Endcap HV Cable Reshuffling

Resistive Plate Chambers

The Resistive Plate Chambers (RPC) are gaseous detectors that can identify the passage of charged particles with good time resolution (from 1 ns to 50 ps) and spatial resolution (up to 30 μm ). They are suitable for experiments that need fast space-time particle detection.

An RPC consists of two parallel plates made with a high-pressure plastic laminate of a phenol resin: bakelite. The plates have a resistivity of $10^{10}- 10^{ 11} \Omega cm$, separated by a gas which occupies a few millimeters. The outer layers of bakelite are covered with graphite paint to form the electrodes of the chamber. One acts as the ground, and the other has a determined potential for the measurement of the ionization produced in the chamber. The output signal is obtained on aluminium strips that are separated by an insulating material (mylar or PET film).


Operation Modes

RPCs can operate in different modes. In "streamer mode" the RPC operates with a strong electric field that produces localized gas discharges in the region near the passage of the ionizing particle, however this mode allows few counts per unit area ($\sim 100 Hz/cm^{2} $). Another mode of operation is the "avalanche mode", where the electric field between the plates decreases (in comparison with streamer mode), and the output signal is amplified. The reduction in voltage reduces the charge generated in the ionization and increase the counting capacity of the chamber. The electrons produced by the ionisation due to the presence of charged particles in the gas are multiplied in avalanche mode by a uniform electric field of about 4.5 kV/mm.

RPC distribution in the CMS experiment

The barrel region of the CMS experiment is equipped with 480 RPC dual chambers, while the endcap region is equipped with 216 dual chambers in the positive endcap and 216 chambers in the negative endcap. In the RPC upgrade it has been installed additional 144 RPC chambers in the endcap region (72 in the positive and 72 in the negative) for a total of 1056 RPC chambers for the entire CMS experiment operating in avalanche mode. (See figure 2)


The RPC Efficiency

The CMS-Muon-System redundancy allows RPC efficiency studies using the readout of other muon-subdetectors (DT/CSC). Using Local Reconstruction algorithms (at chamber level) we can have tags and probes to estimate the RPC efficiency. In this case, the tag is a segment, of a muon track, extrapolated to the RPC surface under study and the probe is the matching RPC-readout with the extrapolation. The efficiency is measured by extrapolating the segments associated to a STA-μ (Stand Alone Muon). Once the efficiency is known, a fit with the following sigmoid function is done:

  \begin{displaymath} Eff= \frac{Eff_{max}}{1+e^{s(HV-HV_{50})}} \end{displaymath} (1)

The parameters of the fit are:

  • "Eff max" is the maximum efficiency, the asymptotic value that the efficiency is approaching when HV → ∞
  • "HV 50" is the HV applied such that the efficiency is Eff max /2
  • "s" is just a scale factor, determines how "horizontal" is the sigmoid.

Apart form the parameters of the fit, the following points of interest are defined:

  • The knee ≡ 0.95 Eff max
  • Working Point Barrel ≡ knee +100V
  • Working Point Endcap ≡ knee +150V
When these parameters are well identified we can say that a given eta-partition is characterized. Once the Working Point has been identified, one can get its HV value and define it as the optimal working point for the chamber under study. For the RPC system it is not possible to set different HV values for every eta-partition. It is necessary to do a "convolution" for the different eta-partition that are fed by the very same HV channel.

Typical High Voltage scan fit parameters

High Voltage scans were performed at the beginning of the 2011 and of the 2012 proton-proton LHC running, aiming at determining the optimal operating HV for each individual chamber. Thevariation of the environmental pressure P and the temperature T inside the CMS cavern was taken into account using the following formula: ...

The goals for the HV scan are:

  • To study the efficiency as a function of the High Voltage for all the RPC chambers.
  • To find the best working High Voltage for every chamber.
  • To assure high efficiency
  • To minimize pressure and temperature dependency
  • To avoid increasing cluster-size
The points were taken between 8.5kV to 9.7kV (points already corrected by pressure)



  • Reshuffle endcap HV cables (RE1, RE2, RE3) to group chambers with close working points.


  • Bring actual working point for every chamber close to ideal WP.


  • 432 chambers in total (endcap only).
  • 13 chambers were removed due to bad fit or bad behavior (can be done separately).
  • Number of channels < Number of chambers.
  • Distributors are 4 to 1, therefore two chambers will have the same voltage.

Distributors (Back)


Distributors (Front)




Physical Constraint

Split every group of interchangeable chambers (in red color) according to rack placement (54 each).



Find one working point per chamber. If ( maxroll < (minroll +100V)) Yes: maxroll No: (minroll + 100V)


Find one working point for every two chambers. Difference was not big take the greater WP.


Working Point Distributions per Chamber



Efficiency Comparison per roll


Efficiency Stability after and before reshuffling

Imagen41.png Efficiency as a function of (working point +100 Volts) minus efficiency as a function of (working point -100 Volts) by roll.

Code Explanation



Combineda.txt - This is a text file that contains all the data from the HV scan


  1. The irrelevant data is removed from the text file. Only rollnames and working points remain.
  2. The data in Combineda.txt is given per roll. Since we want to apply the same HV to the whole chamber, these values must be combined. This code performs this function :
wpc = (avg > minroll+0.1)? minroll+0.1 : maxroll

It averages the three values. If the average is greater than the minimum of the three values by 100 V or more, it takes the working point as min + 100V, else it takes the maxroll.


  1. Endcap Only2.txt - A list of all the endcap chambers and their working points.
  2. Barrel Only.txt - A list of all the barrel chambers and their working points.
  3. After Combined.txt - A list of all the chambers and their working points.
  4. Roll Diff.txt - A list of all the chambers with the difference of the maxroll - minroll



Endcap Only2.txt - From Step 1


This code sorts "Endcap Only2.txt" entries by name. This will make it easier to group the chambers that are interchangeable together.


Endcap Only Sort2.txt: This is a text file that has all the endcap chambers sorted by name.

-- AndresCabrera - 2014-12-17

Topic attachments
I Attachment History Action Size Date Who Comment
JPEGjpg Imagen1.jpg r1 manage 48.6 K 2014-12-17 - 21:18 AndresCabrera  
PNGpng Imagen1.png r1 manage 75.5 K 2014-12-17 - 21:48 AndresCabrera  
JPEGjpg Imagen2.jpg r1 manage 53.5 K 2014-12-17 - 21:18 AndresCabrera  
PNGpng Imagen2.png r1 manage 75.8 K 2014-12-17 - 21:48 AndresCabrera  
JPEGjpg Imagen3.jpg r1 manage 55.1 K 2014-12-17 - 21:18 AndresCabrera  
PNGpng Imagen3.png r1 manage 63.7 K 2014-12-17 - 21:48 AndresCabrera  
PNGpng Imagen4.png r3 r2 r1 manage 406.2 K 2014-12-17 - 22:02 AndresCabrera  
PNGpng Imagen41.png r1 manage 59.6 K 2014-12-17 - 21:54 AndresCabrera  
PNGpng Imagen5.png r1 manage 45.1 K 2014-12-17 - 21:18 AndresCabrera  
PNGpng Imagen6.png r1 manage 38.3 K 2014-12-17 - 21:18 AndresCabrera  
PNGpng Imagen7.png r1 manage 135.3 K 2014-12-17 - 21:18 AndresCabrera  
PNGpng hv_scan.png r1 manage 92.2 K 2015-01-16 - 18:00 AndresCabrera  
PNGpng rpc-schema.png r1 manage 70.9 K 2015-01-16 - 17:22 AndresCabrera  
PNGpng rpc_dist.png r1 manage 860.0 K 2015-01-16 - 17:31 AndresCabrera  
Edit | Attach | Watch | Print version | History: r7 < r6 < r5 < r4 < r3 | Backlinks | Raw View | WYSIWYG | More topic actions
Topic revision: r7 - 2015-03-20 - AndresCabrera
    • Cern Search Icon Cern Search
    • TWiki Search Icon TWiki Search
    • Google Search Icon Google Search

    Main All webs login

This site is powered by the TWiki collaboration platform Powered by PerlCopyright & 2008-2019 by the contributing authors. All material on this collaboration platform is the property of the contributing authors.
Ideas, requests, problems regarding TWiki? Send feedback