D1 Troubleshooting

Detlef Smilgies, CHESS

Remark

The following procedures are written for SAXS and GISAXS/GIWAXS users as well as for the CHESS operators. It is recommended that the users work with the CHESS Operator to start-up the station computers or troubleshoot the detector. Read the whole subsection dealing with your problem carefully, identify which symptom fits your problem best, and then go through the steps in order. If nobody has any idea what I'm talking about, you should give me a yell.

Rock'n'Roll Trouble Shooting

Rock'n' Roll refers to the semiautomatic line-up macros for lining-up samples for grazing incidence scattering and for calibrating the incident angle.
Rock'n Roll works in over 95% of all line-ups, but occasionally there can be a problem, either due to a non-ideal sample or an operator error.

Here is a trouble shooting list

Check whether the shutter was open during the line-up scan; if not, redo line up with shutter open

you can redo "rock" as many times as needed, but do not repeat "roll"
if "roll does not work (only noise in the scans, no reflected beam), your sample may have poor x-ray reflectivity
for poor reflectivity samples run rock a couple of times, and then set "samth" and "samz" to ZERO

Check in the hutch whether something is blocking the beam
Type "wsam" and compare user values and dial values for motors samz and samth.

I carefully align the sample stage relative to the beam; these motor positions are saved as "dial" values
if values differ by more than 10%, use the "set" command to set the motors to the dial values: set samz <dial value>;set samth <dial value>
use the "mv" command to move motors to zero: mv samz 0; mv samth 0

Make sure that you use a proper sample name containing only (A-Z, a-z, 0-9, -, _); otherwise an error results

redo "new" with a proper sample name

Make sure that you supply the correct sample thickness when using the "new" command
Check that the shutter is open.
Run the "rock" and "roll" macros again.

If this does not help

Remove the sample, move samz and samth to ZERO, and carefully line up the sample holder with "rock"; then set "samz" and "samth" to ZERO as you new starting values

you may want to note the dial values shown by "wsam" in your logbook
if the sample stage does not look horizontal, you can level it with a water level, before doing "rock"
if the Imon signal is low or high throught the samz scan in "rock", the sample stage is either too high or too low, respectively

type "gclose" to protect the detector, then "shopen"
move the "samz" manually with the "mv" command
note a position when the beam intensity is starting to decrease
redo "rock"

Put the sample back on, do "new" with a new sample name, and chose the correct sample thickness (wafer: between 0.5mm and 0.7mm, glass slide: 1-1.5mm, ITO/FTO coated glass: 2mm)
Try "Rock" and "Roll" again

Does the sample look flat? rough samples are always hard to align - "rock" is you best bet, "roll" won't work most likely (see above)

The monitor intensity Imon has steadily dropped - what's wrong with the mono?

Call the CHESS operator to help.
Run the macro "monolup"

The report from "monolup" is printed out on the laser printer in the F2 control area
Operator, please file the printout

If this doesn't help, call me or other CHESS staff familiar with mono line-up

Background: Eversince we have been running in the special orbit for the undulators, the D1 beam has been less stable.

The Pilatus detector doesn't take images and/or SPEC gets stuck during taking an image

Call the CHESS operator for help.
If SPEC gets stuck, exit the waitinng loop with "Ctrl-C"
Pull out the screen and keboard of the detector computer, and wait until the detector GUI appears.
Take some (3+) images manually by hitting the "Start" button

if image acquisition does not stop automatically, hit the "Stop" button
check the exposure time, correct it to a smaller number (1s) if need be, and try again to take an image with "Start"

when the detector responds normally, run "movie_setup" with new parameters (filename, exposure time) and check that the GUI values on the detector computer are properly updated

if the detector computer remains unresponsive, close the GUI and click the "Stop" icon and wait until all detector windows are closed. Then click the "Start" icon.
try again with "movie_setup" and "shoot"
if the detector is still unresponsive, have the CHESS operator call in help (Scott Smith - detector group, Phil Sorensen - LINUX software group)

The laser printer does not print - how do I get the printer to cooperate again ?

check whether there is paper in the printer - the CHESS operator can get more paper from the stock room
powercycle the printer and wait 5 min
sometimes the cause for printer failure is a bad job

if printed from LINUX (station computer)

open a new window
check the D1 print queue > lpq
remove the top job > lprm
wait 5 min

if failed job printed from Windows (analysis computer)

check the print manager and remove stalled jobs

see the CHESS operator for help

SPEC refuses to let me open a new data file

SPEC only permits the user to have a finite number of files open. Usually thsi does not cause problems, but under certain conditions that I have not yet clearly identified SPEC "forgets" to close old data files and hence new ones cannot be opened. Here's the fix:

list all open files by typing > on()

note: the parentheses "()" need to be typed!

close all old data files that are not in use any more > close("<data file name>")

note: close only the files that you recognize as old data files!

you should be able to use "newfile" or "new" now

I accidently closed the SPEC window - how do I restart SPEC ?

Call the CHESS operator for help. Often the SPEC process is still active in the computer, and does not allow you to start SPEC again. In this case, first the old process needs to be killed, before SPEC can be restarted.

open a new window (window icon on the bottom left)
type "ps -ef | grep -i saxs" and find the process id of the process called "saxs"

of the two numbers listed, this is usually the second one

type "kill -9 [process id]"

note: read the process table carefully: if you kill the wrong process, the computer may crash

open a new window , type "saxs"

How do I recover from a crash of the D1 station computer ?

Only the CHESS operator or other qualified CHESS staff should reboot the station computer. This following shows how to reconfigure the SPEC interface:

log in and open a window (window icon on the bottom left in the command bar)
type "saxs"
all relevant macros should be loaded automatically when SPEC starts, even from a fresh start

How do I recover from a crash of the D1 analysis computer ?

reboot the Windows PC
start "captureGIXS36" using the icon on the desk top
images are on the S: drive on the Terabyte server
find your user directory on the S: drive, image data should be in the "pilatus" subdirectory

Film thickness monitor hung up

try to quit and restart the soft ware
if this does not help: try to reboot the computer (hutch, on top of the FilMetrics controller)
note: the USB cable has to be connected to start the acquisition program

Microscope computer

check that the USB cable between camera and the hutch computer is properly connected
close and restart "Debut" program

Microbeam stage: motor does not respond

identidy controller (1-3) and channel (1-8) of problem motor
switch to manual by pushing knob on the corresponding manual controller 1, 2 or 3
dial up motor channel using the V or P push buttons
hold knob of manual control down for 2 seconds - motor will be reinitialized and moved to home position
if this does not help: power-cycle the controller

NOTE: After 15 years of faithful heavy duty at D1, the MedOptics detector has been retired in favor of the Pilatus 200k.

What do I do when the CCD camera seems to be hung up?

sympton 1:

no image

go through checklist:

is the personal safety shutter open?
is the garage door open?
is the ion chamber removed?
is the sample at a reasonable position (wsam)?
is the beamstop at a reasonable position (wbs)?
is there signal on Idet in line-up mode (GISAXS users) or on the photodiode (SAXS users) ?
is there an obstacle in the beam, like a fallen down cable, shielding etc. ?

in many cases steps 1-7 fixed the problem, and the CCD turned out to be just fine.

if none of the 7 steps bring an improvement, move on to symptom 2

symptom 2:

always the same image, regardless whether there are x-rays on the sample or not

explanation: the IDL interface does not notice, when it loses connection to the CCD controller - this time the CCD is definitely hung up

troubleshooting steps:

check that the water interlock on the CCD rack is okay (yellow control light on)
check that the CCD controller is on (green LED on front)

if the controller had been off for a while, wait 10 min for the CCD to cool down before proceeding

stop the acquisition program by hitting q on the Windows PC keyboard in the CCD camera rack
restart the program by typing allez
take a test image

if the CCD is still unresponsive:

exit the IDL program
restart by double-clicking the IDL icon on the desk top
type "allez" on the command line
take a test image

whenever you feel completely lost at one of these steps

call Detlef

symptom 3:

the CCD macros return without any delay without producing an image

when an image gets terminated with Ctrl-C before the end of the exposure time (NOT recommended !!!), the CCD can end up in an undefined state

type "ccd_off", then type "ccd_on" to reset the CCD interface