stress values in MPa.cm during 12 hours. (for stress in MPa these values have to be divided by the thickness in cm). The black line is the averaged (over 12) value. Spread is also because the products have a variance in temperature. Example 3, Figure 3.2:

Measured temperatures of the products from figure 3.1 in degree Celsius. Example 3, figure 3.3:

Corrected stress values of the products in MPa.cm during 12 hours. The black line is the averaged (over 12) value. The values from figure 3.1 are corrected for the temperatures as measured in figure 3.2. The correction is an extrapolation to the value that the product should have at a temperature of 20 degree Celsius.

To the top of the page