Testing and Quality Control

ADC thoroughly tests all relevant aspects of the slit systems to ensure the highest quality for our customers. Using state-of-the-art equipment, ADC verifies the accuracy, repeatability, and parallelism of each set of blades. Our technicians then generate reports to be presented to the customer with the completed system.

For ultra high vacuum compatible equipment, ADC performs a bake-out and RGA for each system. The output of the RGA is documented and presented with the final product.

 

Blade Polishing

ADC has worked with the Cornell High Energy Synchrotron Source (CHESS) for over two years developing a polishing process that produces the best slit blade knife-edges in the synchrotron community. We have collaborated in the design and test of slit blades in tungsten and tantalum with lessons learned that could be applied to other materials.

The polishing process is labor intensive and requires highly skilled technicians with years of experience to create an edge useful for x-ray beam definition. Poorly made blades result in secondary radiation sources and an overall muddied beam profile.

ADC produces blades of many different sizes and shapes for its range of x-ray slits. Widths vary from 10 to 150 mm and thicknesses from 1.5 to 12.5 mm. Blades used in ADC's slit assemblies are available in either Tungsten or Tantalum - each presents a unique problem when polishing.

The process begins with a grinding operation designed to minimize the amount of material that must be removed during polishing. An edge and one of the faces are ground to an interior angle of 88° and then a milling operation relieves to facilitate a more rapid polishing process.

The image below shows a typical slit blade with the face ground and polished and the relief cut into the surface.

Fixturing the slit is critical when polishing. The fixture must be extremely hard so that material is removed only from the blades, otherwise the knife-edge can become rounded. Two or more blades are usually polished together in a matched set to maintain parallelism. Both surfaces must be accessible without removing the blade because repositioning is impossible within the necessary tolerances.

Grinding produces an edge that is straight and true, but with pits and scratches that must be removed by polishing. Scratches are removed using a Buehler low-speed polisher with silicon carbide paper and polycrystalline diamond suspensions on fabrics of differing knaps. As finer grits are used the blades and fixturing must be thoroughly cleaned in an ultrasonic bath to remove larger particles. A final polish with colloidal silica is used when surface finish is critical.

The images below show two typical blades after the grinding process, but before final polishing.

Wheel speed, applied force and polishing time vary with each step and are critical to the final quality. Too large a force leads to grain pull-out with Tungsten, which is extremely soft (Vickers hardness of 873). Relatively large force and long polishing time are required for Tantalum (Vickers hardness of 3430) yet over-polishing results in “orange-peel” that destroys the knife-edge. As a result of over two years of continuous improvement to our polishing techniques ADC can produce blades with an RMS surface roughness of better than Ra < 0.013 µm. The feedback that we have received from many of our clients has been critical while developing our processes.

ADC can achieve an excellent surface finish when polishing Tungsten or Tantalum blades for slits as shown in the figures that follow. This does not, however, guarantee that the slit will be satisfactory. The two faces that form the knife-edge may be completely free of scratches or nicks but will still cause diffraction if there is any radius.

These images show the polished surface finish. Color scale in upper image is in nanometers. The trace shown above was taken along the white line in the upper image.

The next two figures show the x-ray diffraction patterns from two sets of blades assembled in slits with nearly identical degrees of polish. The set of blades shown in the left figure have three excellent knife-edges while the fourth blade is acceptable. The second set of blades, shown in the right figure, all have rounded-off edges. Great care must be taken to ensure that the final polishing steps do not round the edges. At ADC we use specially designed fixturing to support the edges throughout the polishing process. We have found that the final polishing step should be done only on the beveled surface of the blade. This allows us to support the knife-edge and prevent rounding. As a result, the surface roughness on the face normal to the beam direction is not polished to better than 1 micron.

An optical comparator is used to position the blades and inspect the edge quality at a magnification of 100X. For the Exit Slits, laser diffraction is used to check the alignment of the blades and measure the positioning accuracy and repeatability. The motors are driven to specified positions and the diffraction pattern from a laser beam, pasting through the slit, is used to calculate the actual opening. Parallelism is tested by making diffraction measurements at the center and near the ends of the slit. Surface finish is measured using a white-light interferometer.

Blades from large precision slit viewed on an optical comparator at 100X.

 

Blades from exit slits at 6.5X magnification (left) and 25X (right).

 

Blade Parallelism

To measure and verify the parallelism between blades for slits other than the Exit Slits, ADC uses a Keyence model LS-7030MT optical micrometer with sub-micrometer resolution to check the gap at numerous points between blades. This gives an RMS parallelism, which is then reported with the unit. Preliminary alignment is performed with an optical comparator. The image below shows the blades from a sub-micron accuracy water-cooled slit enlarged 10x.

 

Blade Accuracy & Repeatability

The accuracy and repeatability of each blade over the length of its travel is measured and documented with the same optical micrometer. It has an accuracy of +/-1 um over its measuring length that is as good as or better than most encoders.

To make use of the sub-micrometer resolution, repeatability measurements are performed over short spans, but at numerous points along the travel of each blade. For example, a blade with a 25 mm stroke is typically checked for repeatability at 5 points.

To verify the accuracy of both encoded and open loop manipulation systems, the data taken by the optical micrometer is compared both to the number of encoder counts recorded and the number of microsteps through which the motor has past. This information is then documented and presented to the customer with the final product.