Electricity: measuring and testing – Fault detecting in electric circuits and of electric components – Of individual circuit component or element
Reexamination Certificate
2000-06-30
2002-07-30
Sherry, Michael J. (Department: 2829)
Electricity: measuring and testing
Fault detecting in electric circuits and of electric components
Of individual circuit component or element
C324S076150, C702S117000
Reexamination Certificate
active
06426643
ABSTRACT:
BACKGROUND OF THE INVENTION
An object of the present invention is a method for the testing of electronic components. It can be used more particularly in the testing of semiconductor electronic components, especially components such as integrated circuits. In the prior art, there is a known method for the testing of electronic components that reduces the unit testing time of each component. The value of the invention lies in the fact that it proposes a test method that first of all reduces the unit testing time and, secondly, optimizes the output efficiency of the goods tested.
The invention is an improvement of the international patent No. WO 97/45748, published on Dec. 4, 1997. The subject matter of this international patent is incorporated herein by reference. An electronic component generally comprises several functions. It is therefore subjected to a series of tests to test each one of its functions. The test method according to the document WO 97/45748 can be applied to each of the tests of a series such as this. A test generally comprises several successive elementary steps. Thus a method for the testing of an electronic component comprises the following elementary steps:
at an initial date Di, terminals of the component, considered to be input terminals with respect to the test, are subjected to an electric potential Vi,
there is a wait, during a period of time, for a response to appear and for this response to get stabilized at the terminals of this component, considered as output terminals with respect to this test,
at the end of this period of time, at a measurement date DM, this response is measured. In one example, a value Vs of a potential at these output terminals is measured.
During the test, the component has thus been subjected to a signal, and the response of the component to this signal has been measured at the end of a certain period of time. Then, to find out if the component has given an acceptable result or not for this test,
the measured response is compared with fixed criteria of acceptance. That is, in this example, the value of the potential Vs is compared with lower and/or higher limits of acceptance.
Since the electronic component has a known structure, it is possible to theoretically determine the period of time at the end of which an expected response can be observed at the output of the component following the application of a potential Vi at the input terminals of this component. With this theoretical time being known, for example as calculated by the designer of the electronic component, a nominal measurement date Do is determined for this test. The nominal measurement date Do is a date after the initial date Di. Moreover, the nominal measurement date is such that the duration that elapses between the initial date Di and this nominal date Do is greater than the duration of the theoretical time. Indeed, to define this nominal date Do a safety margin is chosen.
Then, the nominal date Do is used as a measurement date DM for the components to be tested. This choice, which gives preference to the guarantee of a result that complies with the test, leads to an increase in the total time of the performance of the test. Since this is true for all the tests of the series of tests applied to the component, the total testing time for such a component is thus very considerably increased.
The international application WO 97/45748 envisages a test method to decrease the period of time between the initial date Di and the measurement date Dm. Indeed, since the test method is designed to be applied consecutively to thousands, or even more, of electronic components that are themselves made in batches, the test method applied to the set of these components comprises, according to this document:
a first “learning phase” and
a second “application phase”.
In the learning phase, a population of acceptable components of the batch to be tested is considered. The acceptable components are those that have given a good result in the test performed at a measurement date Dm equal to the nominal measurement date Do. This population is a “population for learning” hereinafter called a “learning population”. In one example, this learning population may comprise a single electronic component.
Then, using the results obtained on this learning population, it is sought to define a measurement date that is the earliest possible measurement date. For this purpose, the test already performed on this learning population is reiterated, by again applying a signal to the input terminals and reading the value of the potential Vs at the output terminals on an intermediate measurement date Dmi, preferably prior to the nominal measurement date Do. If the learning population comprises a single component, then the same test is reiterated by using the same measurement date Dmi or Do to obtain several results for each of the dates tested.
Thus, a dichotomized or step-by-step procedure is carried out to test the measurement dates prior to the nominal date. To choose a measurement date from among these intermediate measurement dates Dmi tested, a comparison is made between:
a statistical image of the results of this learning population, obtained with an intermediate measurement date Dmi, and
a statistical image of the results of this same learning population, obtained on the nominal measurement date Do.
Indeed, for each intermediate measurement date Dmi tested, a statistical image of the results obtained is determined. This statistical image especially comprises the calculation of the mean M, and of the standard deviation S. Furthermore, to compare two statistical images with each other, a criterion of statistical appreciation is used. This criterion requires knowledge of the mean values and standard deviation values of the images to be compared. This a criterion of appreciation D referenced Cpi is defined for each statistical image obtained for a measurement date Dmi. In one example, this criterion of appreciation CPi is equal to a ratio between a difference in limit and the standard deviation. The difference in limits can be given by a manufacturer's tolerance, for example To. Then CPi is equal to To divided by S.
To compare the statistical images with each other, their respective criteria of appreciation Cpi are compared. The earliest possible intermediate measurement date Dmi is chosen as the measurement date. This measurement date is furthermore such that the criterion of appreciation CPi is in a certain proportion of the criterion of appreciation CPo, where CPo is the criterion of appreciation characterizing the statistical image of the values measured at the nominal date Do.
The set of the statistical images determined for each of the intermediate measurement dates gives a representation of the evolution of the values measured at the output terminals after a signal VI has been imposed at the input terminals of this component.
In a first example, if for all the intermediate measurement dates Dmi tested, a wide range of values of the potential measured at the output is observed, without any predominance of any value, then it means that that the behavior of the electronic component under this test with these measurement dates is not reliable. Indeed, until the nominal measurement date Do, the potentials are fluctuating and are never correctly stabilized. In this case, it is impossible to choose a measurement date lower than the nominal measurement date.
In a second example, which is also unfavorable, even at the nominal measurement date Do, a curve is observed representing the values of non-stabilized potential measured at output. Then it is necessary to choose a measurement date which is higher than this nominal date.
In a third example, the representation of the evolution of the values measured at the exit of the electronic component is a curve plotted with a thin line. It is possible then to consider intermediate measurement dates that are correlated with the nominal measurement date. However, even in this example, it is possible to define a minimum measurement date, below which the st
Nguyen Trung Q.
Nilles & Nilles SC
Softlink
LandOfFree
Method for the testing of electronic components does not yet have a rating. At this time, there are no reviews or comments for this patent.
If you have personal experience with Method for the testing of electronic components, we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Method for the testing of electronic components will most certainly appreciate the feedback.
Profile ID: LFUS-PAI-O-2907907