Vin,
This effect is what you see when the goniometer's rotation axis is not
exactly perpendicular to the beam. HKL2000 is compensating for rotations
that are actually due to goniometer misalignment by changing the crystal
orientation angles. It does not cause any harm to your data, unless
things are so far off that refinement gets lost, which does not seem to
have been the case for your data.
Marian Szebenyi
MacCHESS
Vin Purp wrote:
> Hi, sorry for the non-ccp4 related question.
>
> I have one data set that yields peculiar results when I use HKL2000 (or
> 3000) to integrate it.
>
> Background: I have crystals that index to the I222 sg. I have shot and
> collected datasets, integrated, molreped and refined many hundreds of these
> same crystals and know them very well. However, I recently collected data
> that has the following 'problem' when I integrate: I collected 450 frames at
> 0.3 width (135 deg) .The Rot. Change vs. Frame window in the HKL program
> (Integration tab) shows that in the first 150 frames, the x and z deviates
> +30 and frames 150-450 x and z cross to a negative change and finish at
> -120. The Y-axis doesn't deviate from 0 the entire time.
>
> All other statistics look 'normal' for the integration; mosaicity is a bit
> high (0.75-1), the Chi sq. hovers near 1 for all frames, and the 'Cell' and
> 'Distance' vs. frame do not deviate from 0 much at all over all frames. I
> have played with integration box size, spot size, elongation and background
> with the same results.
>
> Other stats: data set is ~1.60 Ang. resolution; unit cell = 128x140x168
> Ang.; Roverall = 4.2%
>
> I was curious as to what this phenomena is due to. Any ideas are appreciated.
>
|