AUTOMATED LEAK DETECTION METHOD USING A ROBOTIC SNIPER LEAK DETECTOR
Patent Information
- Application Number
- DE502022004354
- Authority / Receiving Office
- DE · DE
- Patent Type
- Patents
- Current Assignee / Owner
- Priority Date
- 2021-10-07
- Filing Date
- 2022-07-25
- Publication Date
- 2025-07-10
- Estimated Expiration
- 2042-07-25
AI Technical Summary
Current leak detection methods for heat exchange devices are time-consuming and prone to errors due to human factors, such as omitting test areas or not moving the sniffer probe sufficiently close to a test area.
A robot-guided leak detection system that uses a 3D sensor to automate the movement of a sniffer probe to predetermined test locations, records measurement signals and positions, and correlates these data to identify potential leaks.
The system enables fully automated and reliable leak detection, reducing human error and increasing efficiency by accurately identifying leaks in heat exchange devices.
Description
[0001] The invention relates to a leak detector for detecting leaks in a test object.
[0002] A leak detector is described, for example, in DE 10 2005 022 156 A1 (Inficon). This leak detector comprises a probe whose sniffer tip is placed at predetermined test regions of a test object. The test object is filled with a test gas, e.g., helium. Escaping test gas is sucked in by a base unit via the sniffer tip and fed to a test gas detector, which can be configured, for example, as a mass spectrometer.
[0003] When using leak detectors, the test object, e.g. an air conditioner or the cooling unit of a refrigerator, is filled with a test gas, and a sniffer probe can detect whether test gas is escaping from the test object. During quality inspection of products in industry, the sniffer tip is applied to specific test areas of the test object where there is a possibility of a leak. The probe is then manually moved to the test areas. During this process, it is difficult to check whether the sniffer tip has been moved to all relevant test areas of the test objects. An operator may inadvertently omit certain test areas or skip other test areas that they subjectively consider to be non-critical.
[0004] US Patent No. 4,945,305 (Ascension) describes a position-determining system comprising a transmitter for generating a pulsed DC magnetic field and a receiver disposed on the object. This method is particularly suitable for determining a current position, comparing it with a desired position, and providing feedback. No interference is caused by non-magnetic objects in the line of sight between the transmitter and the receiver. Residual interference from large masses of magnetizable material, such as a compressor block and a refrigerator, can be eliminated by calibration because the assembly under test is static. This method has proven particularly useful for implementing the disclosure.
[0005] WO 2009 / 016160 A1 describes a leak detector comprising a base unit connected to a probe via a tube. The sniffer tip is positioned on test zones or test areas of the test object. If test gas escapes from the test object, this is detected by a test gas detector in the base unit. A positioning system is provided, comprising a transmitter, a receiver located in the probe, and a feed and evaluation unit. This monitors and confirms the presence of the sniffer tip in the individual test areas.
[0006] In the field of automation, it is known to capture digital data from physical objects using a 3D sensor to thereby obtain a point cloud in a 3-dimensional virtual space, where the points each represent a surface point on the external surface of the physical object. Robots used in the automation industry, such as in the automated manufacturing of products such as cars, are controlled by software algorithms that use the captured digital data. This is typically done to sensing the external perimeter or surfaces of the entire physical object in order to locate, grasp, move, or relocate the object, or to paint the external surface of the object using a paint-spraying robot arm, for example in the vehicle manufacturing industry.
[0007] Sniffer leak detection on heat exchange devices, such as refrigerators, air conditioners, heat pumps, etc., is part of the quality inspection of heat exchange devices. A human operator must visually identify the relevant test areas on the heat exchanger, such as fluid-carrying pipes, and manually move the probe to the test areas one by one. This type of sniffer leak detection on heat exchange devices is time-consuming and prone to errors caused by human factors, such as omitting test areas or not moving the sniffer probe sufficiently close to a test area.
[0008] DE 10 2016 226 152 A1 discloses a method and a system for leak testing a container.
[0009] DE 10 2012 008 857 A1 discloses a method and a test system for leak testing a gas system of a motor vehicle.
[0010] DE 41 40 725 A1 discloses a method and an arrangement for leak testing of containers which are pressurized with a measuring fluid in their interior.
[0011] WO 2010 / 028619 A1 discloses a device designed as a robot for the autonomous detection of leaks from piping systems with the release of substances into the environment.
[0012] The invention is based on the object of providing a more reliable and faster detection of a leak in a fluid-carrying element of a heat exchange device.
[0013] The method according to the invention is defined by the features of independent claim 1.
[0014] Accordingly, the probe tip of a robot-guided probe is moved to a test location to be examined using a robot (step a). Once the test location is reached, gas or air is sucked in through the probe tip and fed to a gas detector, which records a measurement signal of the sucked-in gas (step b). At least one first measured value of the measurement signal is recorded at at least one first time point (step c). The probe tip typically sucks in the gas or ambient air as it approaches the test location and is not activated only after the test location is reached. In addition, the first measurement time point is recorded and assigned to the first measured value (step d). Finally, a first measurement position of the probe corresponding to the first measurement time point is also recorded (step e).Position data are therefore recorded that indicate the position of a component of the probe, for example the probe tip, which the component occupied at the first measurement time. Each signal value I of the gas measurement is thus assigned a time t of the measurement and a location or position in space (x, y, z) as I(t, x, y, z). If necessary, in addition to or alternatively to the spatial coordinates of a Cartesian coordinate system, solid angle coordinates in the form of two solid angles Φ, Θ and a distance r (r, Φ, Θ) of the axis along the probe can be recorded. These steps are repeated for at least one subsequent second measurement time (step f), whereby, according to the invention, the measured values are correlated with the respective measurement positions in order to be able to assess, based on the measured values, at which measurement position an extreme value of the measurement signal, which could indicate a possible leak in the test object, was recorded (step g).In this way, fully automated leak detection is to be carried out using a robot, without a human user carrying out the evaluation, operation and / or leak detection or individual steps thereof.
[0015] Steps b) - g) can be repeated after the probe tip has been moved by the robot to another test location. The measurement position of the probe tip, for example, is preferably recorded using a 3D sensor, which can be an imaging system with at least one optical camera and preferably at least one illumination device.
[0016] While recording the measurement signal, the probe can be moved by the robot to continuously assume different measurement positions during the measurement. The measured values can be recorded as a function of the measurement position of the probe and / or the movement speed of the probe. Preferably, the measurement signal is evaluated as a function of the measurement position of the probe tip.
[0017] Advantageously, at least two measurements are carried out at the same measuring position and the recorded measured values are compared with each other.
[0018] The evaluation of the measurement signal can be carried out taking into account the speed at which the probe is moved during the measurement.
[0019] While the probe is being moved to the test location and before the measurement is taken, the probe signal can be zeroed by acquiring a measurement signal with the gas detector at a location known to be free of test gas and / or with a constant background concentration. Alternatively, the signal can be zeroed as the probe tip approaches the test location, at a sufficiently large distance from the test location so that the background signal (zero calibration) can be determined.
[0020] Calibration of the leak detection system can be performed by bringing the probe to a location (e.g. test leak) where a known, defined leak rate is emitted.
[0021] The probe can be moved to a measuring position for which a leak was detected or suspected during a previous measurement in order to carry out a control measurement at the measuring position.
[0022] Those measuring positions for which leaks are suspected from previous measurements are preferably marked.
[0023] The flow rate of the gas stream collected by the probe can be adjusted depending on the movement speed of the probe.
[0024] The gas leak detector can, for example, be a sniffer leak detector with a sniffer probe that has a sniffer tip through which the gas is sucked in. Alternatively, the gas leak detector can have a test gas probe, the detector tip of which contains the gas detector for the test gas.
[0025] An exemplary embodiment of the invention is explained in more detail below with reference to the figure. The figure shows a schematic representation of a leak detection system and a test object in the form of a heat exchanger device. The illustrated exemplary embodiment shows a gas leak detector in the form of a sniffer leak detector, whose probe is a sniffer probe through whose sniffer tip the gas to be analyzed is sucked in.
[0026] Fig. 1 shows a heat exchange device 12 in the form of a refrigerator. The term "heat exchange device" as used in the present disclosure refers to systems or devices that have a heat exchanger, such as refrigerators, air conditioners, heat pumps, etc. The heat exchange device has, on its lower rear side 14, several fluid-conducting elements 16, 18 in the form of tubes that are welded or soldered to the rear of the refrigerator during the manufacturing process. As part of a fully automated quality control, special test areas 20 must be identified for performing leak detection in the test areas 20. A test area 20 is considered to be an area of the heat exchange device in which tubes are mounted, welded, or soldered together, in which a tube ends, in which a tube is connected to another tube, or in which tubes are joined together or cross each other. In the Fig. 1 In the embodiment shown, a first fluid-conducting element 16 is arranged vertically and a second fluid-conducting element 18 is connected by one of its ends to the first fluid-conducting element 16. The area in which the tubes 16, 18 are connected is considered as the test area 20 and is in Fig. 1 shown as a dashed circle.
[0027] The background of the invention is the automatic movement of a sniffer probe 22 of a sniffer leak detector 24 to the test area 20 such that the sniffer tip 25 of the sniffer probe 22 is positioned close enough to aspirate gas escaping from a possible leak in one of the pipes 16, 18 within the test area 20. The sniffer probe 22 is connected to the gas leak detector 24 in a conventional manner via a connecting pipe or connecting capillary 26. The goal is to correlate the movement of the sniffer probe and the associated signal response of the measured signal.
[0028] The sniffer probe 22 is mounted on the distal end 28 of a robot arm 30 of a robot 32.
[0029] A 3D sensor 34 in the form of an imaging system 36, which has two optical cameras 38, 40 and an illumination device 42 in the form of an LED light, captures digital image data from the lower rear side 14 of the heat exchange device 12. The illumination device 42 illuminates the heat exchange device 12, and in particular the lower rear side 14 of the heat exchange device 12, in a generally known manner. The cameras 38, 40 capture the reflected light, and the imaging system 36 generates digital image data from which a point cloud 44 is captured in a 3-dimensional virtual space 48. Alternatively, it is possible to derive the position of the measuring probe from the known position of the robot arm. The robot arm is guided to known measuring positions. Knowing the size and orientation of the measuring probe in relation to the robot arm, the position of the measuring probe can be determined.
[0030] According to the invention, the gas detector 24 records measured values of the measurement signal of a gas stream drawn in by the sniffer tip 25, and the measurement times associated with each measured value are recorded and assigned to the measured values. Furthermore, the respective measurement positions of the sniffer tip 25 are recorded at each measurement time by recording the position of the sniffer tip, i.e., the relative position of the sniffer tip 25 with respect to the test object 12 or the measurement location 20, using the cameras 38, 40 of the 3D sensor 34.
[0031] When evaluating the measured values, they are compared with the respective measuring positions in order to be able to determine the location of a possible leak based on the measuring positions of the respective measured values and the amplitudes of the measured values.
[0032] A typical case of necessary signal correlation is caused by the delay in the signal response due to the gas's transit time through the sniffer line 26. After a test gas cloud is sucked into the sniffer tip 25, the gas flows through the line 26 to the main unit of the gas detector 24. This transit time, also called dead time, can be up to several seconds. This means that by the time a signal response occurs at the gas detector 24, caused by a sucked-in leak gas cloud at one measurement location, the sniffer probe 22 may already have been moved to the next measurement location by the robot arm 30. This time delay must be taken into account when correctly assigning the measurement signal to the measurement location.
[0033] The gas flow time must be measured beforehand. This can be done, for example, during system calibration. The robot arm 30 positions the tip 25 of the sniffer probe 22 in front of a test leak. The delay time between approaching the test leak and the signal response is measured. This delay time is subsequently taken into account when interpreting the signal responses at the gas detector 24 and assigning the leak locations on the test object.
Claims
1. An automated leak detection method using a robotic gas leak detector (56), the method comprising the steps of: a) moving the probe tip (25) of a probe (22) guided by a robot (32) to a test site (20) to be examined, using a robot (32); b) recording a measurement signal from gas recorded by the probe tip (25) by means of a gas detector (24); c) detecting at least one first measurement value of the measurement signal at at least one first measurement time; characterized by the steps of: d) detecting the first measurement time and assigning the first measurement time to the first measurement value; e) detecting and assigning a first measurement position of the probe (22) corresponding to the first measurement time; f) repeating steps a) - e) for at least one subsequent second measurement time; g) correlating the measurement values with each of the measurement positions of the probe (22) in order to be able to use the measurement values to assess at which measurement position an extreme value of the measurement signal, which could indicate a possible leak in the test object, was detected.
2. The leak detection method according to claim 1, characterized in that steps b) - g) are repeated after the robot (32) has moved the probe tip (25) to another test location (20).
3. The leak detection method according to any one of the preceding claims, characterized in that according to step e), the measurement position of the probe (22) is detected using a 3D sensor.
4. The leak detection method according to any one of the preceding claims, characterized in that according to step e), the measurement position of the probe (22) is inferred from the known position of the arm (30) of the robot (32) carrying the probe.
5. The leak detection method according to any one of the preceding claims, characterized in that the probe (22) is moved by the robot (32) during the acquisition of the measurement signal and thereby continuously assumes different measurement positions.
6. The leak detection method according to any one of the preceding claims, characterized in that the measurement values are acquired as a function of the measurement position of the probe (22) and / or the movement speed of the probe (22).
7. The leak detection method according to any one of the preceding claims, characterized in that the evaluation of the measurement signal is performed in dependence on the measurement position of the probe tip (25).
8. The leak detection method according to any one of the preceding claims, characterized in that at least two measurements are performed at the same measurement position and the acquired measurement values are compared with each other.
9. The leak detection method according to any one of the preceding claims, characterized in that the evaluation of the measurement signal is performed with consideration to the speed at which the probe (22) is moved during the measurement.
10. The leak detection method according to any one of the preceding claims, characterized in that during step a) and before step b), a calibration of the probe (22) by a zero adjustment is performed by acquiring a measurement signal by means of the gas detector (24) and storing the same as a background signal.
11. The leak detection method according to any one of the preceding claims, characterized in that the probe (22) is moved to a measurement position for which a leak was determined before, so as to perform a control measurement at the measurement position.
12. The leak detection method according to any one of the preceding claims, characterized in that those measurement positions are marked for which leaks are assumed from previous measurements.
13. The leak detection method according to any one of the preceding claims, characterized in that the feed rate of the gas flow drawn in by the probe (22) is adjusted in dependence on the movement speed of the probe (22).
14. The leak detection method according to any one of the preceding claims, characterized in that the gas leak detector is a sniffer leak detector, the probe is a sniffer probe and the probe tip is a sniffer tip.