FI104998B - Method for testing the reliability of integrated circuits - Google Patents

Description

104998

The present invention relates to a method for testing the reliability of integrated circuits. More particularly, the invention relates to test methods in which the connections in each test circuit are coupled as desired for testing purposes to provide a suitable resistance within the circuit, and the test circuits are coupled to a power supply to supply a measuring current through the test circuits.

New low-volume component fabrication technologies where 10-conductor connections are made directly to silicon substrates, such as flip chip, COB (Chip on Board), CSP (Chip Scale Packaging) and TAP (Tape Automated Bonding), are interesting especially for small and portable but increasingly sophisticated electronic products. . The reliability of these types of circuits is key to reliability, so the manufacturer must ensure the reliability of the component technology connections of his choice before moving on to them.

One of the most common testing methods is to expose the circuits to cyclic temperature variations. This is done by connecting the circuits of the circuits together in a so-called. daisy-chain configuration by placing the number of test circuits required for statistically significant results in a test cabinet and supplying them with a current that causes a voltage drop across the connections. This voltage drop is monitored either directly over the circuit junctions or by measuring it with the exact series resistor in series with it, whereby the change in the junctions causes a change in current and thus appears as a voltage change in the meter.

The problem is that each circuit must have its own power supply, whereby the amount of cabling for the power supply is 2n, where n = the number of components to be tested. If you want to use the so-called. 4-point method, each - * component is additionally equipped with data acquisition cables, resulting in a total cabling of 4n. Thus, parallel power supply increases the amount of cabling and, if there are many components, it may happen that the thick cable bundle does not fit into the test cabinet.

Another problem is that parallel supply will never provide absolutely equal current for all components, because of the connections, 2,104,998 cable lengths, the possible series resistor, etc., there will always be uncontrolled voltage losses, which will change the magnitude of the current. fully r comparable to each other.

5 Of course, the circuits can also be connected in series, reducing the need for cabling. The problem here is that if one of the circuits connected in series breaks down, all circuits will be power-free and will not be tested for them. Thus, in known methods, the circuits are connected in parallel, and each of the chained circuits is provided with its own power supply.

It is an object of the present invention to provide a test method which avoids the above disadvantages and is based on connecting the test circuits in series to a power supply to supply a measuring current through all the circuits under test, wherein the measurement is performed as voltage measurement over each circuit. The method according to the invention is characterized in that a diode downstream of the measuring current is connected across each circuit, so that the measuring current is supplied from a constant current source so that the voltage across the circuits is lower than the threshold voltage of the diode. the constant current source corrects the magnitude of the measurement current.

Preferred embodiments of the invention are characterized in what is set forth in the claims below.

In accordance with the invention, the current is supplied to all components in series instead of the parallel supply using a constant current generator and a bypass diode for each component. If any component breaks down during the test, its circuit 25 will be interrupted, but current will begin to flow through the bypass diode with the constant current generator taking care to maintain the current despite the interruption.

The amount of cabling required by the invention can be significantly reduced, from 2n to 2 for power supply alone, and from 4n to 2n + 2 in four-point metering, which, for example in 10 component measurements 30, reduces the number of cables from 40 to 22. In addition, since the same current circulates through all the components, a precisely comparable voltage across the various components is obtained as the measurement data.

3, 104998

The invention will now be described in more detail by way of example with reference to the accompanying drawings, in which Figure 1 illustrates a way of connecting so-called circuit connections. daisy-chain method; Figure 2 shows a measuring circuit corresponding to the method according to the invention; Figure 3 shows an example of the results obtained by the measuring method according to the invention.

As shown in Fig. 1, the daisy-chain method 10 connects all the connection bumps 1 of the circuit to the joint so that they are connected to each other alternately on the substrate side 2a, alternately on the circuit side 2b. This results in a relatively high resistance value between the power supply points of the circuit. 1. by four-point switching, the measuring current I is supplied to the circuit by a power supply 3 along different wires and at positions la over which the voltage across the circuit (mV) is measured by voltage-15 meter 4. Thus, the voltage losses of the power supply lines and The coupling can be used, for example, to measure the relative changes in resistance values in each circuit in a cycling test. In the known parallel current supply solutions, four wires are required for each circuit.

Figure 2 shows a measuring circuit corresponding to the method according to the invention. Circuits IC1 to IC5 to be tested, shown here as resistors, j. since their resistance is the most important information for the measurement, it is connected in series.

In practice, there may be up to hundreds of circuits connected to Saija. In addition, a bypass diode D1 ... D5 is connected across each circuit. Voltage is measured separately for each circuit, either by four-point measurement as shown in Figure 1, or directly across the terminals of the circuit and the diode. Diodes D1 ... D5 (e.g., 1N5626) are connected in a direction downstream of the measuring current. The measuring current I from the constant current source 3 is (such as 18 mA) such that the voltage across the circuits IC1 ... IC5 is lower, e.g., about 0.2 V, than the threshold voltage of the diode at 0.63V at IN5626. 10mA.

30 When a circuit breaks down, the resulting power failure in the circuit raises the voltage across the circuit and the bypass diode so high that it causes the diode to conduct. but entering a state, that is, a power failure is eliminated. Thereafter, the constant current source 4 104998 corrects the magnitude of the measuring current I again, allowing the test to continue uninterrupted, regardless of the circuits broken during the test.

Figure 3 shows the test results obtained by the method of the invention for 20 circuits out of a total of 60. The test included three different types of flip-chip circuits A, AA, and B, which were tested in the same series connection in a test chamber at which the temperature was allowed to vary from -40 ° C to + 125 ° C for two hour cycles for a total of 1000 cycles. Voltage readings were taken at both high (h) and low (c) temperatures. When the voltage rises to the diode threshold voltage, it is a sign of circuit breakdown (f). Thus, degradation can be detected immediately during the test, which is advantageous for subsequent tests as they are usually designed during previous tests. Differences between different circuit types are also well visible, e.g. circuit A has a somewhat lower resistance than others, and on the other hand has less reliability.

It will be apparent to one skilled in the art that various embodiments of the invention are not limited to the above examples, but that they may vary within the scope of the following claims. Thus, the method according to the invention is applicable to the reliability tests of all components, e.g.

• «*>

Claims (4)

A method for testing the reliability of integrated circuits, wherein the terminals (1) of each circuit (IC1-IC5) to be tested are interconnected in the desired manner (2a, 2b) to provide a resistance suitable for the test purpose in the circuit, and wherein the circuits to be tested are connected in series with a current source (3) for simultaneously supplying a measured current (I) of predetermined size through all circuits to be tested, the measurement being performed as voltage measurement (4) over each circuit, characterized in that over each circuit (IC1-IC5) is connected to a diode (D1-D5) connected to the measuring current in the ratio current, so that the measuring current is measured from the constant current source (3), so that the voltage (U) across the circuits is less than the threshold voltage of the diode, caused by a broken circuit (f) transmits the diode in a conductive state, after which the magnitude of the measuring current (I) is restored to the original with the constant current source (3). Method according to claim 1, characterized in that the connections (1) in each circuit to be tested have been connected to a so-called. daisy-chain method, where adjacent terminals are interconnected in turn (2b) in the circuit, in turn on the substrate (2a). 20 Method according to claim 1 or 2, characterized in that in the test the voltage across each circuit (IC1-IC5) to be tested is determined to determine the condition of the circuit. Method according to claim 1 or 2, characterized in that in the test the voltage is measured over one with each circuit (IC1-IC5) to be tested in series resistance to determine the condition of the circuit. *