 |
KappaCCD User Manual
Revision 1.10, Last Update 1-July-1998
|
Using the KappaCCD server and display programs
Starting the KappaCCD programs
Once the KappaCCD system is powered, the PC will bring
up the Windows environment.
From the program manager open the KappaCCD program group
and double click on the KappaCCD icon. The other applications listed in
the KappaCCD program group will be started along with the main program.
The KappaCCD system will start the initialization procedure,
in which it is slowly moving the axes and camera position towards their
reference positions. The KappaCCD server window displays "GON INIT" during
this action. You will have to wait for this action to complete.
Grabbing the mastership
The KappaCCD system can be controlled remotely over the
network or locally through the KappaPC. After startup the mastership is
set to the host computer. When you do not have the mastership, most functions
will be unavailable to you. This is indicated by the greyed-out appearance
of most entries of the menu. You obtain mastership by using <Acquire>
<Grab Mastership> on the KappaPC.
Mounting the crystal
Mounting the crystal is best done in one of the standard
positions of the KappaCCD.
After using <Acquisition> <View position>
the camera will move to its reference position at the outside of the DX
sledge.
Pull out the beam stop to create some more working space
while mounting the crystal.
If you want to change the collimator, this is the time
to do it.
Centering the crystal
Correct centering should be done from one of the view
positions in order to have the view direction perpendicular to the phi
rotation axis.
Placing a sheet of white paper after the crystal is a
very effective way to get a distinct view on the crystal.
-
Use the view button to go to this position.
When the crystal is far away from the center, the crystal
may not be visible initially.
-
Adjust the height of the crystal and make sure that either
the crystal or its support will be visible. (i.e. put it slightly too high)
-
Press <Rotate 90> until you can see the crystal in the
display area.
-
Use the adjustments perpendicular to the view direction
to center the crystal. When the crystal is visible but very much out of
focus you can adjust the direction parallel to the microscope to
get a sharper image. This is a nifty trick when otherwise the crystal moves
completely out of the display area.
-
Repeat the adjustments of crystal X, Y and Z directions as
necessary.
-
Note that adjustments of crystal height will in general influence
the centering in X and Y, so always finish with X and Y adjustments.
You can use the left and right mouse buttons to rotate the
phi axis in the wanted direction when the mouse pointer is located in the
background area of the centering window.
Visual inspection of X-ray images
At this point you must decide on the type of measurement
you will be doing.
Use the <Acquisition> menu of the KappaCCD server program
and make a 1-degree scan of 15 seconds at DX=25 mm
Spotting a long axes.
You will recognize in the image spots at regular distances.
These usually belong to one of the cell axes.
In general you will be able to see one or two of the
cell axes while the third is more oriented perpendicular to the reflection
plane. Therefor you repeat the scan at several positions of the rotation
axis, e.g. start at 90 degrees further. In case of doubt make also a scan
at intermediate positions.
When you have seen the images, you must be able to tell
if you are dealing with a maximum axis length of shorter or longer than
25 angstrom
Mm distance per angstrom rule.
To avoid spots to overlap, the minimum camera to crystal
distance in mm that can be used is 0.7 to 1 times the maximum axis length
in angstroms, depending on the spot size. The spot size it self depends
on crystal size and mosaicity and on collimator size. In practice a ratio
of 1 is considered to be safe. A smaller value yields a more efficient
data collection, provided spots are still present at the out side of the
sensitive area.
The scan angle per frame.
In an ideal (noiseless) experiment the scan angle is just
as large as possible, but without overlapping with reflections from adjacent
layers in the scan direction. For a 10 A cell axis, 2 to 2.5 degrees is
still acceptable, whereas for a 25 A cell axis 0.5 to 0.6 is preferred.
A large mosaicity results also in earlier overlapping of reflections in
the scan direction.
Like in any real experiment the best measurement is done
with an optimized signal to noise ratio. The scan angle gives you the opportunity
to optimize this ratio for your type of experiment.
Noise is anything that contributes to the background of
the signal you want to measure.
1. Noise originating from the mounting fiber and glue.
Visible as a more exposed region located around the beamstop
area.
Reduces at larger DX. Can be prevented partly by careful
crystal mounting.
2. Reflections from satellite crystal fragments (twins),
an ordered form of noise.
A smaller scan angle reduces the chance that a foreign
spot will intrude the integration area.
Denzo checks the integrity of the background by analyzing
its shape and variation and can therfor mark reflections affected by satellite
spots.
3. Readout noise
The CCD chip of the KappaCCD camera is having very
low noise levels in general. The readout noise is therefor a relative important
contributor to the overall noise level originating from the camera.
Since readout noise is a fixed contribution to every frame, you should
minimize the amount of frames necesary to integrate each reflection. This
can be accomplished by keeping the scan angle in the same range as the
profile width in the scan direction (mosaicity) or larger.
4. Zingers
Zingers fall randomly in the image, usually one every
5 to 10 seconds. Double images are a very effective way to deal with zingers.
The zingers are eliminated by statistically comparing the images.
Experiments with short exposures (up to 10 seconds) can
still be done with single images.
Zingers do not always fall on top of reflections. If
they do, the redundancy of the experiment lets you filter out affected
reflections later in the averaging procedure.
5. Dark current
Dark current is a noise that builds up gradually in the
camera until it is read out. The effect will be more pronounced for longer
exposures. For images longer than 2 minutes it may be better to break up
the exposure times in shorter parts.
The rule of thumb
Considering the effect of the various sources of noise on
the measurement, some experimental summary is given here.
-
High quality crystals with low backgrounds are measured with
a 2 degree scan angle, provided the reflections do not overlap.
-
Fair quality crystals are generaly measured with a 1 degree
scan angle.
-
Crystals with "noisy" background, possibly caused by twins
or small crystal fragments are measured with a scan angle of 0.5 or 0.6
degress.
-
crystals with a long axis or with a high mosaicity are also
measured with 0.5 or 0.6 scan angle.
Resolution range
Minimum and maximum resolution at any distance can be looked
up in a graph to get a rough indication of this range.
Starting the measurement
Though data can be collected on the controller PC, it is
much more convenient to use remote data collection from the host. In this
case images will be stored on the host directly. If for any reason the
data can not be stored on the host, data collection will continue and store
the data locally on the controller PC.
Before starting the measurement, you open a new directory
in your data home directory, and set your default to it.
You can start the data collection program by typing nkcd
at the shell prompt.
The system responds with the CCD> prompt. As this
program will be built in to the hardware control menu later, the dialogue
is kept very simple.
You can type help to list the available commands.
A usual sequence of commands is
zero
pos dx=25
filename comp_001.kcd
scan 30 180 rep=1 phi=1
The measurement is continuing from here automatically.
Instead of the commands listed above, you may use data collection scripts