IPC-4556 印制板化学镍钯浸金（ENEPIG）规范ENG.pdf - 第27页

Pertinent Information from the Protocol Sent to All Round Robin T est Locations: 1) Record the collimator used for the measurements. 2) Record the measurement time. 3) T ake only one reading per pad. 4) If multiple readi…

APPENDIX 3

ENEPIG PWB Surface Finish XRF Round Robin Testing

Gerard O’Brien, President

STandSGroup

INTRODUCTION

Measurement capability for thin immersion coatings has long been a concern for the 4-14 Subcommittee and with the intro-

duction of Electroless Nickel/Electroless Palladium/Immersion Gold (ENEPIG), this has become a bigger issue with the need

to measure three plated deposits. Additionally there is the impact of the phosphorus content in the Electroless Palladium that

was present in the majority of samples submitted for the generation of this speciﬁcation. For previous 455X family speciﬁ-

cations, the committee has typically veriﬁed the deposits submitted by the chemical suppliers using a reference XRF (X-ray

Fluorescence) and the delta between the readings shared with the suppliers and a general comparison chart generated. There

were a signiﬁcant number of ENEPIG users and printed board (PB) suppliers, as well as chemical suppliers involved with

the generation of this speciﬁcation, who expressed an interest in understanding the variability and accuracy issues with mea-

suring using XRF and so it was decided to perform a round robin test with all interested parties measuring a previously

‘‘certiﬁed’’ coupon.

A previously plated test coupon for switch pad testing, but never tested, was chosen for this purpose at random and sixteen

locations on the coupon were identiﬁed for measuring. A letter detailing the protocol was supplied with the sample and sent

initially to the six chemical suppliers prior to plating their test submissions. The test vehicle was then forwarded around the

world to OEMs, CEMs and PB suppliers. The data was returned to S T and S and coded to protect the source of the data.

Feedback was provided to each test site as to where their readings were relative to the actual values for Ni, Pd and Au. Based

on the data generated, some test locations asked for another chance to read the coupon after their XRF had been adjusted

or repaired.

The XRF capability used for this round robin testing ranged from simple proportional counter systems all the way through

vacuum assisted focused capillary systems costing $200K+. It should be noted that the higher cost systems did not always

produce the correct answers and that a correctly set up XRF system of any type with sufficient measurement time can be

capable of correctly measuring the ENEPIG deposit.

It was discovered during this testing that the use of incorrect calibration standards is still a reality. Non-phosphorus contain-

ing nickel standards were a common occurrence as well as gold foil thicknesses outside the measurement range. It cannot

be emphasized enough that correct measurements can only be obtained when using ENEPIG standards with the phospho-

rus content in the nickel and palladium matching that in the plated deposit.

Figure A3-1 shows the test coupon and pertinent information from the protocol sent to the test locations. Figures A3-3, A3-4,

and A3-5 show the recorded test data for the Au, Pd, and Ni measurements, respectively. Level C shows the reference data.

Figure A3-1 Test Coupon Used for XRF Round Robin Measurements

IPC-4556 January 2013

Pertinent Information from the Protocol Sent to All Round Robin Test Locations:

1) Record the collimator used for the measurements.

2) Record the measurement time.

3) Take only one reading per pad.

4) If multiple readings are taken to examine variability of the readings, record them on a separate sheet.

5) Record what material the coupon sat on when measured.

6) Include the calibration ﬁle parameters used to create the ﬁle, whether standards were used or just fundamental param-

eters.

Note: There were no measurements taken in the pads to the immediate left of pad #s 13 & 15, as they have a centered

hole in them. For pads #s 14 and 16, there are also holes present but they were offset enough so as not to interfere with

the XRF readings.

January 2013 IPC-4556

Figure A3-2 Immersion Gold Thickness Readings in Microinches for the Sixteen Test Pads Where C Is the Reference XRF

abcdef gh

test unit

ijklmn

microinches Au

Means and Standard Deviations

Level Number Mean Std Dev Std Err Mean

a 16 0.82562 0.068796 0.01720

b 16 0.85000 0.136626 0.03416

c 16 0.72188 0.071575 0.01789

d 16 0.82188 0.077908 0.01948

e 16 1.63750 0.114746 0.02869

f 16 1.04438 0.129819 0.03245

g 16 0.18038 0.079997 0.02000

h 16 0.82900 0.094297 0.02357

i 16 0.83875 0.086783 0.02170

j 16 0.46513 0.124357 0.03109

k 16 0.58631 0.106453 0.02661

l 16 0.87500 0.134164 0.03354

m 16 0.72500 0.100000 0.02500

n 16 0.96350 0.221891 0.05547

IPC-4556 January 2013