Thermal Imaging Method



Introduction

This page is dedicated to the Thermal Imaging method used at Iowa State University. The main goal of this method is to find flaws in the stave core construction by looking at the thermal profile of the stave core with coolant running through its pipes with an infrared camera. This method of flaw finding will be sensitive to flaws between the cooling pipes, thermal foam, and facing. It will not be sensitive to flaws between the facing and the honeycomb structure. In order to cool the stave core, a system consisting of a chiller and booster pump push cooled fluid through the stave while an acrylic isolation box is used to isolate the stave from its surroundings.

Installation & Equipment

The main setup consists of 7 fundamental parts:

  1. Chiller
  2. Booster Pump
  3. Acrylic Box
  4. Thermocouples
  5. Relative Humidity Sensor
  6. Dry Nitrogen System
  7. Thermal Camera and Research IR software

Extensive and detailed information about every component of the equipment including their manuals, documentation and how to set them up can be found in Equipment and Installation.

Measurements

Below is a description of how a stave core is analyzed step by step.

Camera Calibration

First and foremost the camera must be calibrated and aligned with the box. This step is discussed with more detail in the the equipment and installation section above.

  • After turning on the camera and connecting it to the Research IR software, the box should be checked to see if there is a uniform background temperature.
  • If the camera has not been aligned with the box that should also be done.

Stave Installation

When transporting the stave utmost care must be taken as not to damage it or the titanium pipes.

  • Before installation quick connect fittings must be attached to the pipes on the stave. When screwing on the fittings a copper washer should be placed between the adaptor and the stave. Without this piece there will not be a tight seal and it may leak.
  • Once connected, the quick connect fittings click the stave into the fluid loop. If this is too tight each of the pipes for the input and output can be adjusted slightly by loosening the screws holding the pipes in the stave cradle.
  • Once both sides have been connected to the cooling system, the stave should be pushed up against the two plastic mounts and secured such that the stave is aligned with the wire. If it is having a hard time aligning it may be necessary to adjust the stave cradle.
  • The shutoff valves inside the box need to be opened. This is very important and should be checked before sealing up the box!

Stave Cooling

After stave installation the chiller and cooling system need to be run.

  1. Seal up the box.
  2. Make certain the nitrogen gas is turned on and flowing into the box. Once the nitrogen is turned on air leaks can be found and dealt with.
  3. Check the chiller fluid level to see if it is around 7 cm. If not add more fluid and update the logbook.
  4. Follow the standard procedure for turning on the cooling system. This is attached here.
  5. Once the system is turned on set the pump rate to 22. This rate gives a flow rate of around 1 L/min.
  6. When the humidity gets to 1% or less, it is appropriate to turn the chiller down to temperatures below 0˚C.
When taking data at low temperatures it can take around 45-60 min or more to get the stave down to -40˚C. When going to high temperatures it will generally take less time, though both forms of operation depend on the starting temperature. While using the Novec fluid it is VERY IMPORTANT not to set the chiller to temperatures above 50˚C. This will cause the fluid to evaporate wreaking havoc on the system.

Taking Data

When taking data, it is important to make certain everything is operating within normal parameters. Before taking data check that...

  1. The pump rate is set to 22.
  2. The temperature at the input and output are within only a few degrees (2˚C for high 4˚C for low) of each other and both values has not changed significantly over the last five minutes.
The Research IR software needs to be set in a specific manner for looking at a single side of the stave to find its flaws using our analysis code.
  1. In the record settings, the frame rate needs to be set at 50 frames / sec.
  2. The software should be set to take only 200 frames.
  3. In the control menu turn off the auto NUC.
When confident that all of these variables have been set, you can take data. From this point, only two videos of the stave need to be recorded for each side; one at high temperature and one at low temperature, for a total of 4 videos per stave.

Image Analysis Software

When doing these analyses the data analysis software ROOT and python 2.7 are used. These need to be installed before using pyROOT and the following analysis packages.

AtlStaveQAInfraRedAnalysis Package

The main package that has been developed for the analysis of sequence files from the FLIR camera to generate thermal profiles and then analyze them is the AtlStaveQAInfraRedAnalysis package. It is built up of multiple python scripts that convert a sequence file from the FLIR camera to an analyzed plot of the thermal profile of a stave core. To install use git in the command line:

$ git clone https://github.com/jieyu11/AtlStaveQAInfraRedAnalysis.git

The package is made up of multiple small scripts that do multiple things:

  1. read_sequence.sh: This takes in a sequence file and then converts it to a root file. It requires an emissivity for the conversion. It is built up of many other scripts.
    1. seqToBin.py: Converts the sequence file to a binary one
    2. binarytotext.sh: Takes the binary file and the camera data from the sequence file to generate a text file with the actual temperatures
    3. texttoroot.py: this converts the individual text frames to individual .root along with frame_average.root
  2. frameanal.py: This converts a root file made from read_sequence to images of a stave core from a config file. It then finds the pipe temperature vales and creates a root file that is a summary of these fits called result.root.
    1. configFinder.py: This finds the stave core from an image automatically. It works well (Jan2019) on 13 module stave cores at +50 or -40 C.
  3. seqToProfile.py: This performs read_sequence.py and frameanal.py for multiple sequence files. It allows the user to run many at once and it takes ~11min/seq file. IF the configFinder fails, the script will skip to the next file and may require a manual running of the read_sequence and frameanal.
  4. defectFinder.py: This reads in multiple result.root files generated from frameanal.py and then looks for defects! It can be used to do many different comparisons between each result.root file.
  5. Other packages:
    1. Vignetting.py: A packaged used for measuring vignetting.
    2. SliceFinder.py: A package used to generate data to compare a real thermal image to an FEA image

-- CarlosMiguelVergelInfante - 2017-02-14 -- WilliamDaleHeidorn - 2019-01-31

Edit | Attach | Watch | Print version | History: r10 < r9 < r8 < r7 < r6 | Backlinks | Raw View | WYSIWYG | More topic actions
Topic revision: r10 - 2019-01-31 - WilliamDaleHeidorn
 
    • Cern Search Icon Cern Search
    • TWiki Search Icon TWiki Search
    • Google Search Icon Google Search

    Sandbox 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