April 4, 2006 New orbit testing for X25-suite

When the beam came back after studies, I realigned the beamline. The merit for 300 um collimator (2 dots) at 1.0A= 422.
I moved the ladder slit back to it original value (46.16) and the merit got better. I move the pitch on the mirror; merit =560.
I optimize the v-slit.
I optimized the bending on the mirror, tested the yaw on the table. I need to test the pitch better.
I started a data collection to examin the stability on the beam

stability/1/html

This is really stable!
Lets test a xtal.




April 3, 2006 New orbit testing for X25

Since March 8 there has been a new orbit for X25 without any announcement. We are officialy testing it today. This might impact people in between X25 and X29 in particular.

The beam is injected with the "old" orbit for an hour and the new one will follow for an hour. I started a dummy data collection 15 sec exposure time in order to follow the io. (counter on the tip of the collimator).
The sweep reports can be accessed at
old/1/html
old-suite/1/html
new-25-orbit/1/html

Picture 1 (old)
1-So the beam is injected and the beamline is warming up while collecting

2-alignment of the beamline around frame #30.
It seems like the monochromotor finally reach its temperature after 20 minutes. That's what we would expect.




Picture 2( called new but it's actually the continuation of old)
I realign after frrame #2. It doesn't make a difference in the counts.










Picture 3 newest (the new X25 orbit)
I left the data collection go for about 8 frames before realigning. The count are about 15% down from the previous orbit. The realign did not improve the counts.
X26A is also seeing a 20% decrease in their intensity. Their beam has moved by 120 microns. They get their intensity back by moving their slits.

Realign around frame # 67.

I tested a little moving the ladder slit and it seems like I am getting back the intensity. I need to wait for ops in order the adjust the slit.





May 24, 2005 spring shutdown


The vme crate running the motors has been updated

The computers are now running Fedora 3.0.  There is a problem while trying to FTP. This should be taken care of in the near future.

The computers have a new naming scheme consistent with the other beamlines.
x26c-c is where you start CBASS (old folgers)
x26c-d is the octane which interfaces with the diffractometer. The users are not use this machines.
x26c-b is the new linux machines. It has the HKL2000 license installed.



Dec 15, 2004 winter shutdown

Berilium window
A new berilium window is in place in the hutch. This window is narrower but longuer than the previous one. This will give us more freedom to move the mirror when optimizing the beam. We will not have to move the pipe when we change the experimental setup for 3-beam experiment. We will need to make sure the beam goes thru!
Resulting from this upgrade,  the IG controller didn't want to come back up. It's being sent back for repair. The vacuum in the last segment of the beamline is stable and in the 10^-9.-done

X26C-w
The SGI Octane x26c-w is being replace with a new Linux machine with dual head monitor.  The Octane did not perform anymore compared to a simple Linux machine. It is also less burden on our computer staff as far as maintenance and upgrades to slowly get rid of all the SGI machines. The users also really appreciate having a dual head monitors.
goniometer head
We are in the process of implementing auto-centering on this beamline. We need to install and optimize the parameters for a motorized goniometer head-- The goniometer head is too long and can't be accomodated on this beamline.

Dryer line
Replacement of the dryer line unit for the cryostream with exhaust line of liquid nitrogen from the Light Source--

Detector's controller
Unit back form ADSC. We need to swap controller to make sure it's working properly-- This controller is on X12b now.

Nonius phi spindle
A service call is in place for Andy Gangloff to come and service the phi spindle. We observed signs that it is not as reliable as the omega motor.
camera
We are having a new mount position for the camera at the working station. This will enable a view of the monitor, of the control panel, and all the computers including the detector's computer and controllers. This is critical for the staff when troublshooting from off site.-- work in progress

HV panel
The in hutch panel for all the HV cables of the IG and ionization chamber is being modified to be staff friendly! -- work in progress

Control rack
The control rack where all the VME crate and Origon cards are, in the back of the experimental station, has new electrical outlets on each side. This was implemented after the NSLS Tier I electrical rewiew found we were using power strips powered via electrical extensions. This as been clean up i.e, no more electrical extensions. -done

August 5, 2004

    New CBASS
       A new version a CBASS is now running on X26C and X12C. This will be implemented on the other beamlines in a short time.

In order to start CBASS you still have to ssh or telent to the x26c-h as pxuser.
    telnet pxuser@x26c-h
    password: {same as the pxuser on all the machines}
    cd /img06/data1/x26c/your_directory


When you open CBASS you will be in the COLLECT widow. There are three more windows that you can select to go to via the tabs: ROBOT, SETUP and BEAMLINE.

Please notice the "Group" and "PxID" buttons right beside the command line. You will need to select from the pop up menu your group and a Px number in order for CBASS to start data collection. (click here for picture) cbass_map

Ask us how to do it if you don't know what I am talking about! Have a look at the link posted on our main page http://www.px.nsls.bnl.gov/pxdb


Robot window is not in use at the moment.

Beamline window

It is no longer required to open the "Optix" window. All the normal info like COUNTS and MERIT are now displayed under the BEAMLINE and COLLECT windows.  You can use "Realign" which will perform the tzscan and tyscan macros. Do not be alarm when you realign that the old tzscan is now called a xscan!
On X26C you should not have to use the "Tweak Table" button.


Beamline window

Changing the wavelength or performing a scan for your MAD/SAD data collection is also done under the BEAMLINE tab.

If you need to Calibrate the wavelength please ask us how to do it.

You will have the same plots and CHOOCH as before.
Setup window

There is a new option that will come handy when a robot is installed at a beamline:  SETUP. For now only the first row is useful.

The PxID is automatically taking from the COLLECT window. You have to input the XtalID which should be a unique name for the xtal you are mounting. Select from the pull down menu; SCREEN, CUSTOM.

You can change any of the parameters to your desire.
A nice new feature is the "Repeat at 90 deg"

Collect window

Two lines were created since the" Repeat at 90 deg" button was checked.
The sub-directory for the crystal is automatically created.
A sub-directory for each run is automatically created.
The frames have different run numbers.

**********************************
Notice the "Pause on Beam Dump" button. If checked, CBASS will monitor the "enable beam" signal from the light source. It will pause the data collection if the beam dumps. After the beam comes back, it will wait 15 minutes for the optic on the beamline to warm up. It will do a Realign, back up one frame and recollect it.
**********************************
 

1 June 2004

Beam Alignment

28 may 2004

Maintenance
The ring has been in conditioning for several days. We could not enable the beamline since there was an interlock trip on the mirror proteus. Joe started to work on it and Shai and Annie finished it. The flow was not sufficient for the setting on the proteus. We changed the setting to BC (inter flow).

The beamline was still yellow tagged. There wasn't any yellow tag on Valve 2. This valve is not interlocked. You have to feel on the top of it with your hand to know if it's open or closed.  We enable the beamline so that the mono can be cooked. The vacuum was in the low 10-7.
There was beam all the way in the hutch but after "realign" the merit was only 10% of what it should be. We wait for the real user beam on Tuesday

24 march 2004

Ladder slit Optimization

I want to optimize the vertical position of the beam on the mono. We want to have the ladder slit position aligned with the peak of  intensity of the incoming beam is. Of course this is an iterative process since the height of the mono has to follow were the best ladder slit position is.

I am starting with the 2 dots collimator (300 µm) at lambda 1.1A.

I moved the slit by a small increment (0.2 mm) and waited a minute or two. Then I collected 7-10 frames in order to see how reproducible the counts were.
The exposure time is only 10 sec. 

The data I collected last night were for the ladder slit at 44.945 mm.
On the graph that correspond to light blue series. This position is right at the edge of the plateau (44.745 mm to 43.945 mm). This plateau is fairly larger and a slight change in the orbit should not influence too much the counts observed. I will leave the ladder slit at 44.345 mm. This is a lower position than used previously.





Stability of the beam
Each data point is for 90 sec frames with the pause on dump ( CBASS detects when the beam has dumped, pause the data collection, wait 15 minutes after the beam is enable again, realign, backs up one frames and recollect) and automax (realign macro activated) every 30 minutes. We can locate where the automax occurred since we have a point with a counts of zero where a dark was taken.

First observation the automax slightly improves the counts during a normal data collection within the first hour after a fill.

Second observation     the counts are slowly dropping from the beginning of the injection to the dump of the beam. We do not have more intensity at the end of a fill than at the beginning.

Third observation the next fill gives about the same count as for the previous fill.


Forth Observation   On the C2 counts (blue), we can see a slight fluctuation that seems to be periodic and doesn't correspond to the automax. This is not really seen in the C4 counts (pink). That was also present in the previous graph but at a lesser extent. We believe this is due to a slight change in the orbit of the beam. The smaller collimator can pick it out better than the 300 µm collimator. This should not affect the scaling of data.




Next project; Monochromator's height.




The bourgandy line is what we are getting now. We can see that the values for the small lambdas (0.9-1.1) are in the range of what we expect. The Merit for the longer lambdas are not good at all.



To be done:    The mono needs to be lowered.


23 March 2004

John Skinner has implemented CBASS with a supplementary harvesting feature. When CBASS is started a file called i0.dat is created in the main directory. Every time a frame is collected the counts at C2, C3 and C4, the ring current and the MERIT are harvested. These counts are for 1/10 sec as opposed to the usual 1 sec in the Optix window.
The HTML logger also will have the counts in.

Optimization of the slits

We are dealing with two issues: intensity and stability. The Vslit and the ladder slits have not been moved since the new orbit. This could account for the lost of intensity on the beamline. The fact that I don't get the right counts suggest that I am not in an optimum position which could be responsible for the abnormal behavior of the intensity.


Vslit Optimization


I want to optimize the size of the horizontal beam on the mono. The idea is to get as wide a beam as the mono can take without over heating the xtal.

I am starting with the 2 dots collimator (300 µm) at lambda 1.1A.

I closed down the slit by a small increment (0.2 mm) and waited a minute or two. Then I collected 4-5 frames in order to see how reproducible the counts were.  Considering that the mono takes about 20-30 minutes to stabilize and that those points are separated by about 20 sec, we should not see a big variation. We observe at first that the counts go up as I close the Vslit down. The four blue line are for the same Vslit value. The difference is just a waiting period between taking the measurements. By the third measurement, as I was looking at the shape of the beam on a phosphorus paddle,  realignment of the collimator in the hutch was clearly needed. That gave the forth blue series with better counts. I continued closing the slit until the counts start going down again. That mean that I am cutting the beam smaller than the width of the xtal (last yellow series). I will keep this last value.






Stability of the beam


Each data point is for 90 sec frames with the pause on dump ( CBASS detects when the beam has dumped, pause the data collection, wait 15 minutes after the beam is enable again, realign, backs up one frames and recollect) and automax (realign macro activated) every 30 minutes. We can locate where the automax occurred since we have a point with a counts of zero where a dark was taken.

First observation the automax doesn't improve the counts during a normal data collection. The hutch setup doesn't need to be  realigned.

Second observation     the counts are slowly dropping from the beginning of the injection to the dump of the beam. We do not have more intensity at the end of a fill than at the beginning.

Third observation the next fill gives about the same count as for the previous fill.






22 March 2004

It seems like some PRT users are unhappy about the stability of the beam.

The frames were taken at 500 mm and show about 5-6 spots at very low resolution, ~17 A. The background is ~300 for the first frames while ~250 for the last frame of the data collection.
I called Tony Lenhard and he says that’s possible that the Ir/Ga coating in between the xtal and the copper block reacted over time. The contact would then be less efficient and affect the cooling of the xtal. This phenomenon has been seen  happens over a long period of time. This mono is old and it would be a more plausible thing than the Ir/Ga dripping off as Bob suggested.

Howard, Tony Lanzirotti and I went in the hutch to see if we had any problems with the cooling. We can see water flowing from the gravity tank and both proteus (cooling loop of the tank and mono) are turning. The slits are also water cooled and the proteus are turning.


19 March 2004
Photon shutter

Howard had problems centering his xtal. They were so small that there was no diffraction in one direction (the side view of the loop as oppose to the flat view. He finally was able to center the xtal and collected (1 dot collimator). His scaling shows that the intensity went up over a period of 10 hours. Howard thinks that there is a cooling problem with my mono.

Monochromator’s height:  trying to get the Merit to expected values

Since the last maintenance period, I sometimes get ½ to ¾ of the MERIT I expect according to my standard curves. It varies from one injection to the other. I troubleshoot the mono’s height with Ming. We had the 1 dot collimator. We watched the counts stabilize on C2 (collimator). We did some realign but no tilt scans. We moved lambda from 0.9 to 1.6A to see if the MERIT followed the calibration curves. The merit looked ok but not great for short lambdas. It followed the shape of the curve but with smaller values. We lowered the mono by ¼ turn and it made things worst. We raised the mono back to its position plus ¼ turn. It marginally improved the merit at 1.2 A and didn’t make it worst at 1.0 A. We raised it again by ¼ turn and it barely made any difference. We let it in that position. Howard will collect (px04-082).

18 March 2004
Photon shutter

The photon shutter is sluggish to respond since coming out of maintenance. It finally died. John Gallagher from interlock was called and he diagnosed a faulty solenoid. It was replaced during normal operation within an hour after lunch.