Skip to topic | Skip to bottom
Home
Instrumentation
Instrumentation.HPKr1.10 - 10 Jun 2013 - 15:37 - AlvaroDosiltopic end

Start of topic | Skip to actions

Strip sensor tests

Objectives

Test the Alibava+sensor setup and obtain the S/N value for the sensor.

Setup

The main parts of the setup were: ALIBAVA+STDB(4 BEETLES)+HPK(R-Type strip sensor)

Alibava:

STDB: The Silicon Tracker Daughter Board can only read 2 Beetles simultaneously. It has two sets of switches to select the them. Their I2C directions are: 0x10, 0x12, 0x14, 0x16. As far as we know we must select the pair 0x10-0x12 or 0x14-0x16, never mix them. 0x12 is not bonded yet.

HPK sensor: R-Type Hamamatsu strip sensor. Bias = -130V. Some times, after the end of the bias voltage ramp up, the current grows until ~300 nA without any apparent reason.

Beetle: Only the BEETLE 0x16 has the DATA_VALID signal wired, and is read always and appart from data. So when we want to read the 0x10-0x12 BEETLEs we have to use one Arduino to program them and set DATA_VALID=1. It is very important to remember that the Beetle channels are inverted with respect to the sensor pads, i.e. beetle channel 0 is connected to pad 127 of the HPK sensor and the channel 127 is connected to the pad 0. Beetle 0x14 is connected to the inner most strips and 0x16 is connected to the next ones. The third innermost strips are connected to the Beetle 0x10, and the 0x12 is not connected yet.

Arduino: Warning: the serial port can be tricky. You may need to unplug all the others USB connectors to select the arduino's good one. The procedure is: Load the code -> Select the correct board -> Compile code -> Upload program -> Program Beetles from the arduino terminal

Scintillator: 12 V. (We used the H. Schindler's scintillator. Remember to connect again the banana connectors).

Alibava schematic setup

Alibava field setup

Basic procedure:

The next steps are the basic ones to start a test with this setup:

  1. Connect the Xilinx Platform Cable USB II to upload the correct firmware to the Alibava. Remember that there is a specific firmware for each pair of beetles. (Due to problems with the firmware for the directions 0x14-0x16 we changed the I2C directions 0x14-0x16 to 0x10-0x12 and we used the checked firmware)
  2. Upload the firmware with iMPACT. Strongly recommended to upload the firmware with only the power and I2C cables connected to avoid glitches that could damage the board.
  3. Check the trigger signal.
  4. Connect the trigger and data cables. Check that the STDB switches point to the pair of beetles wanted. (This is independet from the I2C directions and always need to be chanded)
  5. Upload the Arduino code and program the Beetles
  6. Open the Alibava software and change the configurations: Beetle parameters, trigger threshold, the number of events to record...
  7. Take one run of pedestals to check the correct behaviour.
  8. Set the file where you want to save the data you will get in the next run.
  9. Change the power supply current limit to the closest value to 300 nA.
  10. Close the black chamber door and deplete the sensor (~ -130 V). Be careful to start the ramping up at -1 V to avoid any positive voltage and untick the zero check button.
  11. Take a run of pedestals (~10000)
  12. Unbias the sensor. Recomended to stop at -1 V and then stop the power supply.
  13. Locate the radioactive source and the scintillator in the right position around the sensor.
  14. Close the black chamber door and deplete again the sensor.
  15. Change the file where to save the new data.
  16. Take your data.
  17. ...
  18. When finished unbias the sensor, remove the radioactive source and return it to its secure place. Remove the sensor and save it in a humidity controlled atmosphere. Turn off every electronic device.

Analysis

To analyze data from alibava you need the data analysis framwort ROOT.

Test on 06/06/2013 (by P. Rodriguez and A. Dosil)

The setup was the mentioned above.

At the first time we tried to use the two firmwares available for the two sets of I2C directions, but the 0x14-0x16 always got stack between 5 and 140 triggers received. At first sight, it was impossible to distinguish if the problem comes from firmware or software. A. Dosil, P. Rodriguez and P. Vazquez studied the behavior of the system, sending one by one triggers to narrow the problem:

  • We observed the Alibava output trigger signal and determined that the system crashes after sending the trigger signal to the beetles.
  • We monitored the data sent from the beetles to the Alibava and concluded that the system crashes after this communication.
  • We studied the software behavior and determined that the data generated in the last trigger is received, so the problem is after that.

We conclude that the failure can be finishing the communication between the board and the pc, leaving the system in a incorrect status to receive a new trigger or processing the new trigger, just before send it to the beetles.

We changed the I2C Beetle directions 0x14-0x16 by 0x10-0x12, so we had two beetles with the direction 0x10 and one (because the original 0x12 is not bonded) 0x12. We do not expect any problem when programming them because the use the I2C bus, neither with the output data because we use the STDB which has two sets of switches to select each pair. With this configuration, we can use the firmware for the 0x10-0x12 directions, which reported much less stack problems than the 0x14-0x16 one.

The system is not stable at all, and needs a restart when it was working for a long time (>30000 triggers). For this reason we took runs of 20000 particles.

Another issue is a random system failure always preceeded by a "tic" sound. This sound do not comes from the Alibava board, but it hangs the system and we need to restart the board and software, loosing the current run data. P. Rodriguez connected the fridge power to an independent power plug and this problem has never happened again.

Results

In the next plot the raw data is showed for the 256 inner strips of the R-sensor.

Raw data

The next figure shows the histograms for the signal. The left plots is created from the 256 inner strips and the right plot is drawn with data from the strips 257-384 (note that the outer strips are not bonded).

Signal histograms

The next plot shows the signal noise ratio versus radius. The fit tell us that singal/nois is independent from the radius and equal to 31.4 +- 0.5. So, it is also independent from the pitch.

S/N versus radius

The capacity of the strips depends on its length, so we would expect that the noise would also increment with length, contrary to our results. Our pitch adapter is connected compensating the strip lengths, i.e., the pitch adapter shorter strips are connected to the longer sensor strips and vice versa. So, the total strips length is similar and the noise becomes independent from the sensor strip length. -- AlvaroDosil - 03 Jun 2013
to top

I Attachment sort Action Size Date Who Comment
test_setup.png manage 5187.2 K 05 Jun 2013 - 15:14 AlvaroDosil Test Setup
image00.png manage 361.2 K 05 Jun 2013 - 15:27 AlvaroDosil Alibava schematic setup
signal_inner_outer.pdf manage 21.3 K 10 Jun 2013 - 15:28 AlvaroDosil Signal in inner and outer strips
SN_vs_radius.pdf manage 26.5 K 10 Jun 2013 - 15:28 AlvaroDosil S/N vs radius
strip_vs_signal.pdf manage 62.6 K 10 Jun 2013 - 15:28 AlvaroDosil Signal vs strip
signal_inner_outer.png manage 77.3 K 10 Jun 2013 - 15:37 AlvaroDosil Signal in inner and outer strips
SN_vs_radius.png manage 69.7 K 10 Jun 2013 - 15:38 AlvaroDosil S/N vs radius
signal_vs_strip.png manage 150.3 K 10 Jun 2013 - 15:38 AlvaroDosil Signal vs strip

Copyright © 1999-2019 by the contributing authors. All material on this collaboration platform is the property of the contributing authors.
Ideas, requests, problems regarding this material Send feedback