The beam requirements of charmonium experiment E-835 are particularly demanding. For this experiment, beam intensity is not the only relevant parameter; the beam energy distribution is also of great importance for the measurement of resonance parameters. Our main goal is to deliver intense and stable beam with constant energy and small momentum spread; also, we need to make sure E-835 is receiving the relevant ACNET data.
This note is for the P-bar deceleration shift crew and for MCR operators and its aim is to help us in our beam monitoring tasks. If you find errors or omissions in this document, please contact me. For question regarding how to handle specific issues encountered during shifts, please page the on-call deceleration expert.
See also Steve Werkema's note.
Once you enter P69, choose LIFETIMES (instead of ACCUMULATION RATES). The name of the variable device should be E:GJ102, which is the measured jet-target density. A reasonable range for its values is 0 to 5E14 (it is in units of atoms per cubic centimeter). Interrupt on START PLOTTING. It takes a few minutes to get the first trace. This plot will crash if a glitch makes A:IBEAM readback equal to 0; check at least every hour to see if the plot is alive. Alternatively, you can choose the OTHER option (instead of ACCUMULATOR), and plot A:IBEAMV, which should be glitch-free. For the transverse emittances to make sense, A:EMITHS and A:EMITVS have to be set to an integer number plus the fractional part of the tune in that plane (qx=6.695; qv=8.685); for instance, a good choice is A:EMITHS=124.695 and A:EMITVS=125.315 or, simply, A:EMITHS=125.695 and A:EMITVS=124.685.
E-835 relies on the orbit-length measurement for the determination of the beam energy. The BPMs have to be started before running the on-line energy calculation (P77). Enter P51. Make sure the DC gain is 30 dB. Choose difference mode and subtract file #9 from Archive. Start data collection. If the orbit difference is not within a few millimeters or the measurements fluctuate by more that about 0.1 mm in the low-dispersion regions (A10, A30 and A50), display the intensities (this should be done in normal mode, and not in difference mode); they should be larger than 50 mV. Save a raw orbit (ie, normal mode) to the scratch directory at least once every shift.
This is a vital part of the experiment's data collection stream. The program reads all the relevant accelerator parameters, performs some on-line calculations and sends the data to the E-835 data-acquisition system. This is the application that updates A:ECM, A:SIGMAP and A:CENFRQ, for example.
Start P77. After the initialization is complete, read in the ramp file (file #1 for X:POFTT > 6367, file #2 if X:POFTT is smaller than 6367). From the RF CONTROL menu, choose which cavity is being used to tickle the beam for BPM measurements; the corresponding information (frequency and amplitude) is used to calculate A:ECM and put into the data stream. The default BPM and spectrum-analyzer parameters should be ok. Set "SEND TO E835" to YES and "CONTINUOUS MODE" to ON; the application will respond to interrupts only between DAQ cycles, which last approximately 1 minute.
Periodically, at least every hour, interrupt on DUMP TO SCREEN and examine the integrity of the data. Sometimes P77 gets stuck while performing a BPM measurement; this means that E-835 is not getting data. This problem is usually solved just by stopping and re-starting the data collection on P51 or by resetting the BPM nodes from the P51 diagnostic subwindow. When the gas-jet devices A:GJTRB, A:GJPRB, A:GJ102, A:GJ103 and A:GJ100 are in error (P77 will issue a red message and display their values as -666), ask the experimenters at AP-50 to restart the gas-jet Mac (they have a procedure for doing this even when the jet is running); when they are done, restart P77, following the above instructions. This will usually fix the problem.
Application P85 performs four tasks:
In the frequency parameters menu you can choose the period with which P85 checks the BPMs and corrects the RF frequency; set it to a number between 20 and 50 s. The minimum and maximum frequency corrections should be 0.1 and 0.05 Hz. The frequency range should be about ±5 Hz centered around the present value of A:CENFRQ. The maximum RF volts should be between 15 and 30 V. The babysitter program will display a red error message and stop making any RF corrections if the frequency or voltage are outside these ranges. Make sure the right frequency synthesizer is selected: A:RLLFS0 for ARF-3 and A:RLLFS1 for ARF-2. Before you turn on this feature, make sure P77 is updating A:CENFRQ with reliable measurements of the average center frequency of the beam, which come from the narrow-span spectrum (you can see it on p-bar channel 21, alternating with the wide-span spectrum; the narrow-span spectrum usually has a span smaller than 10 kHz); the measurement will be biased, for instance, if the spectrum is truncated or if there are coherent spikes.
NOTE: For stacks larger than 20 mA, we noticed that adjusting the RF frequency often can disturb the longitudinal stochastic cooling system. Ask one of the experts if the frequency-tracking feature should be turned on and how.
In the P77 heartbeat menu, you can set how often P85 checks if P77 is active. Obviously, this number has to be larger than the P77 DAQ cycle; you can estimate the P77 period by looking on a fast time plot how often A:ECM is updated (the expected number is usually about 200 s). On the other hand, you do not want P85 to check P77 too seldom; so, a good range for the heartbeat interval is between 300 and 600 s.
The Ecm tracker will only work if the cooling has enough power to grab the beam and move it; otherwise the 4-8 core momentum plates will be moved until they are not seeing the beam any more, and the cooling will be useless. Turn on this feature only if you will be closely watching the position of the cooling plates A:MARAYD, the cooling powers A:CMTW01 and A:CMTW02, the signal from the plates (p-bar channel 20), the energy spectrum (p-bar channel 21), the CM energy A:ECM and the momentum spread A:SIGMAP.
The Ecm interval should be larger than the P77 period, explained in the above heartbeat paragraph. The feedback time constant should at least be 2 to 4 times the Ecm interval, to allow filtering of fluctuations in A:ECM; currently, we are using 340 s. The sigma-p limit is a way for P85 to check if the cooling is strong enough (of course, this is not the only indicator!); it will not take any Ecm correction action if A:SIGMAP is larger than this limit. Also, no action will be taken if A:MARAYD is outside the specified pick-up lower and upper limits; you can set these limits to -10 mm and 2 mm. The feedback gain indicates what fraction of the measured Ecm deviation (A:ECM minus the desired Ecm) P85 will try to correct at each iteration; a value of 0.1 is what we have been using lately.
There are at least three sets of plots we want to constantly monitor; they are saved in the p-bar directory as number 1, 4 and 9.
The first one shows the center-of-mass energy and the beam width calculated by P77; it also displays the difference between RF frequency and mean beam revolution frequency (make sure you are looking at the right RF frequency: A:RLLFS0 for ARF-3 and A:RLLFS1 for ARF-2). Set the A:ECM scale so that the target Ecm is in the middle and the tolerance requested by the experiment (usually 0.1 MeV at the 1P1 and 0.5 MeV at the chi-0) corresponds to one of the four vertical boxes.
Cooling powers and the position of the 4-8 momentum pickup are shown in plot #4. If the cooling powers A:CMTW01, A:CHTW01 and A:CVTW01 show large spikes, there is a problem. A:MARAYD should be between -10 mm and 2 mm; if the Ecm tracker is on, it should move the plate smoothly (on the order of tens of stepping-motor steps at a time).
Plot #9 shows A:IBEAM, the measured jet-target density E:GJ102, the instantaneous luminosity E:GJ103 and the number of counts in one of the E-835 detectors (A:76SCD0).
Two particularly useful TV displays are P-Bar CATV channels 20 and 21. Channel 20 is usually set to the difference signal from the 4-8 momentum pickup plates and shows a notch if the beam is centered underneath the pickups. Channel 21 is the spectrum analyzer piloted by P77 for the measurement of the beam energy spread. It should toggle between a wide-span display, spanning the whole Accumulator momentum aperture, and a narrow-span display, zooming on the peak. Both spectra are recorded in the E-835 data stream, but it is the narrow-span one that determines A:ECM, A:SIGMAP and A:CENFRQ. Therefore, it is very important that the main peak is well contained within the display; if not contact me on how to change the spectrum analyzer settings from P77.
A summary of the cooling parameters can be found in P36, CORE_M_&_B, subpage 19. The cooling powers are controlled by the attenuators A:CHAH05, A:CVAH05 and A:CMPA01. E-835 experimenters are supposed to call the MCR every time the jet density is changed significantly. Normally, the jet runs in constant-luminosity mode, gradually increasing the density as the beam current decreases. Occasionally, they might need to adjust it manually or to turn it off. If the jet is turned off (besides getting a phone call, you should see A:GJ102 and A:76SCD0 drop), turn down the momentum cooling in order to keep A:ECM and A:SIGMAP constant. You can either turn it down, 3 dB at a time, or turn it off with the PIN switch A:CMPS01. Watch A:ECM and A:SIGMAP and adjust the momentum cooling as the target is turned back on.