Checklist details

Run Checklist Details

BE SURE NUMLOCK IS TURNED OFF. IT CAUSES PROBLEMS IN LINUX.
If resolution on monitor is to low, hit ctr-alt-minus to increase it.
Try pressing F1 if the desktop is frozen.
Or try left click on desktop, exit FVWM, restart.

Pot Motion Software Initialization

If the Pot Motion Software is running, there should be two windows visible. One window is small and contains a button labeled Emergency Stop. The other window is the main screen of the Pot Motion Software. It contains a list of all the pot locations, the current LVDT value for that location and the rates at that location if the HV is turned on and the DAQ. A frog will be hopping along the bottom of the window if the program is active.
If you do not see all of these things or the Pot Motion Software is not responding then close any applications that are running.
To start from d0ol49 (use d0ssh to login if needed):
- Type cd ~d0fpd/python_mux at a command prompt in an xterm window.
- Type ./FPDGui.py& at the command prompt to bring up the software.
This should open the emergency stop button applet and the main Pot Motion Control window with the frog jumping along the bottom. Do not proceed any further until the software is running.
When the program is started, if all pots are not at home, an initialization window will open which reads "Pots are not in home, move them home first". To bring the pots home, acknowledge the message and turn on the 5V control lines. You will see the LVDT bars change color (from blue to red and back to blue) as the power is turned on at the castle. When it has returned to stable (blue), go to the Pots Home --> Init button. The amplifiers and drivers will be turned on, the pots will come home at high speed and the amplifiers and drivers will turn off. Then turn off the control lines.
If any LVDT bars are red there is a hardware failure of some kind. See the Hardware Failure descriptions.
If any LVDT bars are orange, the A/S are turned off. Turn them on from the control menu in the Pot Motion Main Screen.
See Known Issues Page for known problems.

Pot Motion

The control line switch needs to be turned on for the pots you want to move.
The drivers and amplifiers for the individual castles to be moved need to be turned on. To do this, go to COMMANDS->DRIVER and COMMANDS->AMPLIFIER and select the appropriate castle. In the Driver Status panel, it should turn green when it has power (both driver and control line on). Amplifiers should be on at all times.
Go into expert mode in the Pot Motion Software.
Use the group tab to insert pots to different positions or use the insert tab to insert all pots to the same position.
Select the pot(s) you are going to move. Click on the locked button to get a dropdown list from which you select on.
Type the value to which you want to move the pots in mm. speed and click run. A window will pop up to verify your command. Click ok and the pots will start to move.
On the main screen, LVDT bars should be blue for enabled pots, orange for disabled pots (disabled through the software) and red for pots experiencing hardware failure. If they stay red, see the Hardware Failure descriptions.
Back on the main screen, the LVDT bars should turn green if the pot is moving and the slider should change. Click on the name of the pot to open a window with information about the direction, speed, location and desired location as well as any alarms. Click the name of the pot in the main screen again to close this window.
SELF STOP: In expert mode you can click on a single pot that is moving, select stop from the drop down and then click on stop to stop just that pot. It will need to be brought home before it can be reinserted.
ALL STOP: In expert mode you can click on the stop button next to the home button to send a stop command to all pots. If you do this, all pots that were in the middle of a move need to come home before they can once again be inserted.
EMERGENCY STOP: If none of these commands work, you can hit the Emergency Stop button which continuously sends a stop command to all pots. When you click it again, you can again move the pots. Bring home all pots that were in the middle of a move before you insert them again. If a pot doesn't obey any of the stop commands, turn off the 5V control circuit.
To bring all pots home, turn on control line. Select Pots Home -> Init. This will turn on drivers and amplifiers as needed, bring pots home at high speed and turn drivers and amplifiers off when they reach home. Then turn off the control line.

Software Limits / Disables

To establish maximum positions, go into the expert screen and select the pots tab. Type the desired value in the Max Position field (normally, this should be the same as the latest green table value in the insertion tables). You can also disable motion for a pot from this tab (when disabled, the values are greyed out, pot is unable to be moved and LVDT value is zeroed). After making any change, click on the SAVE button (this also updates the values in EPICS) and then the reload button.
See the Pot Motion Document in the binder for further details.

HV Control Software Initialization

From the Pot Motion Control Main window, go to Command and click on High Voltage. After a few minutes, you should see the HV Control Window Open. Or type /online/config/fpd/hv/fpd.hv & (type /online/config/fpd/hv/fpd2.hv & to be abble to access the spare channels).
Verify that the tubes are off. The control state box should be blue for all tubes.
If unexpected lines are blank, press the reconnect button to update the information.
See the HV Control Document for further details.
See Known Issues Page for known problems.

L0 HV Setup

Click on the Trigger Tab
Turn voltages to the value listed in the runplan (can use set 100%). Do not increase the voltage to any tube once the rates have exceeded ~350kHz.
Verify that the current being drawn is as expected. The expected value can be found on the Standard Operating Values page in the binder.
If the current draw window is yellow or red, then the current draw has moved into an alarm zone. Make a note in the logbook.
Set the HV for the Veto counters in the same way to -1500V (can use set 100%) and check currents.
We will be setting the tubes to 0% when not collecting data. This is the Standby value. VETON goes to a PAUSED state when you do this. To clear the PAUSED state (and the attendent minor alarm) turn this channel off, then back on which puts it in the ON state.
If the current limits are too low, make a note of it and follow the epics spy instructions below to temporarily increase the limits.
See the HV Control Document for details.

MAPMT HV Setup

Click on the MAPMT Tab
Turn voltages to -940V (can use set 100% for this) for operating voltage or value given in runplan.
Verify that the currents being drawn are as expected. Expected values can be found on the Standard Operating Values page.
If current is too high, the Current Draw window will be yellow or red. Make a note in the logbook.
When not collecting data, put the tubes to standby (0%). Tubes that are not being regularly used should be put into a LOCKED state when they are ramped off so they cannot be accidently changed. D1I goes to a paused state when ramped to 0. Turn channel off then on to clear this state and the attendant minor alarm.
If the current limits are too low, make a note of it and follow the epics spy instructions below to temporarily increase the limits.
See the HV Control Document for details.

Scalar Initialization

Rate program should automatically start whenever d0olctl38 is rebooted.
If you want to spy on the program:
- Put an xterm from d0ol149 in the upper right corner and at the prompt in the ~d0fpd directory type:
- telnet t-d0-mch1
  login: ioc
  psswd: 0nlined0 (both are zeroes not o)
  con 18
  hit return twice (if you don't get a prompt, d0olctl38 might need to be rebooted, it is below the camac crate in the VME crate. If Fail is red, press the reboot button. In the telnet window you should now see the reboot dialog).
- You should see information from the program scrolling past.
POSSIBLE PROBLEMS / FIXES
- If nothing is updating, there is a possible trigger/electronics error. Verify in the MCH1 porch that the hardware scalars are updating. Check that the trigger generator is operating. Try to follow the signals through the chain to determine where the problem arises. If it is a bad module that needs to be replaced, a temporary fix is to pass the 1MHz signal directly to the CAMAC crate from some point after it has been split and sent to the hardware scalers.
- If you want to restart the program, type ^X in this window to reboot d0olctl38. If rates are stale because of no beam, this is the only way to reset them.
- If you need to exit the program in d0olctl38:
  - Type taskDelete trtdaq
  - Note: If these processes are already stopped you will be told that they are undefined.
- Reboot d0olctl38 to restart the program.

Rate Watcher Program

Verify Software is Running:

This is a cron job. If it is running, a window should be visibile on d0ol49. If you don't see this window, login to d0ol57 (d0ssh -l d0fpd d0ol57), do a ps -efw | grep rate_watcher, find the rate_watcher id and do a kill -9 (ratewatcher id). It will then restart itself.
In addition, to verify that it has not been disabled in cron, do a:
d0ssh -l d0fpd d0ol57
crontab -l
and check that the line containing fpd_rate_watcher.py has not been commented out with a #
To double check that it is currently running type:
ps -efmwwwH | grep rate
and five threads should be running.

To Stop the Software:

To stop it from restarting, log in to d0ol57 as above, do a crontab -e and put a # in front of the line with fpd_rate_watcher.py and then Ctrl-X, Ctrl-C. You should see a note that the crontab is being reloaded. You can then kill the program as above without it restarting. IMPORTANT: Be sure to re-enable the cron job when you are finished!

Functionality:

If the rates get too high, an alarm window will open. A yellow window is a warning set at 350kHz. A red window is a severe warning set at 420kHz and the HV is automatically turned off. The red window will open a window that remains until acknowledged.
If rates exceed 500kHz, a red window is issued and if the control lines are on, stop commands are sent to all pots, a verbal warning is issued telling people to turn off the control lines.
If the LVDT value exceeds the software limit, similar warning to the 500kHz are issued.
If the LVDT is moving in the the direction opposite to the desired direction, similar warming to the 500kHz are issued
To see the log files, look in /online/log/fpd_rate_watcher directory
To see the source code look in cvs at onl_fpd/rate_watcher
If the program is not running, contact Vladimir Sirotenko.
If this program generates an alarm in the SES window, DO NOT ACKNOWLEDGE that alarm. Use the orange acknowlege button that it generates instead. If you acknowledge in the SES window, that bypasses the usefulness of this alarm.

SES Alarm Display

From a prompt type:
setup d0online
start_alarm_display -c /online/config/fpd/ses/alarm_display.config &
This opens up a window that monitors alarms for all FPD Epics values.
Select View->Smaller 5 times to make the window smaller.
This window shows when EPICS values move out of desired range.
You can select an alarm and they click on the Guidance button to get a window with instructions what will hopefully help you diagnose and clear the alarm.
See Known Issues Page for known problems.

Rack Environment Monitor Display

From a prompt type:
/online/config/ctl/pfm.rmi &
This opens up a window that monitors RM alarms. Choose the tunnel tab. Click on reconnect if anything is blank.
Attempt to reset the appropriate alarm.

ACNET Setup

We need to open 4 windows in ACNET in the configuration shown in map on checklist.
Initial Setup (This lists the locations of the windows in the screen when setting up. For detailed instructions of how to setup each screen, see the details below.):
- Setup ICON BOX
  - In first row put the icons: PA, PB, PC (TV icon); GxSA, GxSB, GxXC (line plot icon); Utilities (telephone icon). The move the icons, make the icon box big. Click and drag an icon to move it.
  - Reduce borders of box until only the icons above are showing
- Move ICON BOX to lower right corner
- Double Click on Utilities in Icon Box
  - Click on FTP (Fast Time Plot) in new window that opens
  - Setup Halo variables (C:D0PHTL, C:D0AHTL, C:B0PLOS, C:B0ALOS) and click in the black area to open plot. Set the upper limits for each variable so that the D0 lines are towards the top and the CDF lines are towards the bottom
- Move plot to upper left hand corner
- Double click on the PA icon in Icon box.
  - Type C72 in upper left hand corner of window that appears
  - Reduce borders so window is as small as possible.
  - See details below on how to navigate C72
- Move to lower right hand corner.
- Double click on PB
  - Type D44 in upper left hand corner
  - From options in bottom middle of screen (3rd column), select Recall then D0 FPD. This will setup the desired variables. If D0 FPD is not present, you will need to setup the variables by hand. See the Lumberjack instructions below. When they are setup, select Save and give the name D0 FPD so that the setup is saved again.
- Move to lower left hand corner
- With D44 variables setup and time limits set (T1 and T2) click on any variable name to open the Lumberjack Plot
- Move plot to upper right hand corner
C72:
- This is the FPD window and contains all of our ACNET variables. In the upper left hand corner (line 4), you will see a line with -<1>+. This allows you to move through the subpages. The number between the <> is the current page. Click on + to move up in page number. Click on - to move down in page number. Page 1 shows the home bits. This is 1 if at home, 0 if inserted. Page 2 shows our LVDT values. Page 4 shows our singles rates (if L0 voltage is on and DAQ is running) and halo rates.
D44 (Lumberjack):
- The variables of interest are: C:D0PHTL (D0 Proton Halo), C:D0AHTL (D0 Anti-proton Halo), C:B0PLOS (CDF Proton Halo), C:B0ALOS (CDF Anti-Proton Halo) and then the variables for up to the pots we are moving. They are of the form: C:FP(potname)L where potname is A1U, A2U ... P1D, D1I, D2I etc. The limits for these variables are -10 to 40 for the pots and whatever is appropriate for the halo rates. If you put s at the end of the variable name rather than l, you will plot scaler rates. Set the upper limit in this case to 300000.
  Click on Units in the upper right hand corner and verify that all are Eng-U.
  The variables are broken into two lines, 4 per line. The top spot (Y=) is for the variable name. The next two slots are for the minimum (I=) and maximim (F=) values to plot. The final line is for the logger to be used. Use .dzero unless it is a cdf variable, then use .cdf. Set the start and end times that you want (T1= and T2=) on the line below the last set of variables. Click on any variable name to update the plot.
  Below the time, in the first column, select Trace and make sure all variables are Linear / Yes. Select Symbol and make sure all variables are point.
  In the second column, set X Divs and Y Divs to 5 each.
FTP (Fast Time Plot):
- Setup the halo parameters the same as they are setup in the Lumberjack D44 screen.
Another window of possible intersest is S4<3> which contains the LVDT readings for the CDF Tokyo pots in the middle of the page. Values near 0 mean pots are inserted. Values near -900 mean pots are retracted. Move around in this page in the same way as C72. Type S4 in the upper left hand corner of a window to get to this page.
It is also possible to access ACNET information from any web browser that support javascript. Please click here for instruction on how to do this.

Saving Plots to the web:

Setup Lumberjack plot the way you want.
In the D44 window press on Pgm_tools in upper right corner.
Press Screen Copy.
Press Copies.
Click in the Copy From field and select the window that contains your lumberjack plot
Select Save to GIF
Give the plot a title like "d0fpd_potinsertion" or something.
For directory choose USERB:[PIC.CAPTURE.DZERO]
Click on copy to save the plot.
In netscape go to cns40.fnal.gov/userb/pic/capture/dzero.
Scroll down until you find your title. It will have a date and time appended to the end. Click on it to bring up the picture and get the full URL. You can use ctrl-F in the browser to do a search by the date the file was created. This is usually the easiest way.
If you don't see your picture, go to cns40.fnal.gov/userb/pic/capture/ in netscape and hunt through the directories for you file
To attach the picture to the elogbook, add a text entry, select Insert Image at the top of the window and then type in the URL for the picture.

In case of problems:

If the LVDT or singles values are not being updated in the C72 window, we cannot move the pots until the problem is fixed.
You can look in C72 to see if the values are being sent to ACNET. Even at home, the LVDT values should be fluctuating. Look at Channel 13 on the hermes display (black screen you can access with camera controls) or the beams division page in Netscape. If D0 luminosity is updating there, then we are sending to ACNET properly and the problem is with Beams Division.
We can also get useful information from the lmExpert program. To start this, see the instructions at lmExpert below.
If the console isn't working after verifying that the D0 frontend is fine, call James Smedinghoff (x4509) for help. We currently connect to console CNS47.

ACNET through JAVA

Open a netscape window with the address http://adcon.fnal.gov/userb/www/javaconsole/autogen_javaconsole.html
You can access any of the ACNET screens through this interface. When the page first opens, you should have an ACNET terminal. Type the name of the page you want in the upper left hand corner.
Windows from this applet are time limited. If you leave it untouched for more than 30s it will close.
Sometimes this form of accessing ACNET will not work as it is a shared resource and if too many people are using it, it is unable to connect.

Epics Spy Program

From a prompt type:
setup d0online
PP.py /online/config/fpd/epics/lvps_variables.pp
This opens up a window that monitors Epics values for Low Voltage Power Supply in the tunnel providing power to the A/S.
Other paramater files can be found in the same directory if you need to look at other epics variables.
You can also use PP_SES.py . to look at the current epics value and do a soft change of the values. These values would be replaced by those in the database at the next reboot of the node. Look in the /online/config/fpd/epics directory for *.pp files with examples of variable names.
To change a high voltage alarm or trip limit you could use for instance FPD_HVC_MAA1I/VOLT or /CURR with .HIGH, .HIHI, .LOW, .LOLO at the end or just FPD_HVC_MAA1I.MAXC and change the value.

haloview

From a prompt type:
./python_mux/FpdHaloViewer.py &
This opens up a window that integrates over 10 readings of the 4 pertinent halos to give a smoothed value. When the 10% level is reached it turns yellow. When the 20% level is reached it turns red. If the ceiling is reached it also turns red.
Use this to guide you in pot insertion. You can go a little bit over the limits (except for ceiling).

Camara Setup

Select the desired camera locations and zoom the camera into the appropriate motor at the beginning of the shift.
Zoom the cameras back out at the end of the shift.
Close the iris for all but the dipole cameras.
If you don't know how to do this or there are any problems, please see the Camera Control Document.

Trigger Setup

A single elastic trigger from the standalone system is currently available to the latest trigger list. If there are any problems with this trigger. Check with experts (Duncan Brown).
Integrated triggers will be handled by crate 0x12 which contains the vertex boards and TDC boards.
If there is a problem with crate 0x12, have the Muon shifter try to clear it. If this doesn't work, remove it from the run for now.

AFE Setup

Check with CFT shifter that the electronics are enabled if you don't see expected plots in examine. The electronics needed for the Dipole AFEs include:
- Crate 0x52
- VRB 0x13 (slot 19)
- This will showup as VRB5219 on the CFT display)
- AFEs EA0, EA1, EA4, EA5, EA6
- HDIs properly masked
Instructions on how to access this information yourself will follow later.

Examine Setup

Examine Setup:

Open a telnet window
d0ssh -l d0fpd d0ol61
setup d0online
cd ft_examine_test
./ft_examine_test.sh anything
init
start
If working, things should now be scrolling in this window
To reset the plot, in the examine window with the scrolling text hit enter a few times. You will see no such command, then type report examine then type start. This should reset the histrograms.
To exit the examine process, type stop. Wait for a few minutes while it processes. Then type quit. If the process doesn't dump you to a prompt, type ^C to get a prompt and verify with a ps command that no ft_examine_test processes are running. If they are, kill them.

Browser Setup:

Open a second telnet window
d0ssh -l d0fpd d0ol61
setup d0online
cd ft_examine_test
./ft_browser_test.sh anything
If you click on cycle, it will go through all of the plots

Available Plots:

There are 18 correlation histos (1 per detector). The y axis is for primed layer fibers and x axis for unprimed layer fibers. We want to see correlations. In this way we have:
- at x plane
  from 0 to 19 : U (20 fibers)
  from 20 to 35 : X (16 fibers)
  from 40 to 59 : V (20 fibers)
- at y plane
  from 0 to 19 : UP, XP and VP
- Good correlation occurs when there is a hit either at fiber number N at unprimed layer and at fiber number N, or N - 1 at primed layer.
- There is a root-tuple at /scratch/d0fpd/ft_examine_test/ft_examine_ref.root that you can look at as a reference.
Ignore discriminator occupancy for now. They are being used for depugging of DFE system.

Run Information:

To obtain a list of current run information, prescales etc, in an xterm type:
setup run_summary
runsum.py run#
The current runnumber is listed in the scrolling information of the examine itself.
See Known Issues Page for known problems.

LMB Control Program

From a prompt type:
cd ~d0fpd/python
./fpd_slp.py
LMB will not work unless we are recieving pulses from calorimiter. A particular NIM module will be blinking if we have the pulses. Follow the patch in the chart on the NIM rack to determine the module
Each castle has a voltage and delay. Type in a voltage up to 10V and an appropriate delay so the signal arrives in the ADC gate.
Hit download and read to send the changes to the LMB module.
When finished, return all values to 0 and hit download

Taking LMB data:

Switch the data taking trigger to LMB trigger (remove pin from hole B of the GRAND OR module).
No need for pedestal file since every channel fires with the trigger.
If timed in correctly, you should see a flat distribution in all frames except for dead and weak channels. The ADC distributions should be gaussian.

lmExpert Control Program

From a prompt type:
setup d0online
start_daq lmExpert
Click on Short Term Lum Graph if you want to follow halo values while inserting pots.
If values here are updating properly, then EPICS is getting updated properly and any problem is either on the ACNET side at D0 or with our d0olctl02 module.
The Accelerator Devices tab is useful to see currents in magnets etc as well as associated ACNET variable names

Resetting Trigger Hardware

There are detailed graphical instructions on how to bring up the trigger electronics located here (ps) or here (doc).
There is also a copy in the binder after the emergency tab.

Frequently Asked Questions

Why does lumberjack/c72 show a flat rate when HV is off but it suddenly changes when the HV is turned on?
The reason for this is because of the way the standalone daq used to send the rates to ACNET works. It is only able to update the rates when it is recieving something from the tubes. When the HV is turned off, the daq no longer receives new rates so it continues to report the last known value for the rate. As soon as the HV is turned on, it will start updating again. So basically, ignore the rates being sent to ACNET if the HV is turned off. If necessary, you can reboot the d0olctl38 node to reset the rates to a default value. You would need to do this if there was a messy abort for instance locking the rates at a high value.

Why are some channels LOCKED in the HVGui?
It is a good habit to LOCK channels that are not being regularly used. To use a channel that has been locked, just right click on it and choose unlock.

Why does the VETON counter go to a PAUSED state when it is ramped to standby?
This is a feature of how the HV is controlled. For some reason, the module controlling VETON and D1I reaches the standby voltage, but is not able to pass a certain software check so it goes into a PAUSED state. To clear this state (and the minor alarm that goes with it), turn off the channel then turn it back on so it is in the ON state.

If you have any questions or comments, please mail me at: strang@fnal.gov

This page last updated: 08-Mar-2005 01:01 PM (GMT -06:00)