The key to success: examine key parameters for a successful no-clean assembly.

While investigating a medical product sensitive to stray voltage, my company discovered that the original bare hot-air solder level (HASL) boards passed all process validation and accelerated environmental stress screening (ESS) when built up. However, when the full production units were run, they showed early failures in the field and were classified as no trouble found (NTF).

Investigating NTF Failures

The NTF numbers continued to increase and a detailed investigation took place. The units showed failures in boxed assemblies prior to shipment. Three months into product launch, the assemblies were tested for ionic cleanliness. The assemblies showed levels of bromide that were 10-15 times higher than the original qualification data.

All failures showed no visible corrosion or leakage pathways and passed testing when baked for three hours at 85[degrees]C. Electrical measures confirmed the failures.

We began a detailed investigation of the bare board and assembly process. With a 9% failure rate--growing every month--understanding what variables affected the circuit performance was important.

A data review shows key differences with the weak organic acid (WOA), bromide, chloride and sulfate levels. The current production data showed these levels to be much higher. What was creating the failures and why is there a difference between the two groups?

Bare Boards

The data from the original group was low in all the anions reported. The current production boards show higher chloride, bromide and sulfate levels, indicating a poor cleaning process and possibly a different line. The boards for production came from a new, high-volume automated process line that was not yet qualified. The level of bromide from the HASL flux can cause the stray voltage leakage problem. The levels of chloride and sulfate are only slightly elevated; when the bromide level is reduced, the other residues will also be reduced.

Surface-Mount Technology/Reflow

The surface-mount technology reflow data show similar results for the reported ionic levels. The levels are typical for the no-clean paste and thickness used. The levels of ionics from the reflow step pose minimal risk for electrical leakage or corrosion problems.

Wave and Hand Soldering

The product passed and performed well when exposed to high humidity. But the current production units show much higher WOA levels that can cause the stray voltage due to excess flux that has not been completely heat-activated.

Since the bromide from the HASL flux can also cause these failures, we must evaluate clean, bare boards with the current parameters and the reduced optimal parameters from applying the flux during wave and hand soldering.

Conclusions

The cleaned bare boards show lower ionic residues when built using the current production specs, but fluxing parameters have little effect on the final performance because they still fail. When we use the current bare boards and reduce the fluxing parameters, the boards still fail testing. Only when the cleaned bare boards are built with optimized flux levels do we see good performance and results similar to the original qualification samples.

[ILLUSTRATION OMITTED]

TABLE 1: ESS testing results highlight the crucial parameters.

        all values are                    Ion Chromatography
      [micro]g/i[n.sup.2]

Sample Description               Cl-    Br    S[O.sup.2-.sub.4]    WOA

Original Qualification Boards
and Data

Bare HASL Boards Original        0.27   1.36  0.18                 0.00

HASL / SMT No Clean Paste        0.21   4.36  0.00                 9.35
Reflow

HASL / SMT Reflow / Selective    0.24   5.12  0.00                77.37
Soldering No Clean wave

HASL / SMT Reflow / Selective    0.19   5.39  0.00                99.36
Soldering / Hand No Clean

Current Production Samples

Bare HASL Boads 3 months later   3.54  14.77  1.54                 0.00

HASL / SMT No Clean Paste        2.45  18.91  0.00                11.59
Reflow

HASL / SMT Reflow / Selective    2.27  20.12  0.00               182.36
Soldering No Clean Wave

HASL / SMT Reflow / Selective    2.05  21.39  0.00               223.74
Soldering / Hand No Clean

Assessment of Changes from
Current Production

Cleaned bare HASL boards         0.29   1.08  0.11                 0.00

HASL / SMT Reflow / Soldering /  0.21   4.57  0.00               208.86
Hand / Current Parameters

Cleaned bare HASL boards         0.22   1.11  0.09                 0.00

HASL / SMT Reflow / Soldering /  0.20   5.02  0.00               105.36
Hand / Reduced Parameters

Bare HASL Boads 3 months later   3.21  16.54  1.33                 0.00

HASL / SMT Reflow / Soldering /  2.15  20.98  0.00                97.48
Hand / Reduced Parameters

         all values are              ESS
       [micro]g/i[n.sup.2]         Testing


Sample Description

Original Qualification Boards and  65C/85%
Data

Bare HASL Boards Original          not tested

HASL / SMT No Clean Paste Reflow   not tested

HASL / SMT Reflow / Selective      not tested
Soldering No Clean wave

HASL / SMT Reflow / Selective      Pass
Soldering / Hand No Clean

Current Production Samples

Bare HASL Boads 3 months later     not tested

HASL / SMT No Clean Paste Reflow   not tested

HASL / SMT Reflow / Selective      not tested
Soldering No Clean Wave

HASL / SMT Reflow / Selective      Fail
Soldering / Hand No Clean

Assessment of Changes from
Current Production

Cleaned bare HASL boards           not tested

HASL / SMT Reflow / Soldering /    Fail
Hand / Current Parameters

Cleaned bare HASL boards           not tested

HASL / SMT Reflow / Soldering /    Pass
Hand / Reduced Parameters

Bare HASL Boads 3 months later     not tested

HASL / SMT Reflow / Soldering /    Fail
Hand / Reduced Parameters

Terry Munson is with Foresite Inc., Kokomo, IN; [phone]; email: [e-mail]; Web page: www.residues.com.