Electricity: measuring and testing – Fault detecting in electric circuits and of electric components – Of individual circuit component or element
Reexamination Certificate
1999-03-10
2002-02-19
Mai, Son (Department: 2818)
Electricity: measuring and testing
Fault detecting in electric circuits and of electric components
Of individual circuit component or element
C324S754090, C716S030000, C716S030000
Reexamination Certificate
active
06348805
ABSTRACT:
BACKGROUND OF THE INVENTION
1. Field of the Invention
The invention relates to electrical testing of printed circuit boards, multi-chip modules, and other planar electrical interconnect devices.
2. Background of the Invention
Numerous problems arise in testing electrical interconnect devices. These devices require simultaneously contacting a set of electrical test points distributed in a nonuniform pattern across the surface of the electrical interconnect device (the “test point grid”). The test point grid is typically planar. The test points must be connected electrically to a regular, planar grid of measurement contact points such than an isolated, conductive circuit path is created to each test point. A “measurement contact array” is held parallel to the test point grid at a distance of a few inches from the surface of the test point grid.
For illustrative purposes in the remainder of this disclosure, a printed circuit board (PCB) will be used as the exemplary item presenting the plane of points to be tested. The points to be tested on a PCB are positioned to match terminals of electrical components that will eventually be soldered to the PCB. Electrical components have their terminal connection points positioned in various patterns. The connection points are often more closely spaced than the spacing between, or pitch of, the contact points in the measurement contact array.
FIG. 1
shows a view perpendicular to the two planes (test point grid and planar contact array) for a small section of an exemplary PCB. The large crosses
22
depict measurement points on a measurement contact array
20
. The solid rectangles depict test points
24
on the surface of the PCB (the test point grid). In this example, the density parallel to the rows of test points
24
is four times greater than the pitch of the measurement contact points
22
.
A method of implementing the conductive path in common use in the electrical test industry is to employ rigid or semi-rigid metal “probe pins” to contact the PCB test points. A distal end of the probe pin is held in mechanical contact with a test point on the PCB and the other end is held in mechanical contact with one element of the measurement contact array. A typical probe pin has a length of 3 inches and a diameter of 0.020 inches, providing a very high aspect ratio. Longer or shorter pins are sometimes used.
Defining which measurement contact point of the measurement contact array will be connected to a particular test point by a probe pin is referred to as “assigning” a probe pin to a test point. The particular set of all assigned probe pins for an exemplary PCB is referred to in the art as a “pin assignment pattern.”
Because the pattern of test points on the PCB surface does not present a regular X-Y coordinate grid, and because the area-density of these test points is greater than the measurement points of the measurement contact array, it becomes necessary to angle or lean most or all of the probe pins away from vertical in order that all of the test points be contacted simultaneously. This is called “deflection” in the trade, even though the probe pin itself is not bent. The set of all probe pins defined by a pin assignment pattern, plus the insulating mechanical structure that supports and constrains these probe pins are assembled into a mechanical appliance referred to in the trade as an “electrical test fixture.”
FIG. 2
a
shows examples of the key elements of interest related to an electrical test fixture
28
. Three example test points A, B and C are shown on test point grid
26
. These points A, B and C are connected to three measurement contact points
22
a,
22
b
and
22
c
by three deflected probe pins
30
a,
30
b
and
30
c.
The amount that a probe pin
30
deflects away from vertical to the plane containing the measurement contact array
20
and test point grid surfaces
26
is measured as a linear distance shown as d in
FIG. 2
b.
This distance d is the sine of the angle of deflection a multiplied by the length of the probe pin. Measuring deflection as a distance rather than as an angle is customary in the trade. This deflection assumes a straight probe
30
, and that the measurement contact array
20
is parallel to the test point grid
26
.
Constructing a test fixture
28
requires that a pattern of probe pins
30
be produced which simultaneously connect each test point A, B or C to a unique measurement contact point
22
without any two pins
30
a,
30
b
or
30
c
coming in contact with one another. To construct a usable test fixture
28
, the pins
30
must be mechanically supported in three dimensional space such that they cannot move in any direction parallel to the surface of the planes containing the measurement contact array
20
or the test point grid
26
.
An ideal fixture support structure would be constructed from a solid block of insulating material of thickness equal to the distance separating the test point grid from the measurement contact array. One hole would be bored through the insulating block for each pin. The bored hole may describe a straight line, an arc, or a combination of the two. All holes that emerge from the top side of the block will have (x,y) locations that are a multiple of the grid pitch, aligning with the measurement contact array. The holes that emerge from the bottom (PCB) side of the block will match the unique pattern of test points on the test point grid.
By placing the insulating block on top of the PCB to be tested, inserting probe pins into each hole, placing the measurement contact array on top of the test fixture structure, and applying a compressive force to the entire stack, the desired electrical circuit paths connecting each test point are realized.
In current industry practice, the single, solid block of insulating material is replaced by 3 to 10 parallel plates of insulating material, separated by spacer posts at the periphery and bolted together. This assembly of plates and spacers occupies the identical volume as the solid block of the ideal fixture. It constrains the pins in an analogous manner, using one hole per pin per plate. A typical test fixture
28
will contain between 5,000 and 25,000 probe pins.
Because probe pins
30
can deflect in any direction, a multiplicity of different probe pin assignment patterns may be chosen to build test fixtures
28
for a given printed circuit board. All such fixtures
28
must have the same pattern of holes on the bottom surface for contacting the test points A, B or C in the test point grid
20
. However, some or all of the probe pins
30
will be connected to different points on the measurement contact array
20
. Any of the multiple probe pin patterns define valid test fixtures for a given PCB as long as all test points are connected to a unique measurement contact point.
However, certain pin assignment patterns will provide superior electrical performance to others. A commonly used measure by which to judge the relative quality of several fixtures designed for the same PCB is to compare the deflections of the most deflected pin or pins within each fixture. Given two test fixtures for the same PCB, the one with the lower maximum deflection is considered superior to those in the trade.
There is a sound physical reason for using maximum deflection as a predictor of fixture performance. As its deflection rises, the ability of a probe pin to make a reliable electrical connection to its test point decreases. At some critical deflection, electrical contact is broken. An empirical deflection limit is typically set, below which experience has shown that probe pins operate reliably. The further below this limit all pins of a fixture are deflected, the more reliable a fixture becomes, although below a certain limit, little improvement in reliability is seen.
Current Methods for Test Fixture Pin Assignment
Developing a probe pin pattern to contact 25,000 points simultaneously without shorting any two of the probe pins together is a difficult problem. Even with computer assistance, technicians skilled in the art of fixture design r
Guthrie Thomas L.
Jackson William E.
Mai Son
Mania-Barco GmbH
LandOfFree
Method and apparatus for assigning pins for electrical... 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 and apparatus for assigning pins for electrical..., we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Method and apparatus for assigning pins for electrical... will most certainly appreciate the feedback.
Profile ID: LFUS-PAI-O-2963295