Command: dmcopy "pspc.fits[bin time=1606000:1613000:20,pha=1:100]" timedet.fits
This example is similar to the previous one. We use the same X axis, time binned in 20 second bins. For our Y axis, we choose pulse height (PHA). The PHA is a (low resolution) measurement of the incoming photon energy. In this dataset, the PHA value runs from 1 to 256, but the response of the instrument is very low at large PHA values, so we only display PHA values from 1 to 100 using "pha=1:100". This is interpreted as "pha=1:100:1", a bin size of 1; we could use a larger bin size to improve statistics if we wanted.
The image shows the change in the overall spectrum from the image with time. Most of the photons are are low PHA values (at the bottom of the image), but just before the end of the first good time interval there is clearly extra emission form harder photons, probably due to background from the bright Earth. The user would probably refilter this data to exclude the time period with the bright background.