Date: Sat, 15 Feb 97 20:12:21 EST
 From: "esvax::mrgate::a1::kenyonwg"@esvax.dnet.dupont.com
 To: [log in to unmask]
 Subject: RE: ASSY: Omegameter reading for RMA vs. WS flux.
 Message-Id: <[log in to unmask]>

 From:    NAME: WILLIAM G KENYON
  FUNC: Chemicals/Electronics
  TEL: 302-652-4272                     <KENYONWG@A1@ESVAX>
 To: NAME: [log in to unmask] <"[log in to unmask]"@ESDS01@MRGATE@ESVAX>

 Northern Telecom did an extensive investigation of the response
 of WS flux residues vs. time in an ionic tester of their own
 design when they switched from rosin to water soluble.  Quite to
 their surprise they found the release rate of WS flux residues
 from the laminate surface was much slower than rosin fluxes.  In
 fact, where the rosin fluxes would have released 90-95% of their
 residues into the 75/25 IPA/water solution within the standard 15
 min. test period, the level of WS flux residue release did not
 approach 90-95% until the tester had run about 2 hours.  Thus
 many users in the industry assumed they were getting much cleaner
 boards with the WS flux process vs. rosin.  As you can see, this
 was an artifact of comparing the results from an ionic test
 system optimized for one flux residue (rosin) with the data from
 a system that was not optimized for the residue (WS) being
 tested.
 The military proposed work to come up with an optimized system
 for measuring WS flux residues in the Soldering Standardization
 Plan in 1980-81, but the work was never funded nor done.

 Interestingly, your report states that higher numbers were
 obtained.  There are two factors that could contribute to this:
 first, the levels of halide activator are much higher in WS
 fluxes than in rosin fluxes.  Your flux supplier can give you the
 data here). Thus plain TCA will remove RMA residues to the levels
 needed to pass mil spec tests, where alcohol azeotropes are
 normally needed to get RA fluxes to pass.  In work done early on,
 I was testing Freon TE, Freon TES, the development product that
 was commercialized as Freon TMS and Freon T-E 35 on telephone
 assemblies soldered with Alpha 611 RMA flux.  (The T-E 35 used a
 35% blend of ethyl alcohol in CFC-113 in the boil sump, with the
 4% ethanol azeotrope in the rinse sump.  Both are % by weight).
 We had a steady supply of assemblies to clean and use for ionic
 testing.  Statistical analysis of the results showed that the
 groups were "statistically indistinguishable" -- i.e., that RMA
 flux didn't present the cleaning challenge needed to separate and
 rank various cleaning agents.  The downstream effect of all this
 was that proposed work on engineering an in-line cleaner for T-E
 35 was dropped and the effort was put into support for efficient
 cleaner designs for TMS, and that we used RA wave solder flux and
 RA solder paste for the B-36 assemblies in the CFC Phase 1 and 2
 tests.
 With that bit of cleaning and testing trivia posted, the other
 reason the results might be higher and erratic is the WS flux
 used.  All the work I did on WS fluxes in the early '80s were on
 the polyglycol types- which were the only ones available.  As the
 industry gradually accepted the fact that many of the standard
 polyglycols used in IR reflow fluids, HASL fluids, WS fluxes etc.
 could have a very negative effect on SIR, industry leaders in the
 flux world formulated new materials that comply with the WSF0
 category (no polyglycols). The release rate for these materials
 might be quite different from the older formulations; however I
 don't think anyone has looked into that.  Also, the solids or
 non-volatile loading may be much less, which makes generation of
 a foaming version that applies a precise flux loading more
 difficult.  This could be the reason for erratic numbers.
 In my WSF/RA/SA flux work, I did flux loading studies on the test
 boards at three different conveyor speeds and reported the data
 at Nepcon West in '83.  Top and bottom side pre-heat profiles at
 the three different speeds were reported as well.  The boards
 were cleaned and then ionics, solder defects, SIR and EmR data
 were given.  So there are data availble that could be used to
 compare with the numbers you are getting today.
 --Bill Kenyon
 Global Centre for Process Change, Inc.
 302-652-4272/-5701 Tel/Fax

##############################################################
TechNet Mail List provided as a free service by IPC using LISTSERV 1.8c
##############################################################
To subscribe/unsubscribe, send a message to [log in to unmask] with following text in the body:
To subscribe:   SUBSCRIBE TECHNET <your full name>
To unsubscribe:   SIGNOFF TECHNET
##############################################################
Please visit IPC web site (http://www.ipc.org/html/forum.htm) for additional information.
For the technical support contact Dmitriy Sklyar at [log in to unmask] or 847-509-9700 ext.311
##############################################################