Printed circuit assemblies (PCA's) are typically tested after manufacture to verify the continuity of traces between pads and vias on the board and to verify that components loaded on the PCA perform within specifications. Such printed circuit assembly testing is generally performed with automated in-circuit testers or ICT's and requires complex tester resources. The tester hardware must generally be capable of probing conductive pads, vias and traces on the board under test.
In-circuit testers (ICT) have traditionally used “bed-of-nails” (BON) access to gain electrical connectivity to circuit wiring (traces, nets, pads) for control and observation capability needed for testing. The Agilent 3070 is one typical in-circuit tester and is available from Agilent Technologies, Inc. of Palo Alto, Calif. This necessitates having access targets within the layout of circuit nets that can be targets for ICT probes. Test access points are usually circular targets with 28 to 35 mil diameter that are connected to traces on the printed circuit board. In some cases these targets are deliberately added test pads, and in other cases the targets are “via” pads surrounding vias already provided in the printed circuit.
Lower diameter targets are increasingly difficult to hit reliably and repeatably, especially when a test fixture may contain several thousand such probes. It is always desirable to use larger diameter targets, but this is in fundamental conflict with the industry trend towards higher densities and smaller geometry devices.
Yet another industry trend is to use higher and higher speed logic families. One Megahertz (MHz) designs became ten MHz designs, then 100 MHz designs, and are now reaching the Gigahertz domain. The increases in logic speed necessitate industry attention to board layout rules for higher-speed interconnects. The goal of these rules is to create a controlled impedance pathway that minimizes noise, crosstalk and signal reflections. Printed circuit boards traces that carry high-speed signals tend to have critical layout requirements and require controlled characteristic impedances. When traditional test probe targets are added, this causes discontinuities in the controlled impedances and may damage signal fidelity.
In some higher speed designs it is also important to control the symmetry of the traces. However, routing signals on a crowded printed circuit board necessitates curves and bends in the path, which makes matching lengths and symmetries more difficult. In some cases, series components (such as series terminations or DC blocking capacitors) may be included in the path, and these have dimensions that differ from the layout parameters. Signals may also have to traverse connectors, which add to the difficulties.
Another trend is toward higher and higher density boards, which are also layout critical. When traditional test probe targets are added to a high-density board, the board layout is generally disturbed, in which adding test probe targets to one or more nodes necessitates moving several others out of the way. Such changes to high-density boards in many cases, may not be practical, as there may not be any room to move traces. If any of the signal traces also happen to be high-frequency signal traces, then the bends needed to re-route them may also have a negative performance impact as well as the negative effects of the conventional target itself.
Additional difficulties arise when testing is considered. Testing requires tester access to circuit traces at particular probe targets. Layout rules typically require test targets to be at least 50 mils apart and may require the diameter of the test point targets to greatly exceed the width of the traces.
The positioning of test targets 102a, 102b can be problematic. In many cases the need to keep a minimum separation between targets (typically 50 mils, minimum) is in direct conflict with controlled impedance layout rules. These conflicts lead to either a compromise in controlled impedance integrity and optimal circuit layout for performance and space, or a forced reduction in target placement with a resulting reduction in testability.
While high-speed printed circuit boards are one example of layout-critical circuits, another more general case is that of high-density boards. Adding conventional probe targets to a high-density board will most likely disturb the layout. For example, adding test points to just one or more nodes may require moving several other traces out of the way. In many cases, in a high-density circuit design, this may be impractical, if not impossible, as there may not be any room to move these traces in a crowded circuit layout. If any traces are also high-frequency signal traces, then the re-routing may have an additional negative performance impact as well as the negative effects to the optimal circuit layout itself.
As discussed more fully in U.S. patent application Ser. No. 10/670,649 entitled Printed Circuit Board Test Access Point Structures And Methods For Making The Same filed Sep. 24, 2003 by Kenneth P. Parker et al. and U.S. patent application entitled Method And Apparatus For Manufacturing And Probing Printed Circuit Board Test Access Point Structures filed Oct. 25, 2004 by Kenneth P. Parker and shown in
As shown in
As noted above, an important factor in probing is obtaining a good electrical contact between the fixture probe and the test target. In the old paradigm, this was typically handled by probing with a sharp probe tip 120. However, this method may be problematic with bead probes or test access point structures 18a and 18b, as the test access point structures may have very small dimensions in the x-y plane, in the order of 3-5 mils wide by 15-20 mils long or less. Current design for test guidelines for ICT (in circuit testers) require a minimum 30 mil diameter test pad or target by a chisel or spear tipped probe. The small dimensions of the test access point structures in the x-y plane in combination with the small dimensions of the chisel or spear tipped test fixture probe would very likely create test access accuracy and reliability problems. Simply, it may be hard to hit a small target with a sharp tipped probe with current industry test fixture standards. Also, while a chisel or spear tipped test fixture probe may be effective in disturbing surface contaminants on a 30 mil or greater test pad, assuming it could reliably hit a 3-5 mil test access point structure, it would very likely catastrophically damage the test access point structure. There is a need in the industry for a method to ensure good electrical contact between bead probes or test access point structures and fixture probes.
The present invention solves the problem of how to obtain a good electrical contact between a solder bead probe or a test access point structure and a fixture probe for printed circuit board testing.
In one embodiment a surface of a solder bead probe may be cleaned by a rotational motion of a twisting fixture probe as the twisting fixture probe is brought into compressive contact with the solder bead probe.
In an exemplary method for implementing the twisting fixture probe, a substantially flat tip is attached to a rotating member, which is brought into operational contact with connected to a rotating mechanism and a spring force.
In an exemplary method for cleaning and probing a solder bead probe on a printed circuit board, a twisting fixture probe is aligned with a solder bead probe and brought into compressive, twisting contact therewith. Testing may then be run with the fixture probe in contact with the solder bead probe.
A more complete appreciation of this invention, and many of the attendant advantages thereof, will be readily apparent as the same becomes better understood by reference to the following detailed description when considered in conjunction with the accompanying drawings in which like reference symbols indicate the same or similar components, wherein:
Turning now in detail to the invention, on a trace defined in an x-, y-, z-coordinate system where the x-dimension represents the trace width, the y-dimension represents the trace length, and the z-dimension represents the trace thickness, it will be recognized by those skilled in the art that present techniques for test access point placement on a printed circuit board utilize only the x- and y-dimensions. The present invention takes a different approach by taking advantage of the z-dimension, that is, the trace thickness. In this regard, the test access point structure of the invention is a localized “high point” on a printed circuit board trace that does not significantly perturb the impedance of the trace and that can be targeted with a probe. Throughout this specification, test access point structure and bead probe structure are used interchangeably.
As discussed more thoroughly in the above referenced patent applications,
As discussed above and shown in
For the purposes of discussion, assume the fixture probe has a flat surface that comes in contact with the test access point or bead 18. If the radius of curvature is tight enough, the test access point structure 18 made of solder is subject to deformation when a fixture probe contacts it with a predetermined amount of force. A typical fixture probe force is approximately 4-8 ounces. The yield strength of typical solders (both leaded and lead-free) is approximately 5000 psi. Thus, when a fixture probe compressively contacts a newly formed bead or test access point 18 for the first time, the test access point 18 will deform with a substantially flattened top, as shown in
By way of analogy, one can think of a potato as a bead probe and the potato skin as surface contaminates, residues or oxides. The potato is placed on a flat, hard surface. A second object, representing a fixture probe, having a flat, hard surface with a diameter at least as large as that of the potato is brought into compressive contact with the potato until the surface of the potato begins to deform and flatten. As this happens, the potato skin will be deformed and the flat surface of the second object, representing the fixture probe will come into contact with the inside of the potato, representing uncontaminated solder of the bead probe.
As an exemplary model, the yield strength of solder is approximately 5000 psi or 0.005 pounds per square mil, or 0.08 ounces per square mil. Thus, to support a typical 4-ounce flat probe, the flattened area 32 of the test access point 18 must be 4/0.08 or 50 square mils. Assume a 5 mil wide by 20 mil long bead 18 that is approximately 3 mils high. Assume that when the fixture probe first touches the bead 18, there is no flattened surface area. Then as the fixture probe pushes down on the solder, the area that is flattened 32 approximates an ellipse with a 1:4 width/length ratio. As this area increases, the solder yield begins to slow until there is a “footprint” of 50 square mils, or about ½ the total area of the bead itself. Once the surface area is large enough to support the fixture probe force, no further deformation occurs. Subsequent probing does not produce any further deformation.
Once probing has flattened a bead probe, the solder bead is subject to growing a new layer of oxide coating or other surface contaminants as time passes. Therefore, if the board is returned from the field for re-test or if the board had to pass through a repair and re-test cycle, there may be a new oxide or contaminant layer on the surface of the bead probe 18. Since the flattened bead is already capable of supporting the probe force, no new deformation should occur to displace this oxide or contaminant layer. While lead-free solder (containing a majority of tin) may grow a conductive tin-oxide layer, older leaded solder may grow a lead-oxide layer that is a poor conductor. Thus, this oxide layer or other surface contaminants may adversely affect re-testing.
This problem may be addressed by using a smooth faced probe that rotates or twists about its long axis (around the z axis).
It will be appreciated from the above detailed description of the invention that the present invention uniquely solves the conflict problems faced by traditional techniques for test access point placement on printed circuit boards. Although this preferred embodiment of the present invention has been disclosed for illustrative purposes, those skilled in the art will appreciate that various modifications, additions and substitutions are possible, without departing from the scope and spirit of the invention as disclosed in the accompanying claims. For example, the tip of the twisting fixture probe may be a shape other than circular, as shown in
The subject matter of this patent application is related to that of U.S. patent application Ser. No. 10/670,649 entitled Printed Circuit Board Test Access Point Structures And Methods For Making The Same filed Sep. 24, 2003 by Kenneth P. Parker, Ronald J. Peiffer and Glen E. Leinbach and assigned to Agilent Technologies, Inc., which is hereby incorporated herein by reference. The subject matter of this patent application is also related to that of U.S. patent application Ser. No. ______ entitled Method And Apparatus For Manufacturing And Probing Printed Circuit Board Test Access Point Structures filed Oct. 25, 2004 by Kenneth P. Parker and Chris R. Jacobsen and assigned to Agilent Technologies, Inc., which is hereby incorporated herein by reference.