JPH0466883A - Monitor burn-in apparatus - Google Patents

Abstract

PURPOSE:To enable the detecting of deficiency in a burn-in board or a socket by providing a memory to store the results of a monitoring history for several times of testing for each burn-in board and each socket. CONSTITUTION:A monitor burn-in device section 1 performs a burn-in testing of an IC and monitors 4 the results thereof. Here, the device section 1 is provided with a socket history memory 6, which is made up of a memory map 7 per serial number of a burn-in board 2 and corresponding to sockets in the board 2 individually and the results of each test are stored for each socket. A parameter is set and a socket is considered as defective depending on how many times the same socket of the same board 2 is rejected continuously. In other words, a continuous deficiency matrix value 8 is provided and a judging circuit 9 compares with the set value a defective level for each socket as outputted from the device 6 to judge deficiency of the socket. This clarifies the number of the burn-in board or the socket which causes abnormality continuously or many times thereby enabling judgment of abnormality in the IC or the socket.

Description

DETAILED DESCRIPTION OF THE INVENTION [Field of Industrial Application] The present invention relates to a burn-in test device for semiconductor integrated circuits (hereinafter referred to as IC), and particularly to a monitor burn-in device.

[Conventional technology]

Conventionally, this type of monitor burn-in device supplies a test signal from a characteristic test circuit 3 to each IC of each socket of each burn-in board 2 mounted on the monitor burn-in device section 1, as shown in the configuration diagram of FIG. However, when an abnormality in the electrical characteristics of the IC is detected by the monitor circuit 4, the address specified by the RAM memory 5 Heburn inboard number and socket number (Al...Nn) stores this data. The fail data is stored.

After the burn-in is completed, the IC that became abnormal during the burn-in
The burn-in board number and socket number on which the burn-in board is mounted are printed out, and the defective ICs on the burn-in board are removed and classified according to the printout data. The fail map in the memory is not accumulated for each test count, and will be erased and overwritten at the next burn-in.

[Problem to be solved by the invention]

The conventional monitor burn-in device described above has a drawback in that even if the burn-in board or socket becomes defective due to an abnormality, it cannot be distinguished from the abnormality due to the IC.

That is, in the case of a socket abnormality, even if the IC in that part is a good product, it is determined to be defective every time, leading to an increase in costs. Therefore, if ICs in the same socket on the same burn-in board become defective consecutively, it is assumed that there is a high probability that the socket is abnormal, and maintenance must be performed at an early stage.

The present invention differs from the conventional monitor burn-in device described above in that it has monitor history results for the number of tests for each burn-in board and socket, and can detect defects in burn-in boards or sockets.

[Means to solve the problem]

The monitor burn-in device of the present invention includes a storage device capable of storing monitor history results for the number of tests for each burn-in board and each socket, and circuit means for determining whether the burn-in board is defective or the socket is defective based on the results. are doing.

〔Example〕

Next, the present invention will be explained with reference to the drawings.

FIG. 1 is a system configuration diagram of an embodiment of the present invention, in which the present invention is added to a conventional monitor burn-in device.

As described in the related art section, the conventional monitor burn-in device section 1 has, in addition to the burn-in function, a characteristic test circuit 3 that supplies test signals to ICs mounted on the burn-in board 2 via sockets, It is comprised of a monitor circuit 4 that monitors the data, a RAM memory 5 that temporarily stores the monitoring results, and the like.

Conventionally, defective ICs on the burn-in board 2 can be known from the fail map on the RAM memory 5. Or - In order to classify a plurality of burn-in boards 2 for boat fishing, a serial number of each burn-in board 2 is formed on the burn-in board 2 using a diode matrix, etc., and this can be known electrically. There is.

Here, this embodiment has a socket history storage device 6, which has the number of bits corresponding to the expected number of tests as one piece of socket information, and this is stored for each serial number of the burn-in board 2 and for each socket in the burn-in board 2 ( Al・
. . .Nn), respectively. In addition, a parameter for setting the number of consecutive failures of the same socket on the same burn-in board 2 to be considered as a socket failure, that is, a continuous failure setting value 8, is set, and the value of the parameter and the socket history storage device 6 are set. It has a determination circuit 9 that compares the failure status of each socket and determines whether the socket is defective, and an output circuit 10 that outputs the result.

FIGS. 2(a) and 2(b) are bitmap diagrams corresponding to individual sockets (or ICs) for explaining the memory map 7. FIG. In this case, each bit has a good value for each number of uses.
Defective results (eg, 0=good, 1=defective) are shown and are recorded from the lowest to the highest. Naturally, it is assumed that the number of bits corresponding to each socket is sufficient to withstand the expected number of tests as described above. Here, if the value of continuous failure setting value 8 is set to °°3'', this condition, that is, failure (1'') does not occur three or more times in a row in the bit map 11 in Figure (a), and in this case, the socket It is assumed that there is no possibility of a defect, and on the other hand, since the bitmap 12 in FIG. 12(b) satisfies this condition, the corresponding socket is determined to be defective, and the output circuit 10 prompts the socket to be repaired.

As an application example of the present invention, socket history storage device 6. The socket defect detection section consisting of the continuous failure set value 82 judgment circuit 9 and the output circuit 10 is completely separated from the conventional monitor burn-in device section 1, and information is exchanged between them via LAN or floppy media. Of course it is also possible.

Alternatively, the socket defect detection section can be treated as fail map information and used to categorize good and defective products when removing ICs from burn-in boards.

〔Effect of the invention〕

As explained above, the present invention can store IC and socket abnormality data in a storage circuit and record a history for each socket of each burn-in board. This has the advantage that the socket number can be determined, and an abnormality in the IC or socket can be distinguished.

[Brief explanation of the drawing]

Figure 1 is a system configuration diagram of an embodiment of the present invention, and Figure 2 (
FIG. 3 is a block diagram of a conventional monitor burn-in device. 11. , monitor burn-in device section, 2... burn-in board, 3... measurement test circuit, 4... monitor circuit,
5... RAM memory, 6... Socket history storage device, 7... Memory map, 8... Continuous failure setting value, 9
... Judgment circuit, 10... Output circuit, 11... Bitmap, 12... Bitmap.

Claims (1)

[Scope of Claims] 1. A monitor burn-in device that performs a semiconductor integrated circuit burn-in test and monitors the results, comprising a storage means for accumulating and storing the test results for each socket of each burn-in board, and a circuit means for reading them. A monitor burn-in device characterized by comprising: 2. Claim 1, further comprising circuit means for determining burn-in board failure or socket failure based on stored test results.
Monitor burn-in equipment as described.