PROBE-BURN · Probe burn phenomena: Predictive modeling and characterization for high power wafer test applications
FP7 — People (Marie Curie Actions)
- Duration
- 2011-03-01 → 2016-08-30
- EU contribution
- €100,000
- Participants
- 1
- Scheme
- MC-IRG
Lines connect the coordinator with its partners.
Results in brief
Probe burn phenomena: Predictive modeling and characterization for high power wafer test applications
During the past four years, the researcher has established a materials and testing lab for undergraduate engineering class he taught, also helped setup a physics and electronic labs for the university use. In addition, he completed establishing a research lab and experimental setup for probe burn test and numerical simulation facilities at the university. He targeted, in his research, developing a scheme for simulating and experimentally determining probe burn phenomenon in wafer testing environment. The designed experiment was based on the measurement of current carrying capability (CCC) for wafer probes. The CCC is a critical parameter for probe cards. The current carrying capability depends on many parameters such as probe tip diameter, ambient temperature, thermo-physical properties of material, the duration of applied current and the contact resistance at the probe tip-bond interface. Measurements were made on probe test cards and results were published in reputable journals and presented in international conferences. The standard measurement approach used in the test industry is conducted to define the mechanical degradation of the cantilever probe on the wafer card and temperature distribution along the probe body is conducted using a conduction heat transfer equation via computational discretization. In this work, it is established that the maximum current carrying capacity is defined and the probe burn phenomenon is observed at the tip region of the tungsten–rhenium cantilever probe due to effects of Joule heating for both experimental and numerical results. Reasonably good agreement is observed between experimental and computational results. General results can be summarized as follows: • Probe burn phenomenon is investigated with experimental and numerical methods. • The relation between mechanical degradation and temperature distribution of the probe is shown and examined. • Results suggest that both experimental and numerical approaches are comparable and complementary. Coupled thermal-electric computational mechanics techniques have been developed to understand the temperature distribution along a special design spring and cantilever probe body in order to model the probe burn phenomenon for conduction. The experimental maximum current carrying capability tests have been performed and compared with numerical solutions. Reasonably good agreement was observed between experimental and numerical results. A predictive model was developed as a design tool to enable faster probe design for cantilever or vertical types, assembly and test cycle for a wafer sort environment. In addition to the first mode, transient heat transfer between a heated spring probe and its close environment is investigated. A continuum finite volume simulation is used to analyze the heat flow within and from the resistively heated probe to its environment. Experimental results are conducted for spring probe with laminar air flow and without air flow. The numerical and experimental results are compared and strong similarity is observed.
Data: CORDIS, © European Union
Project objective
As the probe size shrinks due to finer device pitches, its current carrying capability (CCC) is also reduced, conflicting with the future wafer test requirements for power delivery. Probe burn can result in damage to probe-card or pad of wafer tested, causing significant semiconductor wafer yield loss. The approach in the test industry has been mostly experimental for baseline data for determining this key parameter-CCC of probes on probe cards. Often, the CCC of a probe calculated empirically, does not agree well with measurements for dc and especially pulsed current applications. Accurate analytical models are needed for predicting probe CCC under both dc or pulsed current test loading for probes under compression. The main goal of the proposed research is to establish fundamental understanding of the probe burn phenomena through development of accurate models of probe current carrying capacity and verification of numerical and analytical models by data collected using a precise experimental measurement methodology. We will develop advanced analytical models representing electro-thermal characteristic of contacting probe and its environment by using advanced meshing techniques to perform simulations for any probe geometry. Time dependent algorithms to predict pulsed current loading will be included in the predictive tool. There will be special focus on cantilever or MEMS-type probes where the current restriction and probe burn happens on the probe tips. The software tool developed will be help users to optimize probe design and predict CCC limits accurately, thus prevent provide probe burn failures before wafer test. Another objective is to develop a cost-effective current limiter tool via thermistors mounted on a printed-circuit-board for advanced probe card technologies. This will be demonstrated using probe-cards for specific device power requirements.
Original text from CORDIS.
Participants
- ISTANBUL SEHIR UNIVERSITESI · Dragos Kartal IstanbulCoordinatorTürkiye
Links
Data: CORDIS, © European Union
