This entry is the continuation of the cascade effects on the pg peak shape entry:

https://exp-astro.physik.uni-frankfurt.de:8080/E108/475

The main point is that having more cascade makes the impulse carried by one photon smaller --> recoil cone is
smaller --> pg peak gets more centered

When one wants to make an "all-inklusive" simulation the problem comes, that there are too many populated states
after CN* decay (many are just theoretical), also many cascades afterwards as well.
Instead, one can make a "statistical approach":
 1, take an average 1st populated state after the gamma emission of the CN*.
 2, take an average of the gamma energies after the p-capture. (it includes also the photons from the CM* to the
1st state after g emission with energy of [(E_CM - Q) - E_1st_state] --> should these be excluded from the avg.
gamma energy calculation?)
 3, the excited compound nucleus first decays to 1, then cascading down with the avg energy of 2,

    number_of_cascades = E_1st_state/E_avg_gamma. 

    For the 124Xe at 7AMeV using the TALYS code the above values are:
    E_1st_state = 6.74 MeV
    E_avg_gamma = 2.75 MeV
    number_of_cascades = 2.45

To reproduce the resulting avg. pg peak, I took a 2 and a 3 cascade simulations produced by MOCADI and mixed
them as 55% of the 2-cascade and 45% of the 3-cascade for this naiv, statistical model.
Then, to test the peakshape, I fit this mixed distribution to a part our measurement data at 7AMeV after a
sloppy Rutherford background removal (the background removal can be more improved, I used this as a quick method
for now). The calculated Chi^2/NDF ~ 3, which is not a super good value, but taking into account that this is a
strongly naiv model, and the background removal can be also improved, the chi^2/NDF value can be also
interpreted as promising.

I have made some simulations at 8AMeV, 7AMeV and 4AMeV energies for the main 118Te + p reaction channels. For the
Rutherford simulations credits to Yuanming! For each energy there are 3 simulations: without the scraper, with the
scraper (online), scraper (online) + E-truncation (offline). The scraper is treated as an "ideal scraper" meaning
that there is no scattering at the edges (maybe worth a GEANT4?). The slit position is at -3cm from the beam, the vertical
length is 7cm (centralized). The following channels are combined in the
simulation:

-8AMeV:
  Rutherford
  pg channel, 5cascade model. Photon emission is treated isotropically
  pn channel, including: -->gs, -->1-->gs, -->2-->gs, -->3-->gs, -->3-->1-->gs, -->4-->1-->gs, -->5-->3-->gs,
-->5-->3-->1-->gs, -->5-->4-->1-->gs decay chains with their weights. Neutron and photon emission is treated isotropically.

-7AMeV:
  Rutherford
  pg channel, 4cascade model. Photon emission is treated isotropically

-4AMeV:
  Rutherford
  pg channel, 3cascade model. Photon emission is treated isotropically


The detector position is -2.5cm from the beam in the radial direction and centered vertically with a 45° tilt. 
The cross section values for pg and pn, the pn channel mixing, and the pg cascade number is based on TALYS simulations.
The other two input parameters are the luminosity and the measurement time. For the simulation, a little
pessimistic (or realistic?) scenario is taken: L = 10^24 cm-2s-1, and t = 10^4 s. From this values, from the N=CS*L*t
equation, the pg counts (based on the TALYS
cross sections) are the following: 8AMeV ~267, 7AMeV ~163, 4AMeV ~5! This means that to reach the Gamow-window
energies is challenging, but with a well-working scraper it is not impossible to reach.


As it is visible, the 8AMeV 118Te(p,g) case is very similar to the 7AMeV 124Xe case. The p,n threshold is
somewhere ~7.5MeV.

The attached spectrum is for the 90 degree X-ray detector taken for about 3 hours with 118Te52+.
We see two peaks at ~29 keV (K-alpha) and 44 keV (K-REC).

The resolution is about 1.5 keV, and the signal-to-background ratio is also really bad.

The increase of background events at higher E might be induced by Compton scattering of a high E gamma ray from the BaF detectors, which are active and sit close by (not anymore).

Here are the data file for the rates of the recoil detector for different values of the target bump. The average rate over background is proportional to the target overlap. 

The MCP was operated at a voltage of 2500 V, where it sees mostly photons, so not really recoil ions.

Column "X" is time in seconds, Column "Content" is rate in Hz.

With scraper, we have roughly 150-200 (p,g) counts now.

X-rays look fine!

Without scraper, we have very few (p,g) counts, hard to get numbers quickly, maybe ~50.

X-rays look fine!

Here is the analysis of the test runs with the 2nd DSSD (gen2) using the quick_si_plots.c script attached.

All 32 Si channels are working with acceptable performance. 

In run001, there are some additional low energy peaks in nearly all x-strips, which I do not understand yet. They are around 3 MeV and are not visible in the y-strips. It doesn't look like an electronic problem, because there are at least 4 
peaks, so not a low amplitude copy of the 3 major alpha peaks between 5 - 5.8 MeV.

However, in run002, the peaks have mostly disappeared, only in x6,x7,x8,x9 is a broad structure at somewhat similar energy... maybe this has to do with the small incident angle of the alphas?

Another run to confirm and double check this would be nice.

//a simple code (template) for offline analysis
//made by Laszlo, serves as a simple demonstration for enthusiastic shifters
//it creates a "no double counting Si 2D pos" histo
//usage:
//
//save the file as eg. "simple_code.c"
//root -l
//.L simple_code.c++
//run()
//when counter finished: ".q"

#define INPUT1 "input.root"//first unpack the lmd, then give the path of the unpacked .root file.

#define OUTPUT "./"//folder of the output. minimum input: "./"

#define ROOT_NAME "dummy.root"//name of the output

#include <cmath>
#include <string>
#include <sstream>
#include <cstdlib>
#include <cstdio>
#include <ctime>
#include <fstream>
#include <iostream>
#include <stdint.h>
#include "TROOT.h"
#include "TAttText.h"
#include "TAxis.h"
#include "TCanvas.h"
#include "TChain.h"
#include "TCut.h"
#include "TF1.h"
#include "TFile.h"
#include "TGraph.h"
#include "TGraphAsymmErrors.h"
#include "TGraphErrors.h"
#include "TH1.h"
#include "TH2.h"
#include "THistPainter.h"
#include "TKey.h"
#include "TLatex.h"
#include "TLegend.h"
#include "TMath.h"
#include "TMatrixD.h"
#include "TMinuit.h"
#include "TMultiGraph.h"
#include "TNtuple.h"
#include "TPave.h"
#include "TPaveText.h"
#include "TPoint.h"
#include "TRandom.h"
#include "TRint.h"
#include "TStyle.h"
#include "TString.h"
#include "TTree.h"
#include "TH1F.h"
#include "TH2F.h"
#include "TSystem.h"
#include "TProfile.h"
#include "TVirtualFitter.h"
#include "TCanvas.h"
#include "TLegend.h"
#include "TColor.h"
#include <time.h>

using namespace std;

inline bool exists_test0 (const std::string& name) {
    ifstream f(name.c_str());
    return f.good();
}

///////////////////////////////////////////////////////////////////////////////////////////////

void loop(TChain *fChain){
	
	//setting pedestal values
	int PEDESTAL_LOW=400;
	int PEDESTAL_HIGH=8000;
	

	//this normally should be in a separate header, branches are defined.
	UInt_t          trigger;
	fChain->SetBranchAddress("TRIGGER",&trigger);
	UInt_t          E_SiY[17];
	UInt_t          E_SiX[17];
	UInt_t          t_SiY[17];
	UInt_t          t_SiX[17];
	for(int a=1;a<17;a++){
		fChain->SetBranchAddress(Form("E_SiY%d",a),&E_SiY[a]);
		fChain->SetBranchAddress(Form("E_SiX%d",a),&E_SiX[a]);
		fChain->SetBranchAddress(Form("t_SiY%d",a),&t_SiY[a]);
		fChain->SetBranchAddress(Form("t_SiX%d",a),&t_SiX[a]);
	}

	//creating histos
	TH2D *h_pos_si_xy=new TH2D("h_pos_si_xy", "h_pos_si_xy",16,0.5,16.5,16,0.5,16.5);

	//creating and initializing some variables used in the event loop (for "no double counting")
	int r_pos_x=0,r_pos_y=0;
	int dc_Ex_max=-999;
	int dc_Ey_max=-999;
	
	
	Long64_t nentries = fChain->GetEntries();
	Long64_t nbytes = 0;

	//starting the entry loop
	for (Long64_t i=0; i<nentries;i++){

		nbytes += fChain->GetEntry(i);
		//event countdown
		if ((float(i)/100000.)==int(i/100000)){cout << "event: " << i << " \tof " << nentries << endl;}


		if(trigger==1){//trigger 1 = TargetON

			dc_Ex_max=-999;
			dc_Ey_max=-999;

			for(int i_x=1;i_x<17;i_x++){
				for(int i_y=1;i_y<17;i_y++){

					if( ((int)t_SiX[i_x])>0 && ((int)t_SiY[i_y])>0){
						if( PEDESTAL_LOW<((int)E_SiX[i_x]) && PEDESTAL_HIGH>((int)E_SiX[i_x]) &&
						    PEDESTAL_LOW<((int)E_SiY[i_y]) && PEDESTAL_HIGH>((int)E_SiY[i_y])){

							//assign the hit to StripX and StripY where the most energy is deposited (rel. Ecalibration is needed, but roughly ok)
							if(dc_Ex_max<((int)E_SiX[i_x]) && dc_Ey_max<((int)E_SiY[i_y])){
								r_pos_x=i_x;
								r_pos_y=i_y;
								dc_Ex_max=E_SiX[i_x];
								dc_Ey_max=E_SiY[i_y];
							}
						
						}
					}
					
				}
			}

			//Filling pos. histo
			if(dc_Ex_max!=-999 && dc_Ey_max!=-999){h_pos_si_xy	-> Fill(r_pos_x,r_pos_y);}
			
		}//trigger==1

	}//entry loop
	

	//writing out the root output file
	TFile *graphfile = TFile::Open((OUTPUT + (string)("") + ROOT_NAME).c_str(), "RECREATE");
	graphfile -> mkdir("map");
	graphfile -> cd("map");
	h_pos_si_xy -> Write();
	graphfile -> Close();

}//loop

//////////////////////////////////////////////////////////////////////////////////////////////////////////////

void run(){

    const char *command = new char[1000];

    char filename[100];
    TChain *fChain = new TChain("h101");

    sprintf(filename,INPUT1);

    if(exists_test0(INPUT1) && exists_test0(OUTPUT)){

    cout<<"\033[0;37m//loading run: "<<filename << "\033[0m" <<endl;
    fChain->Add(filename);

    loop(fChain);
    }
    else{cout<<"\033[0;31mError 404: non-existing INPUT1 or OUTPUT file!\033[0m" <<endl;}

	command = "rm  *.so";
	gSystem->Exec(command);
	command = "rm  *.d";
	gSystem->Exec(command);

	return;
}

int main(){
    run();
    return(0);
}

The two attached plots show the collected Si data up to run0115.

Within this ~15h we have about ~70 counts in the peak region.

This translates to ~4.5 counts per hour.
And shall be compared to the 1 count/hour we had last year.

Attached is a plot of the uncalibrated energy spectra taken with the 90 degree xray detector for 118Te (red) and 124Xe (blue).
Shown is the region covering the K-alpha and K-REC peaks.

Calculating the ratios (Te/Xe) for K_alpha and K-REC energies from theory one gets:

K_alpha (theory): 0.9236  
K-REC   (theory): 0.9308

Without energy calibration we get the following ratio (using channel numbers):

K_alpha (exp): 0.9293  
K-REC   (exp): 0.9344

Taking into account that the channel numbers are not calibrated, the agreement is still very good.

In summary we can be sure that we have 118Te stored at 7AMeV.

This is the summary of all Si data collected at 6 MeV/u until run174 before the Alvarez problem.
We have roughly 50-60 counts in the peak region.
This data corresponds roughly to 36hours of beam, which means ~1.5 to 2 counts/hour.

measured beam lifetime after the second deceleration using parallel plate Schottky detector and the RSA5103 spectrum analyser. Analysis code can be found here:

https://github.com/xaratustrah/e127_scripts/

The plots here were produced using the script:

https://nbviewer.ipython.org/github/xaratustrah/e127_scripts/blob/main/lifetime_calculator.ipynb


UPDATE 2022-06-30:

Life times have been slightly corrected. Please see attachment.


Filename	Start Frame	Correct Lifetime [s]	Mean [s]
pgamma_lifetime-2021.05.25.21.56.37.593_on.tiq	29500	1.43	
pgamma_lifetime-2021.05.25.22.02.16.618_on.tiq	29500	1.55	1.49
pgamma_lifetime-2021.05.25.22.07.22.369_off.tiq	53500	1.75	
pgamma_lifetime-2021.05.25.22.13.24.787_off.tiq	34900	1.89	1.82

The target overlap was optimized on xray and si rates.
The setting now is -10mm in SC9

Logbook from start of FRS until 118Te injection

The full cycle length of ESR with 30 stacks is about 7:45 minutes, out of which 15.5 sec are with target switched ON.
This is without CMB/HTD in parallel.