At 09:25 PM 2/12/96 EST, you wrote:
>From:	NAME: WILLIAM G KENYON              
>	FUNC: Chemicals/Electronics           
>	TEL: 302-652-4272                     <KENYONWG AT A1 AT ESVAX>
>To:	NAME: [log in to unmask] <"[log in to unmask]"@ESDS01@MRGATE@ESVAX>
>
>
>Many of us would be interested to know how you made your ionic 
>contamination measurements; what instrument, what test solution, 
>what test time etc.
>Northern Telecom had determined that the low ionic results often 
>obtained with water washing of water soluble fluxes were in fact 
>due to the slow "release rate" of such residues from the PWA 
>surfaces.  They determined that 90+% of rosin flux residue was 
>released into the 75/25 (v/v) test solution within 15 minutes, 
>but that it took almost 120 minutes to achieve the same level of 
>extraction of flux residues from water solubles.  Thus the PWA's 
>soldered with water soluble flux "appeared" to be cleaner than 
>the traditional rosin flux--- when a 15 minute test time was 
>used.  You might wish to run a follow-up experiment to determine 
>the release rate of your water soluble fluxes and pastes vs. 
>ionic test time to check this point out and assure yourself that 
>the low ionic readings are indicative of clean PWAs.
>W. G. Kenyon
>
>
Bill,

What I have found, in my travels, is that people are not performing the
ionic test properly. Belcore has a procedure that is quite good. A sample
board is tested with the results plotted, real time. The plot results are
monitored and when the extraction curve flattens out, the test is stoppped.
The time that represents the 90% mark is the time used for testing the
boards in production. 

I have found people select a time with no foundation based on board size, or
surface area. This method will most certainly provide results indicating a
clean board, when in essence, the boards were not tested long enough for
accurate results.

Regards,

Gary Ferrari
Tech Circuits, Inc