SPC data from the Post Etch Punch provides the total change due to all of the above as a measured difference between the intended panel target location and the actual location at the time of punching.

This value is referred to as SPREAD and the number can either be positive or negative depending on whether the innerlayer panel has grown or shrunk.

SPREAD, or dimensional change, can be a function of laminate material, the source (vendor) of the material, the thickness of the core, copper weight, the image etched on the surfaces (power or ground planes vs. signal planes), and even how the individual panel has been sheared from a larger sheet (warp and fill). SPC data of SPREAD is used to determine the acceptability of layers at the punching operation, provided it is known what to expect for dimensional change.

In the lamination process step, additional and possibly more significant dimensional changes take place due to the forces generated by heat and pressure. Influencing this dimensional change are the materials used, press cycle, lamination equipment, individual core thicknesses, copper weight, and circuit pattern on each of the innerlayers.

After lamination, the traditional process step was to mount the laminated panel using the four slot lamination tooling holes to soft tooling plates on the drill.

One, or more panels would be drilled and the results analyzed by x-ray viewing the drill registration to the circuit pattern, or to some coupon for that purpose. From this, the Drill Room Supervisor would make a determination of offsets to put into the drill program to compensate for the shift that always resulted. The entire lot would most likely be drilled using these somewhat arbitrary offset values. Recently, machines that drill pinning holes utilizing x- ray to locate innerlayer images have become more wide spread. The Optiline PL, Post Lamination Drill, is a system that sights on a target stack of all the innerlayer panels using x-ray technology coupled with a vision system controlling an X-Y-Theta panel positioning system.

The vision system aligns the targets to reference fiducials, splits the error caused by material movement, and drills new tooling locations for drilling. The SPC feature in the system reports the SPREAD of the viewed target stack. Since the Optiline PL has split the error in every panel, offsets to correct shrinkage or expansion caused by lamination is not required as each panel has been compensated for this during the x-ray drilling.

In addition to providing the data of SPREAD, the Layer Analysis Package (LAP) software available in the Optiline PL compares the distance between additional targets on each individual layer to the intended (reference) distance. This value is again reported as SPREAD and can either be positive or negative. It is now known how the multilayer panel has grown or shrunk and how each and every layer has grown or shrunk. In addition, since the SPREAD data has been recorded at the Optiline PE, a comparison can be made between data at the punch and data prior to drilling and determine what the influence is by the lamination process alone on a layer by layer basis.

Take the SPC gathering effort one step further back in the total process.

Typically with a new job circuit CAD data is added to a standard frame. This frame contains all of the nomenclature required for part identification, test coupons, and fiducial marks used for various checks and alignment. Once the features are all added, the entire image including the frame is scaled based on some history or experience. Typically scaling is different in the two axes.

The Automatic Artwork Punch also provides the same sort of SPREAD data as the Optiline PE and the Optiline PL. After plotting the film and during the punching of the film, the artwork punch will report the change in the distance between the fiducials. This is a check on whether or not the artwork has been scaled properly.

A suggested method to the use of these Multiline SPC gathering tools follows utilizing a test panel or test lot of panels.

• Plot the phototools in the normal fashion. All of the necessary fiducials should be added at their nominal positions. Do not scale the image, but record the SPREAD values when punching the phototools, do this for each film.
• At the Post Etch Punch record all of the SPREAD values reported for each innerlayer. There will be a change from nominal.
• After lamination and at the Post Lamination Drill, record all of the LAP data for each innerlayer in the book.
• Use the data from the Optiline PL Layer Analysis software to go back, if necessary, and scale the artwork by layer to yield the best possible product at the end step.
• After rescaling, use the SPC data collection capability of the Artwork Punch to confirm that the scaling has been done properly.

It may be fair to ask, why record the SPREADs at the Post Etch Punch at all, why not just use the Layer Analysis software data for artwork compensation? Once this initial control lot has been analyzed by specific layer, the Optiline PE can be set up to look for a specific SPREAD value during processing. This is accomplished by setting the high and low acceptance tolerance during the job set-up. That is, if layer B behaves differently than the other layers through the process there will be a specific SPREAD value at the Post Etch Punch that will satisfy the end requirement after lamination. This provides an on-going production gauge without an additional operation. This information is provided on every panel processed. A panel or lot of panels that falls outside of the tolerance window is not allowed to move to the next step and, more importantly, this is a signal that something has changed in the process.

The Automatic Phototool Alignment Unit used for phototool alignment in outer layer or soldermask imaging also provides SPC data on SPREAD. Although in this case the system reference is not “hard tooled” into the unit. The system reference is the part or multilayer panel with all of the dimensional changes that it will probably have. The data reported is the difference between the panel target and the phototool target. If the data indicates an unacceptable error, then rescaling the film should be a direct route to resolve the problem. This can be done in two axes, by merely providing an additional pair of fiducials at 90o on both the panel and the film on the test lot.

In the adjustment phase of the test lot, rescaling for change after etch and rescaling for change after lamination, the film can be checked using SPREAD data to ensure that no scaling error has been made. This same check can be made when reproducing working films (diazo) from file films (silver). A master film held in a file can possibly change with time.