Chandra Aspect Operations

Plate Scale Pre/Post CCD Cooldown

Summary

The evaluation of changes in the plate scale that we performed (Plate 2006) did not answer the following question: How did the plate scale change as a direct result of CCD cooling from -10 to -15 degrees C? We set about answering that using the star pair data from one year before the cooldown (May 2002 to May 2003 inclusive) and one year after the cooldown (Aug 2003 to Aug 2004 inclusive). We use the method of the plate scale calibration to solve for new plate scale coefficients and compare the changes.

Figure 1

Postcool( Housing@289.22K) - Precool ( Housing@287.9K )

Unfortunately, the IRU swap occurred just as the cooldown occurred, so we don't have enough data to separate out changes in the CCD due to cooling and changes in the plate scale due to the changes inthe housing temperature. So, we'll take a look at this in a few different ways, but we can't come to any firm conclusions without more cooldown data.

Silicon has a linear coefficient of thermal expansion between 1 (at -100C) and 2.6 (at 0C) * 10^-6 per degree C. A CCD in a mixed package might have a CTE of up to 5 * 10^-6 per degree C.

If we use 3*10^-6 to create some estimates, we expect a change at the edges of the CCD (~2500 arcsec) to be

Δarcsec = 2500 arcsec * 5 deg * 3*10^-6 / deg
Δarcsec = 0.0375 arcsec

As we see a change at the edges of closer to 0.3 arcsec, the change due to CCD cooling is either much larger than it should be or just a small part of the change.

Estimated Change Due to CCD Contraction

Here we've compared the primary difference plot with one created to show what we would expect if the first order precool coefficients were set to values solely based on CCD contraction.

Table 1

Postcool( Housing@289.2K) - Precool( Housing@287.9K )	Precool ( Housing@287.9K, 1st order terms manually set to Si contraction values ) - Precool ( Housing@287.9K )
The plot at right is a plot of the difference between the two plots above. The plot can also be thought of as the extent to which the CCD contraction model explains the plate scale change. If the CCD contraction completely explained the change, this plot would be completely green.

Estimated Change Due to Housing Temperature Increase

Here we've compared the primary difference plot with one created to show what we would expect if the precool coefficients were used at the housing temperature used during the postcool year.

Table 2

Postcool( Housing@289.2K) - Precool( Housing@287.9K )	Precool (Housing@289.22K) - Precool (Housing@287.90K)
The plot at right is a plot of the difference between the two plots above. The plot can also be thought of as the extent to which the ACA housing temperature increase explains the plate scale change.

Combined Estimates of CCD Contraction and Housing Temperature Increase

Here we've combined estimates of the effects of both changes, CCD contraction and ACA housing temperature change.

Table 3

Postcool( Housing@289.2K) - Precool( Housing@287.9K )	Precool ( Housing@289.22K, 1st order terms manually set to Si contraction values ) - Precool ( Housing@287.9K )
The plot at right is a plot of the difference between the two plots above. The plot can also be thought of as the extent to which the CCD cooling and ACA housing temperature increase might explain the plate scale change before and after CCD cooling.

Conclusions

Analysis data

Last modified:02/27/06

The Chandra X-Ray Center (CXC) is operated for NASA by the Smithsonian Astrophysical Observatory. 60 Garden Street, Cambridge, MA 02138 USA.    Email: aspect_help@head.cfa.harvard.edu Smithsonian Institution, Copyright © 1998-2004. All rights reserved.