Controlling flux deposition: methods for measuring uniformity and volume.

When applying flux, the key issues are:

* How the flux is applied.

* What type of flux is used.

* How much flux is applied.

* Uniformity of the deposition.

Once a flux has been selected and approved for a product, it is essential that a consistent amount be applied to all boards and divided equally over each one. Techniques used in flux application include spraying with ultrasonic or atomizing air spray nozzles. In both instances, controlling certain parameters will affect the amount of flux deposited.

The ultrasonic fluxer creates a fine mist of flux that may result in less penetration in the copper barrels of the board and nonuniform deposition. The atomizing spray fluxer has more vertical power to spray flux into barrels, but is more sensitive to clogging. The system has a large number of parameters that define deposition, so proper selection of settings is required.

Controlling uniformity. There are no tools to measure the uniformity of flux deposition on boards. Process engineers use glass plates, fax paper and other methods to control this parameter. There is a tool used to define flux layer uniformity. Since this tool is a visual inspection method, an agreement analysis was performed to validate this measurement. The measurement tool (Figure 1 Online) provides easy, rapid qualification of the spray flux topside board penetration and uniformity. Test paper is placed on top of a test mesh. The tool is fluxed like a board. After fluxing, the test paper is removed and the user is able to analyze the topside board penetration and uniformity.

To validate this method, a statistical technique called Kappa-Iota is used. A total of 20 samples were made using different spray settings. Three appraisers were asked to quantify the samples. After that, the samples were shuffled; second and third inspections were made, and the consistency of the appraisers and repetitions were investigated (Tables 1 and 2).

Table 2 can be interpreted as follows: Using Object 7, for example, all appraisers validated this board as having good uniformity and penetration of the flux. But for Object 3, the appraisers did not agree on the level of penetration (Figure 2).

The experiment's outcome was that kappa = 0.635 (kappa > 0.8 is required). For the individual performances of the appraisers, the intra-rater kappa values are better:

Appraiser 1 = 0.91, appraiser 2 = 0.78 and appraiser 3 = 0.73.

[FIGURE 2 OMITTED]

Measuring flux amount. A simple method to measure flux amount is to use a balance. Measure a PCB before and after spraying returns the wet amount of flux. A more accurate method is to dry the board and use a more sensitive balance to measure the weight of the solids of the flux. Commercially available flow meters are thermodynamic or Coriolis-based. These devices measure the amount of flux flowing through the nozzles. The amount on the PCB will be less because of overspray and solvent evaporation during spraying. For every assembly, one is able to define the correlation between flow measured with a flow meter, and the real amount of flux on the PCB.

To validate a Coriolis flow meter, a gauge R & R experiment is done. A typical layout of a gauge R & R experiment is used; e.g., spray different boards with 10 different settings by each of three appraisers and repeat the experiment (Table 3).

For flux amount, [+ or -] 10% tolerance is good and more than acceptable. The results of this validation are:

* The Coriolis flow meter has a gauge R & R tolerance of 18% (<30% is acceptable).

* Operator versus sample interaction is not significant.

Flux uniformity can be evaluated with a flux measurement tool. However, the criteria should be well defined, and the inspectors should have sufficient experience with the process to understand indications regarding deviation of the flux. A Coriolis flow meter is able to measure the flux flow accurately. A thermodynamic flow meter showed less accuracy.

Gerjan Diepstraten is a senior process engineer with Vitronics Soltec BV (vitronics-soltec.com); [e-mail]. This column appears monthly.

[ILLUSTRATION OMITTED]

Table 1. Ranking Scheme

Ranking  Penetration  Uniformity

1        Good         Good
2        Good         Poor
3        Poor         Good
4        Poor         Poor

Table 2. Appraiser Results

        Appraiser 1  Appraiser 2  Appraiser 3
Sample  1  2  3      1  2  3      1  2  3

1       2  2  2      2  2  2      2  2  2
2       2  2  2      2  2  2      2  1  2
3       2  2  2      4  4  4      4  2  4
4       2  2  2      4  4  4      4  4  2
5       2  2  2      2  2  3      2  2  2
6       2  2  2      2  4  4      2  2  2
7       1  1  1      1  1  1      1  1  1

Table 3. Coriolis Flow Meter Data

Coriolis  Appraiser 1   Appraiser 2   Appraiser 3
Board     1      2      1      2      1      2      Average

 1         7.59   7.69   7.71   7.71   7.63   7.63   7.66
 2         7.50   7.41   7.52   7.54   7.45   7.47   7.48
 3        12.24  12.27  12.39  12.42  12.30  12.30  12.32
 4         8.29   8.31   8.42   8.42   8.31   8.33   8.35
 5         5.55   5.55   5.62   5.62   5.56   5.58   5.58
 6         6.46   6.45   6.53   6.54   6.48   6.48   6.49
 7         8.97   8.97   8.97   9.10   9.04   9.04   9.01
 8         7.94   8.00   8.06   8.04   7.98   7.98   8.00
 9         9.35   9.35   9.47   9.50   9.40   9.40   9.41
10         6.49   6.49   6.57   6.57   6.52   6.52   6.53
Average    8.04          8.14          8.07         mg/[cm.sup.2]