Method and apparatus for testing welds between dissimilar metals by using the seebeck effect

• US 4,019,364 A
• Filed: 09/08/1975
• Issued: 04/26/1977
• Est. Priority Date: 09/08/1975
• Status: Expired due to Term

• Current Assignee: AT&T, Inc.
• Original Assignee: Western Electric Co., Inc. (Nokia Corporation)
• Inventors: Maddox, Harry L.
• Primary Examiner(s): Goldstein, Herbert
• Application Number:
  US05/610,964
• Time in Patent Office: 596 Days
• Field of Search: 73/15 R, 73/15 FD, 73/15.4, 73/359, 219/109, 219/110, 219/118, 219/121 LM, 324/32, 324/51, 340/228 R, 340/248 W, 340/253 W
• US Class Current: 374/5
• CPC Class Codes: G01N 25/72 Investigating presence of flaws
• CPC Class Codes: G01N 27/002 by investigating the work function voltage

Abstract

To test certain structural characteristics of a weld formed between two dissimilar metals in order to test the quality, or structural integrity, of the weld, the emf generated at the weld bond interface during welding, as a result of the Seebeck effect (thermocouple), is monitored. As the weld cools, the emf decreases at a first essentially rapid rate with respect to time until the weld begins to recrystallize, or fuse, at which point the emf decreases at a second and slower rate since the latent heat of fusion of the metals decreases the cooling rate of the weld. When recrystallization is complete, the emf decreases at an essentially exponential rate. It has been discovered the rate of change of the emf with respect to time during recrystallization is an indication of the quantity of the metal involved in the weld where the weld is exposed to a constant heat sink, such as the atmosphere, since the more metal that is involved in the weld the slower the weld will cool during fusion, and vice versa. The quantity of metal involved in the weld is indicative of the quality, or structural integrity, of the weld, since the more metal that is involved in the weld the greater the likelihood that a secure weld has been formed. The rate of change of the emf with respect to time during fusion is compared with a predetermined rate of emf change with respect to time which is indicative of a predetermined weld mass of the metals to obtain one indication of the quality of the weld. Further, the occurrence of any discontinuity in the emf during cooling of the weld is detected, and indicates the development of a crack or a venting gas bubble at the weld bond interface, either of which decreases the integrity of the weld.

Claims

1. In a method of testing a weld formed between two dissimilar metals by the application of welding energy to the weld, where an emf is generated across the weld bond interface and has a value related to the temperature of the weld bond interface as a result of the Seebeck effect:

monitoring the emf across the weld as a result of the emf generated across the weld bond interface;

detecting the occurrence of a loss of the monitored emf subsequent to the application of the welding energy to the weld and during the cooling thereof, and

generating an indication of a detected loss of the monitored emf.

2. In a method of testing a weld as set forth in claim 1, the detecting step includes detecting the occurrence of a loss of the monitored emf across the weld from the time that the weld begins to cool and until recrystallization of the weld is essentially complete.

3. In a method of detecting the occurrence of a crack through a weld formed between two dissimilar metals by the application of welding energy to the weld, where an emf is generated across the weld bond interface and has a value directly related to the temperature of the weld bond interface as a result of the Seebeck effect:

monitoring the emf across the weld as a result of the emf generated across the weld bond interface;

detecting when the monitored emf stops increasing in a first direction as the weld is heated upon the application of the welding energy thereto, and increases in a second and opposite direction as the weld cools subsequent to the application of the welding energy thereto;

detecting, upon the monitored emf increasing in the second direction, the occurrence of a loss of the monitored emf, and

generating an indication of a detected loss of the monitored emf.

4. In a method of testing a weld formed between two dissimilar metals by the application of welding energy to the weld, where an emf is generated across the weld bond interface as a result of the Seebeck effect and has a value directly related to the temperature of the weld bond interface:

monitoring the value of the emf across the weld as a result of the emf generated across the weld bond interface;

measuring, subsequent to the application of the welding energy to the weld, the time required for the monitored emf to change from a first predetermined value of emf representative of a first predetermined weld temperature at which recrystallization of the weld has essentially begun to a second predetermined value of emf representative of a second predetermined weld temperature at which recrystallization of the weld is essentially complete, the measured time being representative of the weld mass of the metals;

comparing the measured time required for the emf to change from the first to the second predeermined values of emf with a predetermined time which is equal to the time required for a predetermined weld mass of metal to cool from the first to the second predetermined temperatures;

detecting the occurrence of a loss of the monitored emf, and

generating an indication of a detected loss of the monitored emf.

5. In an apparatus for detecting the occurrence of a crack through a weld formed between two dissimilar metals by the application of welding energy to the weld, where an emf is generated across the weld bond interface and has a value directly related to the temperature of the weld interface as a result of the Seebeck effect:

means for monitoring the emf across the weld as a result of the emf generated across the weld bond interface;

means for detecting when the monitored emf ceases to increase in a first direction as the weld is heated upon the application of the welding energy thereto, and increases in a second and opposite direction as the weld cools subsequent to the application of the welding energy thereto;

detector circuitry, operative upon the monitored emf increasing in the second direction as the weld begins to cool for detecting the occurrence of a loss of the monitored emf, and

an indicator for generating an output upon a detected loss of the monitored emf.

6. In an apparatus for testing a weld formed between two dissimilar metals by the application of welding energy to the weld, where an emf is generated across the weld bond interface as a result of the Seebeck effect and has a value directly related to the temperature of the weld interface:

means for detecting and for amplifying the emf across the weld as a result of the emf generated across the weld bond interface;

means for detecting, subsequent to the application of the welding energy to the weld, the rate of change of the emf across the weld with respect to time as the emf changes from a first predetermined value of emf representative of a weld temperature at which recrystallization of the weld has essentially begun to a second predetermined value of emf representative of a weld temperature at which recrystallization of the weld is essentially complete;

means for comparing the detected rate of change of emf with respect to time with a predetermined rate of emf change with respect to time which is indicative of a predetermined weld mass of the metals, and

means for detecting, subsequent to the application of the welding energy to the weld, the occurrence of a loss of the detected emf.