AXAF Project Science:
The approximate diameter of the x-ray-emitting spot in the Electron Impact "Point Source" (EIPS) is < 0.5 mm, depending primarily upon the fractional bias voltage on the EIPS cathode plate. For the XRCF source distance of about 520 m, this corresponds to an approximate angular size < 0.2 arcsec. Owing to the excellent angular resolution of AXAF's High-Resolution Mirror Assembly (HRMA), the effects of the finite size of the EIPS spot are observable in focal-plane images, on this scale.
To characterize the EIPS spot size, we used a simple x-ray pinhole camera, comprising a pinhole and an imaging detector --- the High-Resolution Imager (HRI) microchannel plate. The pinhole is a 40-micrometer-diameter hole in a 40-micrometer-thick gold foil, mounted in a thick support with a 3.08-mm inner diameter. Note that, for images obtained at the higher energies, the foil becomes transparent and the mount's hole is visible.
To produce the data shown here, we converted the raw HRI data into FITS event lists using the Fortran program HRIFIT, which also produces de-gapped coordinates. We then used LEXTRCT to read the event lists and produce thumbnail GIF files and FITS images. LEXTRCT compressed each full 4096x4096 HRI images to a more manageable 512x512 image; thus, each FITS-image pixel corresponds to 8x8 HRI pixels. We have applied neither flat-field corrections nor background subtraction. The CRDELTx keywords in the FITS file correspond to the physical distance, in microns, at the spot.
The de-gapping algorithm is not perfect; thus, some de-gapped images exhibit weak anti-gaps. Furthermore, one of the HRI's preamps --- number 16 of 17 --- no longer works. Hence, no real events were detected with pixel coordinates between 3584 and 4096, and the positioning algorithm erroneously places some photons between channels 4096 and 4224. However, this does not affect the spot images, which were located away from this defective area.
We performed the copper (Cu) sequence twice. The first run was a test which led to changes in the setup, to fix light leaks and to add internal baffles to reduce scattering and focussing by the tube's walls. Do not use the first Cu data set: We provide these only to demonstrate consistency, in that 2 different runs, 6 days apart, give essentially the same images.
The EIPS MP1 data were processed using versions 1998-02-26 of HRIFIT and 1998-02-26 of LEXTRCT; we used sumb.gap for de-gapping corrections. The FITS files record these version numbers. Accept nothing less.
Cu-K data first run (do not use)
Cu-K data second run (use this instead)
The data listed below are for the MP1 cathode tests made in June 1998. A comparison of the these spots with the previous runs indicate that there are some differences. It is believed that the change occurred in the period 1997-Jan-17 to 1997-Jan-20. At that time, a leak was discovered in the MP1 cathode and the MP2 cathode was inserted for a few days. In searching for the leak the MP1 cathode was partially disassembled. Since the spot is very sensitive to precise alignment, it is likely the change occured when the cathode was reassembled. Thus the data listed below are appropriate to all data taken after 1997-Jan-20 (late in phase D and extending to the end of the test). For tests prior to that date, the data listed above should be better.