Target_Studies

Target Studies

Figure 1 - Run 34912, systematic current change, LUMIs 5-8.

Figure 2 - Run 34912, systematic current change, LUMIs 1-3.

Figure 3 - Run 34923 LUMIs 5-8 with strange multiple peak structure.

Figure 4 - Run 34921 LUMIs 1-3 with strange multiple peak structure.

RATIO BOX QUESTIONS

Exploring the strange behavior of the ratio box signal. For some of the runs the raw signal is being sent to the "0" bin, yet there appears to be an asymmetry. The lowest currents have this behavior, and except for the 20 &mu A setting at 30 Hz, it seems like the channel number is scaling with current down to zero at aroun 5 &mu A for 30 Hz and 15 &mu A for 250Hz. So I have a couple of questions:

1) Does it make sense for the ratio box to be outputting a zero channel? Mark said he thought that it would have some minimum other than zero.

2) Assuming we can figure out the answer to 1), why is there an asymmetry when it appears that all of the raw signals are in the "0" bin?

Figure 1 - Example of a ratio box "raw" signal in "0" bin with reasonable? asymmetry.

Figure 2 - Channel number vs current for ratio box.

With this rough calibration I get the following pedestals (compared to "beam off" values).

For 30Hz:

Monitor Pedestal Beam off

LUMI5 6844 6813

LUMI6 7397 7327

LUMI7 6994 6962

LUMI8 6713 6649

BCM1 8393 8352

BCM2 8359 8350

For 250Hz:

Monitor Pedestal Beam off

LUMI5 804 802

LUMI6 867 863

LUMI7 822 820

LUMI8 789 785

BCM1 984 982

BCM2 982 982

Figure 2 - LUMI and BCM channels vs. "set" current for 30Hz.

Figure 2 - LUMI and BCM channels vs. "set" current for 250Hz.

The LUMI (and BCM) pedestals may make a difference in the asymmetry widths. The 30Hz pedestals are pretty similar to the ones used for all the plots below, but the 250Hz are way off. It would probably be good to add the necessary calibrations to the database, reanalyze the runs, and re-do all the plots to get the new widths.

In Database:

Monitor Pedestal

LUMI5 6966.26

LUMI6 7080.87

LUMI7 6474.33

LUMI8 6599.81

BCM1 8343.95

BCM2 8341.77

30 Hz, fan 8 Hz

1.2x1.2 raster

LH₂

30 &mu A

25 &mu A

20 &mu A

15 &mu A

10 &mu A

5 &mu A

250 Hz, fan 8 Hz

1.2x1.2 raster

LH₂

30 &mu A

25 &mu A

20 &mu A

15 &mu A

10 &mu A

5 &mu A

30 Hz, fan 8 Hz

1.2x1.2 raster

C

25 &mu A

15 &mu A

5 &mu A

250 Hz, fan 8 Hz

1.2x1.2 raster

C

25 &mu A

15 &mu A

5 &mu A

30 Hz, fan 40 Hz

1.8x1.8 raster

LH₂

25 &mu A

15 &mu A

10 &mu A

5 &mu A

250 Hz, fan 40 Hz

1.8x1.8 raster

LH₂

25 &mu A

10 &mu A

5 &mu A

Plots of phi, log scale
Plots of phi

colored plots of all that get through 1st
dot plots of all that get through 1st

colored plots of all that hit bar
dot plots of all that hit bar

Figure 1 - Plot of what hits the cerenkov bar at z=-637.91.

Figure 2 - Plot of what gets through first collimator at z=-637.91.

Figure 3 - Plot of what hits the cerenkov bar at z=-637.91, colors.

Figure 4 - Plot of what gets through first collimator at z=-631.91, colors.

Figure 5 - phi plot of what hits the cerenkov bar at z=-631.91, 3.5

Figure 6 - phi plot of what gets through first collimator at z=-631.91, 3.5

The cell containing the hydrogen in all "standard" simulations is a 2in OD tube of aluminum. The tube wall is 10 mils thick and the end-windows are 3.5 mils. The hydrogen itself is 35 cm long, and 1in - 10 mils in radius. There is a mysteriously small effect due to the target attachments shown in Figures 1-3, however, in each of these cases only a small fraction of the particles that hit the GEMs or Region II chambers (and to an even smaller extent the cerenkov bars) have interacted in the attachment.

These pictures show the geometry of the designs used in the simulations. The sleeve is solid aluminum, but versions 604 and 605 have tubes filled with hydrogen. In all cases the design was simplified; verison 604 does not have holes along the length of the target cell, for example.

Figure 1 - Target with 6in Al sleeve.

Figure 2 - Version 604 of the transverse target design.

Figure 3 - Version 605 of the transverse target design.

These tables show the rates of various process at the various detector locations. The comparison is between the "original" design, the largest sleeve overlap recommended (6 inches), and the new transverse designs.

In addition, new simulations with 3inch od with and without Aluminum sleeve were done. The effect of increasing the target cell OD is shown in Figure 8.

The photon rates at the cerenkov bar were not calculated, because they are notorious for being inaccurate so far, but should probably be done carefully at some point.

These rates were calculated for the octant at 12 o'clock, regardless of the design. Hopefully tomographic plots will be able to show if there is any new backgrounds coming from adjacent octants.

Version
ep Rate
(kHz/nA)
< Q² >
((GeV/c)²)
< &theta >
(degrees)
Inelastic Rate
(kHz/nA)

"original"
4.85
.0257
7.891
.0025

Al sleeve 6in overlap
4.90
.0257
7.891
.0018

"original" with 3in OD
4.96
.0257
7.882
.0022

Al sleeve 9in overlap with 3in OD
4.88
.0257
7.885
.0020

transverse flow 604
4.93
.0257
7.889
.0017

transverse flow 605
4.94
.0257
7.889
.0021

approximate errors
.4%
---
---
10%

Table 1 - Comparing elastic electron information at the cerenkov bar.

Version
GEM moller Rate
(kHz/nA)
GEM photon Rate
(kHz/nA)
GEM inelastic Rate
(kHz/nA)
Region II moller Rate
(kHz/nA)
Region II photon Rate
(kHz/nA)
Region II inelastic Rate
(kHz/nA)

"original"
6300
6793
4.5
2443
1564
2.0

Al sleeve 6in overlap
6133
9656
4.6
2388
1553
2.0

"original" with 3in OD
6444
6545
4.6
2497
1250
2.0

Al sleeve 9in overlap with 3in OD
6265
5823
4.7
2451
3037
2.0

transverse flow 604
6985
6033
4.7
2610
1326
2.0

transverse flow 605
6675
5961
4.7
2617
1299
2.0

approximate errors
.3%
3%
.4%
.5%
5%
.5%

Table 2 - Rates for various processes at the GEM and Region II locations.

These pictures show the origin of photons hitting GEMs in the upper (12 o'clock) octant in the simulation. Very few events hit the attachments and even fewer create backgrounds that actually make it into the acceptance. By far most of these don't make it to the region II chamber and none of them make it to the cerenkov bar (in the limit of the statistics).

Figure 4 - Plot of origin of photons hitting upper octant at GEMs with version 604.

Figure 5 - Plot of origin of photons hitting upper octant at GEMs with version 605.

These plots show the realtive photon rates for photons hitting the GEMs. The black plot is for stuff coming from the target or the tube around it and the red is for backgrounds coming from the attachments. In both versions it seems that the rates coming from the attachments are about 6 or 7 orders of magnitude smaller.

Figure 6 - Relative GEM photon rates (kHz/nA) from target
cell (black) and attachments (red), version 604.

Figure 7 - Relative GEM photon rates (kHz/nA) from target
cell (black) and attachments (red), version 605.

So it appears that the rate increases as the radius of the hydrogen increases - while for a given cell radius, it decreases as the thickness of the tube increases. I believe this is because as the radius of the cell increases, less and less of the particles see the tube, and more of them are coming through the end window. With an outer diameter of 3 inches, pretty much all of the scattered particles are exiting the target cell through the end window. These plots are not directly comparable because the one on the right had a different geometry than for the one on the left (which used a more recent geometry).

Figure 8 - Rates at cerenkov bars at different hydrogen cell diameters.

Figure 9 - Rates at cerenkov bars for various tube thicknesses.

Figure 10 - All octants for the GEMs with a 3inch OD target cell.

Figure 11 - All octants for Region II with a 3inch OD target cell.

Figure 11 - No picture yet.

Figure 12 - No picture yet.

Jefferson Lab | Virginia Tech Main Page | Jefferson Lab's Qweak Page | VT Physics

Page Maintained by Juliette Mammei Last Update: 7/19/2006

Email Juliette

This work is supported by the National Science Foundation.