Low mode dispersion with positive and negative orders

 Did a comparison of low resolution in order = +1 and order = -1.  Looking at wn0 = 399 cm-1 with H2O lines from the gas cell.  Used slit 1 to limit light and dekker ~33 degrees for short slit.  

Note that dispersion direction is reversed from what we're used to in negative orders.

I rotated the order=-1 image by 180 degrees so it was the same orientation as the +1 image.  I then rectified by hand (slightly different angle required: 5.88 degrees for +1 order and 5.85 degrees for -1).  I then extracted 11 rows near the center of each spectrum and summed.  I subtracted off the median and took the absolute value so now got emission spectra.

Looking carefully, you can see that the negative order dispersion is less than the positive

I fit gaussians (with baseline) to all 7 lines in both spectra.  I then called the centroid of the first line on the left as pixel 0.  I found the difference in separation between the leftmost line and the rightmost line and took the ratio.  The order = -1 dispersion was 98% of the order = +1 dispersion.  

After correcting for the dispersion to first order, the lines lie pretty much on top of each other.