Hubble space telescope is famous for its spectacular space images and flawed mirror. This simulation lets you explore why a 0.01% error in mirror curvature required a space walk to fix. Technical details of the flaw can be found in the Hubble Space Telescope Optical Systems Failure Report[1] and offer insight to prevent similar problems.
Below is an optical model of the Hubble telescope. The right side shows how a star would appear using the original wide field camera[2], and if you zoom in, you should observe a blur around the spot of light in the center. That is the infamous blur of the Hubble telescope. If you change the drop-down box to the Designed System, you can get a sense for how blurry the as-built system was.
In this system, star light begins far to the left side of the image, and some of it encounters the large M1 mirror. This is the mirror we are most interested in. After reflecting from the mirror, light travels to the left where it encounters the smaller M2 mirror. One reflection later, the light travels to the detector through a hole in the center of the M1. The light on that detector is shown in the image on the right. Dragging your cursor inside the detector plot shows how the star image changes depending where it is in the sky. Pressing the Ctrl key exaggerates the off-axis effect, and pressing the Shift key eliminates camera pixelization. Note: zoomed in images are centered on the star while the non-zoomed case shows the actual star location.
To appreciate the precision of telescope optics, pay attention to the numbers as you swap between the designed and as-built systems using the drop-drown box. The important number is the M1 'ROC' value which stands for Radius Of Curvature and describes how steeply curved the mirror surface is. Almost the entire Hubble error occurs because the M1 mirror focused light 1.7mm long of its 11,040 mm design. In more practical terms, the edge of the surface was 8 μm (0.0003 inch) lower than it should be!
To better understand how radius of curvature and conic constant affect the mirror shape, try adjusting the sliders below. The blue circle shows the radius of curvature for a cross-section of the mirror.
- ROC: acronym for Radius Of Curvature of the mirror at its center. Larger values means less steeply curved.
- conic: number indicating shape of the mirror. Conic = 0 is a sphere. Conic = -1 is a paraboloid. Other conic values indicate ellipsoid (conic > -1) and hyperboloid (conic < -1) shapes.
A mirror can only be made as good as it can be measured. The people manufacturing the M1 mirror at Perkin-Elmer corporation knew this, and took great effort to correctly measure the radius of curvature. An important step in that process was a precision metering rod fabricated from special material insensitive to thermal expansion. This was crucial for spacing test equipment. Incorrect spacing would result in an incorrect radius of curvature.
Furthermore, the rod needed to be well centered, so end caps were fabricated with a hole such that the end of the rod could be seen while maintaining proper lateral position. The most likely cause of the Hubble error is related to the end cap, and best illustrated by figure 7-4 in the Hubble Space Telescope Optical Systems Failure Report[1]. Instead of measuring to the surface of the rod, they inadvertently measured the surface of the end cap!
As with all accidents, there were a series of events which could have prevented this incident. The report mentions independent measurements which showed a problem, but whose significance was not recognized. Management pressure also contributed to the incorrect manufacturing. As with most accidents, a series of unfortunate events all lined up to cause this catastrophe.
Fortunately, this story has a happy ending since the system was optically corrected and improved with future missions and space walks. In the 30+ years since the mirror flaw was recognized, the Hubble space telescope continues to advance our understanding of the cosmos.
On a personal level, this piece of history offers important lessons for my work fabricating the next generation of large telescope mirrors. Having a trusted, independent measurement is a crucial part of every plan. No matter what your job is, it is also important for everyone to keep an eye open for overlooked problems.
References
- National Aeronautics and Space Administration (1990). "The Hubble Space Telescope Optical Systems Failure Report" (Publication NASA-TM-103443) https://ntrs.nasa.gov/citations/19910003124
- National Aeronautics and Space Administration (1990). "Hubble Space Telescope: Wide field and planetary camera instrument handbook. Version 2.1" (Publication NASA-CR-189753) https://ntrs.nasa.gov/citations/19920008266