Precision angle measurement is a critical component in the construction of many modern electro-optical assemblies. Laser-based interferometers and autocollimators are among the most common optical tools for non-contact angle measurement. All of these systems direct an optical signal at a reflective surface and derive angle information either from interference effects of the reflected signal with a reference signal or by sensing the position of the returned signal (see figure A).
Visual autocollimators project an optical image (commonly a crosshair) onto a target mirror at infinity focus. The instrument detects a change in the angular position of the mirror as displacement of the reflected signal when the signal is re-imaged on a reticle with a graduated scale via a beamsplitter through the shared objective lens. Visual autocollimators can measure angle changes as small as 0.5 arcsec under ideal conditions but are generally used only to resolutions of a few arcseconds. Common applications in measurement and alignment include prism angle, wedge angle, laser cavities, and components having a reflective surface or having a mirror attached.
Autocollimators require that the target mirror be flat to λ/4 or better to ensure that the return image is sufficiently well focused that the operator can accurately determine its position against the graduated scale in the eyepiece. (It is possible in some cases to gain valuable angle information from poorly focused images.)
Visual autocollimators require the operator to judge the position of the return image in order to make an angle reading. The operator can evaluate the relative position of multiple images produced by different surfaces within an optical system, as well as the multiple images produced by a single component such as a prism. Even autocollimators with micrometer-driven reticles used with averaged multiple readings require a skilled operator to achieve accurate, repeatable results. The addition of a CCD camera, software and frame-grabber board can automate some measuring tasks.
Electronic autocollimators remove operator dependence, decrease measurement time, and increase repeatability. Measurement resolution and accuracy are improved by approximately two orders of magnitude over visual autocollimators. Advances in electronic autocollimator design during the last 30 years have led to instruments with very broad measuring ranges. The addition of CCD arrays for detection, digital electronics for signal processing, and the direct export of data to spreadsheets and other specialized software greatly speeds data collection. This is especially useful for such arduous tasks as recording and calculating all of the information required to qualify rotary tables, machine ways, and polygons.
For all of their high accuracy and speed, electronic autocollimators have some peculiarities that require careful consideration. An out-of-focus image is impossible for an electronic autocollimator to resolve into good angular data. The same is true for multiple reflections; however, in many cases, users can temporarily obscure unwanted surfaces to allow measurements of individual surfaces, or they can set the software to reject unwanted images. Consider the alignment of three fixtures in a system. By blocking three of the four mirrors to measure one mirror at a time, three fixtures in a system are aligned to a reference mirror within tenths of arcseconds (see figure B). An electronic autocollimator is especially suited to this task because interrupting the signal has no effect on the value stored for the reference mirror.