Comparing surface isolation resistance measuring data

Introduction
The isolation resistance of a PCB surface can be affected by the behaviour of the flux residues that might remain on the PCBs after soldering. The effect on the surface is depending on the climatic conditions and on the type of flux that has been used.
Often the flux supplier gives test data for the behaviour of these residues under a defined climatic environment for a restricted time period. This information can be used to compare the data with the users demands that are specified for the correct functioning of the circuit on the PCB, e.g. the product specification.
Comparing test data
If one wants to compare isolation resistance test data one must know with which test pattern the measurements were done. Often so called comb track patterns are used for such tests, like those from the IPC-B-25 test board. This board contains however three test patterns with different spacing between the tracks. So it is important what pattern has been used for the test. Obviously the pattern with the large spacing between the tracks will give a higher value than the pattern with the moderate or fine spacing.
There is however a simple method to compare such tests done with different patterns. For that it is necessary to know the relation between the pattern length(l) and the width in-between the pattern tracks (w).The factor l/w can be used to compare different isolation measurements (Ris) and convert them to the relative surface isolation resistance (r).
Calculation example
For a glass epoxy base material the surface isolation r during the test should be >16.105 MΩ.
If we use the finest pattern from the IPC-B-25 test board, what should be the minimum level of Ris during the test?
From the finest pattern, the A pattern, the l/w ratio is 2000. This means that the minimum boundary for Ris during the test will be:
Ris = r / (l/w) = > 16.105 MΩ / 2000 = > 8.102 MΩ.
If the same test was done with the B pattern, having a l/w ratio of 500, the minimum value of Ris should be:
Ris = r / (l/w) = > 16.105 MΩ / 500 = > 32.102 MΩ.
Remark
If one find relatively very high surface isolation values, the test has probably been applied at a pattern that has a low l/w ratio. In general a ratio of > 25 is recommended for such tests. It is also possible that only the Ris value under dry conditions is expressed. The value requirement for Ris under dry conditions is commonly a factor 5 higher than the measurements under humid conditions.