A User's Guide to BT-1

NIST Center for Neutron Research

Rules for BT-1 Operation

Information board and log book.: You MUST write the sample composition along with your name(s) and telephone number(s) on the white board. Also fill out all information requested in the BT-1 log book and on a BT-1 sample tag.
Shutter/Collimation operation.: This is a two-step procedure. First turn the knob to the desired position, then press enable. Do not turn the knob again until the green locked light comes on. Remove and take the shutter key when working in the beam path.
Monochromator/Collimation selection.: DO NOT CHANGE MONOCHROMATORS. At this time changes are to be made only by Brian, Judy or Qing. Be sure to set the collimation in your buffer if you are not using 15' (this is reset each time you edit the buffer unless A- mode is set).
Sample position.: The beam is 5/8" wide, and is focused to a 2" height at the sample position. To check your sample positioning with a Polaroid camera, close the shutter at the top closed position, place the camera behind the sample, turn the shutter control to the second closed position and press enable. This will rotate the drum through the 7' open position and give a reasonable exposure.
Zero point.: BT-1 does not have encoders, so the zero point should be checked periodically: Remove the beam stop from the X-rail; drive 2theta to zero (d4=0); center the primary beam in detector 1 (fpd 1,4); fit a Gaussian and drive there (y); initialize 2theta to zero (ini4=0). Replace the beam stop.
Scan range, step size, and time.: The default scan range of 3-13 degrees 2theta (1.3-11.3 for Ge311) with a step size of 0.05 is highly recommended so that each data point is counted in two different detectors. Make sure you have at least 4 prefactors. W+ produces a log file.
Sample checking.: If you are unsure of the quality of your sample, run a quick scan (15' collimation, range 5-10 degrees, 0.1 degree steps, 10 sec count).
ICP and data files.: Do not create files in someone else's subdirectory (especially mine). Exit ICP and use SET DEF [BT1.xxx], or log off and then restart.
Sample changer.: Do not attempt to mount the 6-position automatic sample changer without instruction/approval of Brian, Qing or Judy. The ICP command next advances to the next sample position.
Unloading Samples.: Survey your sample before removing it from the instrument. All V sample cans are to be stored in the BT-1 cabinet when not in use. Log sample cans to be removed from the BT-1 area. Never open cans in C100. At this time, Health Physics must clear irradiated powder samples before they are removed from their cans.

DO NOT ENTER THE DETECTOR AREA WHEN THE BEAM IS ON

BT-1 Sample Handling Procedures

This outlines recommendations for use BT-1 and storage and handling of samples. Exceptions from these recommendations will be made for experiments with unusual needs. All experiments at BT-1 will have a NIST employee either as an active participant or as a local coordinator. It is the responsibility of this NIST employee to ensure that users are properly trained, and to ensure that NIST procedures are followed with respect to sample handling.

Prior to experiments

BT-1 users are strongly recommended to evaluate the potential activation for their material using Web form http://www.ncnr.nist.gov/~toby/sample_info.html or by contacting Health Physics.

Qingzhen Huang (x-6164, B105) will provide vanadium sample containers, if available, to users. Requests for cans made at the last minute may not be accommodated.

A BT-1 Sample Tag should be filled out when the sample is loaded into the container. The sample tag should be attached to the sample, or to the outside of a cryostat, furnace, etc, when the sample is not in the instrument.

During Experiments

It is NCNR policy that all samples must be recorded in the log book as well as the instrument monitor (mrat) reading. For proprietary measurements, an approximate empirical chemical formula may be used.

The whiteboard must be updated to show the experiment, the name and phone numbers for an emergency contact and the names and institutions of all experimenters. The BT-1 sample tag(s) are to be hung on the whiteboard, when sample(s) are loaded in BT-1 for measurement.

After Experiments

A survey must be performed before the sample is removed from the instrument.

Until such time as appropriate facilities are available, all irradiated powder samples must be checked by Health Physics before they can be removed from their containers. Since samples are frequently unloaded by someone other than the person who loaded the sample, if special precautions or non-ambient conditions are needed for unloading the sample, note this on a pink sample tag.

The local contact is responsible for assisting users with unloading samples, clearing samples through Health Physics and shipping samples. When users must leave NIST without their samples, either because their experiments are not complete or the materials are neutron activated, the user must complete a BT-1 Sample Shipping Form and either provide a Material Safety Data Sheet (MSDS) or certify their sample is not hazardous (flammable, toxic, etc.). Samples without shipping forms may not be shipped.

Sample containers are to be stored after experiments in the unlocked cabinet adjacent to BT-1. If this is not possible because the sample requires special storage conditions, this should be discussed with a BT-1 instrument scientist prior to the experiment and the location of the sample container should be noted on the signout sheet on the outside of the cabinet. Likewise, if the sample and container will be used on another instrument, the signout sheet must also be used and the container be returned after the measurement. Qingzhen Huang will ensure that cans are returned after experiments and that local contacts unload cans when appropriate. As time permits, he will assist local contacts with unloading and shipping of samples.

Running a single sample

A single sample is run by setting up the parameters to be used for measurement in ICP and then initiating data collection. It is possible to set the data collection parameters in a second (ExtraICP) window while data collection on another sample is in progess. Note that the function key references (e.g. F17) refer to the labels on the console computer, not to the the letters on the keys.

Summary of steps

1.: Define parameters for run in ``Prepare mode.''
2.: Set the data collection time using Automon.
3.: Switch to control mode (F17) and check the run timing (HOWLONG I#).
4.: Check and log the neutron monitor (MRAT) as well as document the sample on: a sample tag, the white board and in the log book.
5.: Start the run sequence (RI#).

Detailed discussion

1.

In ``Prepare Mode,'' define a run to be measured in what is called an ``increment buffer'' (see Figure 1). Each ``buffer'' line defines the parameters for a single diffactometer scan.

**Figure 1:** Defining a single run in ``Prepare Mode.''
$\begin{figure} \centerline{ \epsfig {figure=setupm.ps,height=2.6in} }\end{figure}$

For use without a temperature controller, you will typically need to set the following fields in the buffer: Comment, T0, Monit, Prefac. and M-typ, which are used as follows:

Comment: This sets a 1-line file header and the name of the data collection file. Be sure to use letters and numbers (A-Z and 0-9) and no other characters for the first five letters of the Comment as this is used for the file name. If an invalid name is used the file will be named DEFLTxxx.BT1, where xxx is a number in the range 001 to 999.
T0: This specifies the nominal temperature for data collection. T0 should be 0, when temperature control is not being used. This causes the Wait, Err, Inc-T, and Hld0 values to be ignored.
M-typ: Is either ``NEUT'' or ``TIME''. NEUT is used for most data collection, where the data collection time is adjusted to match the neutron flux on the sample.
Prefac: Each data point is measured ``Prefac'' times and if Prefac is 4 or greater, the measurements are checked for statistical agreement, so that significant noise spikes can be discarded. A rule of thumb is that Prefac should be 4 for runs of 6 hours or less. It may be desirable to increase Prefac by 1 for each additional 6 hours of length, but 4 is a good default value regardless of the data collection time.
Monit: This value, along with Prefac, determines the length of the data collection period. If M-typ=TIME, this specifes a count time in seconds. Most commonly, M-typ=NEUT and Monit is set using the AUTOMON (AMON) feature.

It is very unlikely that you will want change the default values for some fields: A3-beg, A3-end, Inc-3, A4-beg, A4-end, Inc-4 and #pts. The Col field informs ICP of the in-pile collimation (15' or 7'). The default, 15' is usually correct. Note that the A4-*, #pts and Col values are reset every time a buffer is edited. There is one exception to this. If you are setting up runs while the instrument is collecting data and plan to use a different monochromator than the one that is currently in use, you may need to change the A4-beg and A4-end values to match the monochromator you plan to use. Use 3-13 degrees for Cu311 and Si531 and 1.3-11.3 degrees for Ge311. Note that the A- command in control mode turns off the automatic resetting of A4-beg and A4-end. Note that the field Hld should always be 0. Hld creates a delay that is executed at each data point. This is never of use at BT-1.

2.

Determine the data collection time using the Automon feature. The appropriate monitor value is computed so that the current run will finish at a specified time. Automon is initiated by moving the cursor to the AMON field and pressing Enter. The screen shown in Figure 2 then appears.

**Figure 2:** Using Automon to compute a run length.
$\begin{figure} \centerline{ \epsfig {figure=amon.ps,height=2.6in} }\end{figure}$

The number of days and the end time for the run are entered in the Automon page. Use 1 for delta-days if the run will go past midnight even if the run length is only a few hours. A run starting at 21:00 (9 pm) and ending at 9:00 (9 am) the next morning, would be entered as delta-days=1 and Time=9:00. The computed Monit value is set when Automon completes.

3.

Switch to control mode by pressing the F17 (actually F9) key. The length of a run sequence can be estimated using the HOWLONG I# command, where # is the buffer number (see Figure 3).

**Figure 3:** Computing the length of an ``increment buffer'' with the `HOWLONG I#` command.
$\begin{figure} \centerline{ \epsfig {figure=howlongm.ps,height=2.6in} }\end{figure}$

4.

Before starting the run be sure to:

(a): be sure the shutter is open
(b): measure the monitor using the MRAT command (see Figure 3).
(c): enter the sample composition and contact info on the white board
(d): put the sample tag in the holder on the white board
(e): enter the sample information in the log book

5.

The run is started with a ``RI'' command, as shown in Figure 4.

**Figure 4:** Starting a single run with the `RI` command.
$\begin{figure} \centerline{ \epsfig {figure=ri.ps,height=2.6in} }\end{figure}$

Using a ``Displex'' closed-cycle He refrigerator

The closed-cycle He gas refrigerators used at the NCNR are not properly Displex devices, since that term is a trade name. I mention the term ``Displex'' though because the closed-cycle He gas refrigerators at the NCNR are commonly and incorrectly referred to using that name.

Most units can be used to control sample temperatures over the range 10-300 K, but a few units have been modified to allow operation to 450 K or higher. Do not use the standard units above 310 K as this can destroy the cooling head.

While it is possible to manually set the temperature controller attached to the refrigerators, the ICP program can control the temperature of the refrigerator, so a series of measurements can be made at different temperatures automatically. The steps to be followed for automatic usage are outlined below and then will be discussed subsequently in more detail.

Note that the function key references (e.g. F17) refer to the labels on the console computer, not to the the letters on the keys.

Summary of steps

1.: Load a sample, pump down and cool.
2.: Connect the temperature controller to the VAX.
3.: Set the temperature controller type in ICP (TDEV).
4.: Confirm communication (PT).
5.: Define parameters for run in ``Prepare mode.''
6.: Set the data collection time using Automon.
7.: Copy and edit the parameters for additional runs.
8.: Set up a run sequence (RS=...).
9.: Switch to control mode (F17) and check the run sequence timing (HOWLONG/RS).
10.: Check and log the neutron monitor (MRAT) as well as document the sample on: a sample tag, the white board and in the log book.
11.: Start the run sequence (RS).
12.: Remove the refrigerator from the instrument.

Detailed discussion

1.

You should ask for help loading a sample, if you are not comfortable with the task. It is better demonstrated than described. Samples are usually sealed with at least a few percent He gas to promote thermal equilibration and then sealed with a indium gasket, but note that indium melts at temperatures in the operating range of the high-temperature refrigerators.

Attach the sample can to the copper block with screws that are the correct length as screws that are too long can damage the heating element or temperature sensor. Also, make sure these screws are tight, so that the screws do not loosen due to vibration.

The inner heat shield cans should not be screwed on tightly, as they may be impossible to remove after temperature cycling. Back them off by a quarter turn.

Attach the outer vacuum can and pump down with a turbopump. As soon as the pressure is in the range of $10^{-3}$ torr, the cooling can be started. It will typically take about 3 hours to bring the sample to the minimum temperature. I do not recommend attempting to collect room temperature data with the set point at 295 K, while the displex is initially cooled as temperature control may not be good. When I collect data over a range of temperatures, I tend to start at the lowest temperature first, but I cannot defend this choice to be the best.

For runs of a day or so, one can usually operate the refrigerator without continual pumping, but cooling can fail if the pressure rises, so it is usually worth the effort to leave a pump attached while data are collected. Always use a pump with the high-temperature refrigerator, when operating above room temperature in case of outgassing.

2.

Physically connect the BT-1 temperature controller RS-232 connector to the refrigerator. The cable from the VAX is marked ``BT1 Temperature Control'' and is connected either to the rear or the top of the refrigerator temperature controller. The cable can usually be found in the vicinity of the shield wall with the ``white board'' attached. Take care not to use the nearly identical ``BT1 it Magnet Control'' cable.

3.

Instruct ICP (in Control Mode) which temperature controller you will be using with a tdev command (see Figures 5 and 6). Note that the tdev command must be reentered each time ICP is restarted and must be reentered if you change the controller model when a new refrigerator is mounted.

**Figure 5:** Menu of temperature controllers from the `tdev` command.
$\begin{figure} \centerline{ \epsfig {figure=tdev1.ps,height=2.6in} }\end{figure}$

**Figure 6:** Selecting a temperature controller with the `tdev` command.
$\begin{figure} \centerline{ \epsfig {figure=tdev2.ps,height=2.6in} }\end{figure}$

The tdev command sets the T+ flag in ICP, so that the temperature is recorded at each data point.

4.

Check that the temperature controller is reading correctly using the pt command (optional, see Figure 6).

5.

In ``Prepare Mode,'' define a run to be measured in what is called an ``increment buffer'' (see Figure 7). Each ``buffer'' line defines the parameters for a single diffractometer scan.

**Figure 7:** Defining a single run in ``Prepare Mode.''
$\begin{figure} \centerline{ \epsfig {figure=setup.ps,height=2.6in} }\end{figure}$

For use with a temperature controller, you will typically need to set the following fields in the buffer: Comment, T0, Wait, Err, Hld, Monit, Prefac. and M-typ, which are used as follows:

Comment: This sets a 1-line file header and the name of the data collection file. Be sure to use letters and numbers (A-Z and 0-9) and no other characters for the first five letters of the Comment as this is used for the file name. If an invalid name is used the file will be named DEFLTxxx.BT1, where xxx is a number in the range 001 to 999.
T0: This specifies the nominal temperature for data collection. This value is sent as the set point to the temperature controller. For the controllers attached to the He refrigerators, this is a temperature in K. If T0 is set to 0, and the T+ flag is set, the temperature will be recorded, but will not be changed and the Wait and Hld0 terms (below) are ignored.
Wait: This specifies a maximum amount of time in minutes that ICP will wait for the sample temperature to be in range (see ERR, below), before starting data collection time. If the desired sample temperature is reached in less time, the remaining time wait is not used. Typical practice is to use a wait that is much longer than the expected time needed to reach the desired temperature, for example 120 to 180 minutes. If you do not want data collection to wait for the temperature to be reached, Wait can be set to 0.
Err: The temperature is considered ``in range'' if the temperature is between T0+ERR and T0-ERR. Note that the value for ERR does not affect the actual stability of the temperature (which is determined by the PID parameters set in the temperature controller) so setting ERR to a small number, can cause data collection to be suspended for long periods when temperature control is flaky. Typical values for ERR are 2 to 5 K for low temperature measurements, but may be 5 to 10 K near room temperature or above.
Hld0: This specifies an amount of time in minutes to wait for temperature to equilibrate after the temperature is reached (or Wait expires) before data collection is started. The desired value for this parameter will depend on the experiment to be performed. A value of 20 minutes is common, but so is 0 as well as longer times.
M-typ: Is either ``NEUT'' or ``TIME''. NEUT is used for most data collection, where the data collection time is adjusted to match the neutron flux on the sample.
Prefac: Each data point is measured ``Prefac'' times and if Prefac is 4 or greater, the measurements are checked for statistical agreement, so that significant noise spikes can be discarded. A rule of thumb is that Prefac should be 4 for runs of 6 hours or less. It may be desirable to increase Prefac by 1 for each additional 6 hours of length, but 4 is a good default value regardless of the data collection time.
Monit: This value, along with Prefac, determines the length of the data collection period. If M-typ=TIME, this specifies a count time in seconds. Most commonly, M-typ=NEUT and Monit is set using the AUTOMON (AMON) feature.

Note that two fields, Hld and Inc-T, should always be 0. Inc-T causes the temperature to be changed for each data point and Hld creates a delay that is executed at each data point. These processes are almost never of use at BT-1.

6.

**Figure 8:** Using Automon to compute a run length.
$\begin{figure} \centerline{ \epsfig {figure=amon.ps,height=2.6in} }\end{figure}$

The Automon computation can either use or ignore the Wait and Hld0 values. If you answer Y for ``Use Holds,'' the time needed for the Hld0 (and Hld) hold is included in the run length computation. If you answer Y for ``Use TempWait,'' the entire Wait period is included in the run length computation. Since the entire Wait period is usually not used, it is best to say N for ``Use TempWait'' but the answer for ``Use Holds'' is a matter of personal convenience. The number of days and the end time for the run are then entered. Use 1 for delta-days if the run will go past midnight even if the run length is only a few hours. A run starting at 21:00 (9 pm) and ending at 9:00 (9 am) the next morning, would be entered as delta-days=1 and Time=9:00. The computed Monit value is set when Automon completes.

7.

Duplicate the run information for other temperatures that you will wish to run. This is done either by highlighting the buffer to be copied and then pressing the F14 key (actually the F6 key), which will copy the information to all other buffers, or (preferably) the buffer can be copied selectively by pressing the F18 key (actually F10) to enter the ``Buffer Ops'' mode where a buffer can be copied by entering a command such as COPY 2,3 (see Figure 9).

**Figure 9:** Copying a buffer in ``Buffer Ops'' mode.
$\begin{figure} \centerline{ \epsfig {figure=copy.ps,height=2.6in} }\end{figure}$

This copies the parameters in buffer #2 into buffer #3. Exit ``Buffer Ops'' mode by pressing the F20 key (actually F12). Each buffer can then be quickly modified to change the temperature (T0) and possibly the Comment, Hld0, Err and Wait values.

8.

Once a series of runs has been defined, a ``run sequence'' can be defined by pressing the F19 key (actually the F11 key). This brings up the menu shown in Figure 10. If a previous sequence is present, it can be cleared by typing DEL and return.

**Figure 10:** Beginning a ``Run Sequence.''
$\begin{figure} \centerline{ \epsfig {figure=rs1.ps,height=2.6in} }\end{figure}$

Commands are added to the run sequence by typing RI#, where # is the buffer number of the run in the list. Commands may be entered one a time or several commands may be entered at once, separated by semicolons (;). The run sequence in Figure 11 will cause buffer #1 to be collected three times and then buffer #2 to be collected twice. Note that the files will all be named NALICxxx.BT1, so if no other files exist, the data files will be named NALIC001.BT1, NALIC002.BT1 and NALIC003.BT1 for 15 K runs and NALIC004.BT1 and NALIC005.BT1 for the 295 K runs. Exit the RS menu with the F20 key (actually F12).

**Figure 11:** Entering a ``run sequence.''
$\begin{figure} \centerline{ \epsfig {figure=rs2.ps,height=2.6in} }\end{figure}$

9.

Switch to control mode by pressing the F17 (actually F9) key. The length of a run sequence can be estimated using the HOWLONG/RS command (see Figure 12).

**Figure 12:** Using the `HOWLONG/RS` command to determine the expected run time for a ``run sequence.''
$\begin{figure} \centerline{ \epsfig {figure=rs4.ps,height=2.6in} }\end{figure}$

Note that the estimated lengths will be estimated assuming the maximum delay allowed by Wait and the minimum assumes that all Wait times are negligible. In the example shown in Figure 12, there are no actual temperature changes, except between the third run and the fourth, and perhaps before the first run. Assuming that the sample is already at the appropriate temperature and the refrigerator will need approximately one hour to heat from 15 K to 295 K, a good estimate is that the runs will require 32.2 hours. Your mileage may vary.

10.

Before starting the run be sure to:

(a): be sure the shutter is open
(b): measure the monitor using the MRAT command
(c): enter the sample composition and contact info on the white board
(d): put the sample tag in the holder on the white board
(e): enter the sample information in the log book (see Figure 12).

11.

The run sequence is started with a RS command, as shown in Figure 13.

**Figure 13:** Starting a ``run sequence'' with the `RS` command.
$\begin{figure} \centerline{ \epsfig {figure=rs5.ps,height=2.6in} }\end{figure}$

Note that it is possible to modify the run sequence or change the measurement parameters for the runs that have not been started in another (ExtraICP) window while the data collection is in progress.

12.

When removing the refrigerator, close the vacuum valve before turning off the vacuum pump. Let the pump vent completely before removing the vacuum hose - this takes 5 to 10 minutes. The compressor may be turned off at any time. If you plan to unload your sample as soon as possible, you may wish to set the temperature set-point to 295 K, though the cold-head will remain very cold for many hours even when the sample has reached room temperature. When possible, let the refrigerator warm up slowly by letting it sit for a day or so before releasing the vacuum and removing the sample. Remember to leave the sample tag on the refrigerator, so that the identity of the sample is known.

If samples will be changed quickly, it may be necessary to use a heat gun to drive off condensation, but be careful not to heat the cold-head or sample stage to much more than room temperature. The refrigerator can be severely damaged by heating it above about 50 C.

Using the six-position sample changer

The six-position sample changer can be used to collect room temperature data on up to six different samples under control of the data collection program, ICP. The steps to be followed for automatic usage are outlined below and then will be discussed subsequently in more detail. Note that the function key references (e.g. F17) refer to the labels on the console computer, not to the the letters on the keys.

Summary of steps

1.: Mount the sample changer and connect the motor controller.
2.: Ensure the changer works and the correct sample is rotated into position (NEXT).
3.: Define parameters for run in ``Prepare mode.''
4.: Set the data collection time using Automon.
5.: Copy and edit the parameters for additional runs.
6.: Set up a run sequence (RS=...).
7.: Switch to control mode (F17) and check the run sequence timing (HOWLONG/RS).
8.: Check and log the neutron monitor (MRAT) as well as document the sample on: a sample tag, the white board and in the log book.
9.: Start the run sequence (RS).

Detailed discussion

1.

Physically move the sample changer into position and connect the two motor controller cables marked ``Sample Changer'' A and B. They are polarized so that they can only be connected in one way.

2.

Load your samples into the changer, paying careful attention to the position number of each sample. A good practice is to fill out a sample tag for each sample before you load the samples and then mark down the position number on the sample tag.

3.

Check that the sample changer advances properly by typing NEXT. Note that it takes about 1 minute for the sample changer to completely advance and the Motor 15 display to be set correctly. If the display is not correctly synchronized with the actual position, the display can be reset using the INI15=# command, where # is 1,2,3,4,5 or 6.

4.

In ``Prepare Mode,'' define a run to be measured in what is called an ``increment buffer'' (see Figure 14). Each ``buffer'' line defines the parameters for a single diffactometer scan. For convenience, you may wish to use the buffer numbers that correspond to the positions of your samples in the sample changer.

**Figure 14:** Defining a single run in ``Prepare Mode.''
$\begin{figure} \centerline{ \epsfig {figure=setupm.ps,height=2.6in} }\end{figure}$

For use with the sample changer, a temperature controller, you will typically need to set the following fields in the buffer: Comment, T0, Monit, Prefac. and M-typ, which are used as follows:

Comment: This sets a 1-line file header and the name of the data collection file. Be sure to use letters and numbers (A-Z and 0-9) and no other characters for the first five letters of the Comment as this is used for the file name. If an invalid name is used the file will be named DEFLTxxx.BT1, where xxx is a number in the range 001 to 999.
T0: This specifies the nominal temperature for data collection. T0 should be 0, when temperature control is not being used. This causes the Wait, Err, Inc-T, and Hld0 values to be ignored.
M-typ: Is either ``NEUT'' or ``TIME''. NEUT is used for most data collection, where the data collection time is adjusted to match the neutron flux on the sample.
Prefac: Each data point is measured ``Prefac'' times and if Prefac is 4 or greater, the measurements are checked for statistical agreement, so that significant noise spikes can be discarded. A rule of thumb is that Prefac should be 4 for runs of 6 hours or less. It may be desirable to increase Prefac by 1 for each additional 6 hours of length, but 4 is a good default value regardless of the data collection time.
Monit: This value, along with Prefac, determines the length of the data collection period. If M-typ=TIME, this specifes a count time in seconds. Most commonly, M-typ=NEUT and Monit is set using the AUTOMON (AMON) feature.

Note that the field Hld should always be 0. Hld creates a delay that is executed at each data point. This is never of use at BT-1.

5.

**Figure 15:** Using Automon to compute a run length.
$\begin{figure} \centerline{ \epsfig {figure=amon.ps,height=2.6in} }\end{figure}$

6.

Complete buffers for all the samples you want to measure. If many parameters are similar, it may be useful to duplicate the run information for other temperatures that you will wish to run. This is done either by highlighting the buffer to be copied and then pressing the F14 key (actually the F6 key), which will copy the information to all other buffers, or (preferably) the buffer can be copied selectively by pressing the F18 key (actually F10) to enter the ``Buffer Ops'' mode where a buffer can be copied by entering a command such as COPY 2,3 (see Figure 16).

**Figure 16:** Copying a buffer in ``Buffer Ops'' mode.
$\begin{figure} \centerline{ \epsfig {figure=copy.ps,height=2.6in} }\end{figure}$

This copies the parameters in buffer #2 into buffer #3. Exit ``Buffer Ops'' mode by pressing the F20 key (actually F12). Each buffer can then be quickly modified to change the appropriate fields.

7.

Once a series of runs has been defined, a ``run sequence'' can be defined by pressing the F19 key (actually the F11 key). This brings up the menu shown in Figure 17. If a previous sequence is present, it can be cleared by typing DEL and return.

**Figure 17:** Entering a ``Run Sequence'' for the sample changer.
$\begin{figure} \centerline{ \epsfig {figure=rsnext.ps,height=2.6in} }\end{figure}$

Commands are added to the run sequence by typing RI#, where # is the buffer number of the run in the list. Commands may be entered one a time or several commands may be entered at once, separated by semicolons (;). Be sure to use a NEXT command between each RI# command so that the sample changer will be advanced to the next sample. The run sequence RI1;NEXT;RI2;NEXT;RI3 buffer 1 to be collected and the sample changer will be advanced and buffer 2 will be collected. The sample changer will be advanced again and buffer 3 will be collected. To skip a sample position, two next commands can be used: RI1;NEXT;NEXT;RI3 and it is also possible to collect multiple data sets on a sample: RI1;RI1;NEXT;RI2;NEXT;RI3. Exit the RS menu with the F20 key (actually F12).

8.

Switch to control mode by pressing the F17 (actually F9) key. The length of a run sequence can be estimated using the HOWLONG/RS command (see Figure 18).

**Figure 18:** Using the `HOWLONG/RS` command to determine the expected run time for a ``run sequence.''
$\begin{figure} \centerline{ \epsfig {figure=howlongnext.ps,height=2.6in} }\end{figure}$

9.

Before starting the run be sure to:

(a): be sure the shutter is open
(b): measure the monitor using the MRAT command
(c): enter the sample composition and contact info on the white board
(d): put the sample tag in the holder on the white board
(e): enter the sample information in the log book.

10.

The run sequence is started with a RS command, as shown in Figure 19.

**Figure 19:** Starting a ``run sequence'' with the `RS` command.
$\begin{figure} \centerline{ \epsfig {figure=rs5.ps,height=2.6in} }\end{figure}$

Note that it is possible to modify the run sequence or change the measurement parameters for the runs that have not been started in another (ExtraICP) window while the data collection is in progress.

Retrieving and Processing BT-1 Files

Retrieving files

**Figure 20:** The initial window from `rem_fetch`.
$\begin{figure} \centerline{ \epsfig {figure=fetch1.eps,height=1.2in} }\end{figure}$

After data has been collected, it can be transferred using the UNIX command, rem_fetch. When this command is invoked, a graphic user interface (GUI) appears, as seen in Fig. 20. For the ``Remote User Name'' enter the directory used to collect data, for example, guest1 for [BT1.GUEST1] or sam.nano for [BT1.SAM.NANO]. Press the ``Get File List'' button to see a list of files displayed. By default, all files in the appropriate CNBS directory are then displayed (as seen in Fig. 21), except that files previously transferred into your current directory are not included. This behavior can be changed by deselecting the ``Omit existing files'' checkbutton, which causes all files to be displayed. It is also possible to select the files that will be listed using the ``File filter.'' For example, use of *ac* for a file filter restricts the file list to files containing the string ``ac''.

**Figure 21:** A list of files displayed in `rem_fetch`.
$\begin{figure} \centerline{ \epsfig {figure=fetch2.eps,height=3.0in} }\end{figure}$

After the ``Get File List'' button is pressed, the window is modified to include the file list, as seen in Fig. 21. At this point it is possible to modify any of the previously described fields and then press the ``Get File List'' button to obtain an updated list.

Multiple files are selected by clicking on them while holding the shift key down, or groups of files can be selected by using control while clicking on them. Once a set of files has been selected, pressing the ``Transfer File(s)'' or the ``Transfer File(s) and Quit'' button causes the files to be transferred to the local computer. If the ``Run proprep & gformat'' checkbutton is selected, each file is individually processed by the proprep and gformat programs (see below) as the files are transferred.

Processing files

Program proprep is used to convert BT-1 data files to an intermediate format where data are separated by detector number. The gformat program may then used to scale, interpolate, and combine the data to appear as if it were collected using a single detector.

proprep

Program proprep is used to convert .bt1 files into .raw files used by a number of programs, such as REFINE and are used as an intermediate for conversion to GSAS format. It can be used to add multiple runs, but this is better done in program gformat, if files for GSAS are to be produced.

The program can be used by typing:

       proprep file

       proprep file.bt1

Either command causes file file.bt1 to be read and a new file, file.raw, to be created. Several (up to 50) files can be processed at one time:

       proprep file1 file2 file3

or even,

       proprep abc*.bt1

where the .bt1 extension is assumed if not specified. The program will produce an error message and fail if the .raw file exists.

It is important to look carefully at your data before execution of proprep. For instance, you may wish to exclude data below 5 degrees in detector 1, 10 degrees in detector 2, and 15 degrees in detector 3 using the option -cx.x (see below), as the background levels in these regions is often anomalously large. The default is to exclude data below 3.0, 8.0, and 13.0 degrees, respectively for detectors 1, 2 and 3, which would affect only data taken with the Ge(311) monochromator under standard data collection conditions.

A number of options can be supplied to proprep:

       proprep [-s] [-k] [-1] [-cx.x] [-dx,y] [-p] file1 [file2] ...

or proprep can be used as a filter:

       proprep -f [-k] [-1] [-cx.x] [-dx,y] < file.bt1 > file.raw

The full list of options accepted by proprep follows. Options can be combined together where appropriate.

-s: sum all specified files into a single file. The file will be named using the first input file. Use of gformat rather than proprep for this purpose is recommended because gformat checks that the files are in statistical agreement.
-k: keep all data from detectors 1, 2 and 3.
-1: keep all data from detector 1, but discard data below a threshold from detectors 2 and detector 3 (used to preserve very low angle peaks).
-cx.x: Sets the threshold for cutting low angle data from detectors 1, 2 and 3 (unless -k or -1 is used). The default threshold is 3.0 meaning that data below 3.0, 8.0 and 13.0 degrees are cut from detectors 1, 2 and 3, respectively. This means that by default, data is only cut from Ge(311) scans.
-dn: drop data from detector n, or if more than one number is used, for example, if ``-d1,3,5'' is used, data from banks 1, 3 and 5 will be dropped.
-p: pipe output to stout rather than create an output file.
-f: run as a filter: read from stdin and write output to stdout.

gformat

Program gformat is used to convert .raw files into pseudo-single-detector .gsas files that are used by GSAS and a few other programs. The zero corrections and scale factors found in the .bt1 file header (and written in the .raw file) are applied to the data. The program can be used to add together runs of different lengths, but the first file specified should start at the lowest angle. When points are averaged together, a statistical check is applied to see how well the points agree. Warnings about points that differ by more than $4\sigma$ are appended to the end of a file named stat.check. When warnings are generated, a message is displayed on the user's terminal.

In the simplest form, program gformat is used by typing,

       gformat abc.raw

to process abc.raw into abc.gsas. Note that the .raw extension is assumed if not specified for input files, so that the command

       gformat abc

is equivalent to the previous command.

Up to 50 files can be specified on the command line. Each will be processed individually (unless -s is present, which causes the contents of the files to be averaged, see below). To process a series beginning with the letters ``abc'' you can use the commands

       gformat abc*.raw

Background corrections are applied to each detector bank to give the best agreement for overlapping data. This correction can be omitted using the -o option. The program will not overwrite an existing .gsas file unless -c is specified.

When a zero correction is applied to each detector bank, data points will no longer ``line up'' with the data points in the output file. Thus, gformat will interpolate the intensity for each point when it bins. This creates some correlation between adjacent points, which is not strictly appropriate for correct least-squares statistics. If -m is used the points are placed into the closest bin without interpolation. This introduces a bit more scatter into the data, but is statistically more valid. The -m option causes your output step size to be double that of your input, which reduces the scatter that is introduced, so if you are going to use -m, you should probably collect data with a smaller step size 0.02 or 0.025 degrees).

The data for all detector banks is included in the .gsas file, unless the -L option is used (see below). A .inst file is created with the same file name as every .gsas output file. This contains the wavelength and other sundry information.

The .gsas and .inst files created by gformat can be e-mailed, ftp'ed etc. However, before using the files in GSAS, they need to be converted to direct access. This is done automatically for use with UNIX versions of GSAS by gformat. The direct access data files are named with all capital letters. When using e-mail or for non-UNIX versions of GSAS, transfer the lower-case named files.

A number of options can be supplied to gformat:

       gformat [-s] [-c] [-o] [-m] [-Lxx] [-a] file1 [file2] ...

or gformat can be used as a filter:

       gformat -f [-Lxx] [-o] [-m] [-a] < file.raw > file.gsas

The full list of options accepted by gformat follows. Options can be combined together as appropriate.

-s: average all specified files into a single file. The file will be named using the first input file.
-Lxx: discard the data from detector number xx and subsequent detectors.
-c: overwrite existing .gsas file(s) rather than stop.
-m: move points to closest bin rather than interpolate.
-o: don't apply a detector-by-detector background spline correction.
-a: apply an absorption correction using the coefficients from file abs.corr
-f: run as a filter: read from stdin and write output to stdout.

The -g option requires an absorption correction input file, abs.corr, which consists of a free-format file containing a one-line title followed by the number of Chebyshev terms (up to 30) and the Chebyshev polynomial coefficients. The polynomial is evaluated for

$\begin{displaymath} x = 2\theta/90. - 1\end{displaymath}$

and the intensity and standard uncertainty (esd) are multiplied by the result. Typically this polynomial is fitted to A* values from the International Tables.

Plotting BT-1 files

The cmpr program can be used for plotting data in a variety of formats. Figure 22 shows the dialog box used for reading data. Note that pseudo-1D data is processed by the gformat program while multi-detector data does not combine the individual detectors (but does apply the zero offset and efficiency corrections). To read a file, select the format, select the file to be read and click on ``Read Selected File.'' It is also possible to read a file by double-clicking on it.

**Figure 22:** The cmpr read dialog
$\begin{figure} \centerline{ \epsfig {figure=cmpr1.eps,height=1.6in} }\end{figure}$

Files can also be read by starting cmpr with a command option:

      cmpr -readopt file1 file2 -readopt file3

The following read-options are available:

-1d: The following files are to be converted to pseudo-1d data
-md: The following files are to be read as muilti-detector data
-gsas: The following files are to be read as GSAS data
-x7a: The following files are to be read in BNL X7A format

After data has been read, it can be plotted using the dialog box in Figure 23. Multiple files can be selected for plotting using shift plus the left mouse button and then using the ``Update Plot'' button (or double-clicking). It is also possible to change the way one or more files are displayed by selecting the files then selecting options for the Line, Color, or Symbol and then pressing ``Apply Changes.''

**Figure 23:** The cmpr plot dialog
$\begin{figure} \centerline{ \epsfig {figure=cmpr2.eps,height=1.6in} }\end{figure}$

Note that when a file is plotted, it is possible to ``zoom-in'' using the left mouse button to click on two corners of the new region. The right mouse button zooms out and the middle mouse button opens a window for manual scaling.

The HKLGEN and EditCell dialog boxes (see Figure 24) are used to generate the allowed reflection positions for a given set of unit cell parameters and optionally using space group extinctions. In EditCell, these reflection positions can be superimposed as vertical lines on one or more sets of data. The cell parameters can be changed and the reflection positions will shift as the lattice constants change. Also, extinct reflections can be highlighted. To see the indices for a reflection, shift-click on the line.

**Figure 24:** Generating and displaying reflection positions in cmpr
$\begin{figure} \centerline{ \epsfig {figure=cmpr3.eps,height=1.6in} \epsfig {figure=cmpr4.eps,height=1.6in} }\end{figure}$

The cmpr program can also be used for peak fitting, as shown in Figure 25. The program uses the GPLSFT program, as written by David Cox and Larry Finger, to do the actual fitting.

To fit one or more peaks, first select the dataset to fit, then zoom-in to the region to fit using the left-mouse button and press ``Set range from graph'' (or manually enter the 2-theta range). Peaks may be entered or modified by clicking on the appropriate ``Set'' button and then by clicking on the appropriate location on the graph. The ``Use'' checkbuttons select if a peak is included in the fit. The remaining checkbuttons determine the variables that are refined when the ``Run GPLSFT'' button is pressed.

**Figure 25:** Peak fitting in cmpr
$\begin{figure} \centerline{ \epsfig {figure=cmpr5.eps,height=1.6in} \epsfig {figure=cmpr6.eps,height=1.6in} }\end{figure}$

The cmpr program has other options. For example, the Rescale dialog can be used to multiply or add a constant to a dataset's x-axis or y-axis. It can also be used to change the units for plotting a dataset. For example, intensities can be shown on a logarithmic scale or 2-theta can be converted to 2-theta at different wavelength, or d-spaces or Q. PlotOptions controls how PostScript output is treated. A command for plotting the output, or the name to be used for saving output, can be specified here.

Writing files to floppy disk

Files can be transferred to a DOS format floppy disk quite simply.

1.

Create a UNIX window on the console computer by pressing the ``BT1'' button on the button bar to the lower right of the screen.

2.

Insert a formatted 3.5'' floppy disk in the approporiate slot.

3.

Type the following command:

       mount -v /a

the computer will respond:

       /dev/fd0 on /a type vfat (rw,noexec,nosuid,nodev)

or will report an error.

4.

Transfer files to the floppy. This can be done in a single step, by transferring files directly, or by moving files from another computer, such as rrdjazz. One should not run gformat on directly on a floppy, however, because DOS disks cannot contain two files having the same name, but differing in the use of capitalization (e.g. MYRUN001.GSAS and myrun001.gsas).

(a)

Transfer files directly to floppy:

        (cd /a/; rem_fetch)

(b)

Copy files from jazz. The following command copies all files names beginning with a lower case letter, to the floppy from subdirectory sunysb in account guestbt1 on jazz.

        scp "guestbt1@jazz:sunysb/[a-z]*" /a/

to include subdirectories use scp -r ... in the above command.

5.

Check the contents of the floppy

        ls /a/

6.

Unmount the floppy. (This is important!)

        umount -v /a

The computer will respond

        /dev/fd0 umounted

If the computer responds

        umount: /a: device is busy

you have a process running in the /a/ directory. Type

cd

in all UNIX windows or if need be log out or all windows (other than the ICP process) on the console until you can run the umount command without an error.

7.

Remove the floppy.

About this document ...

A User's Guide to BT-1

This document was generated using the LaTeX2HTML translator Version 97.1 (release) (July 13th, 1997)

The command line arguments were:
latex2html -split 0 -no_math -no_navigation -scalable_fonts -local_icons guide.

The translation was initiated by Brian Toby on 4/22/1999

Brian Toby
4/22/1999