Hi Techies

HAPPY NEW YEAR!

Now that the mental fog is clearing from the Christmas and New Year indulgences, my response to Bev’s comments:

Bev asked: How can a non-ionic species cause an ECM problem?  I do not understand how one can detect a non-ionic species by SIR or ECM testing.  You need high pressure liquid chromatography for non-ionic compounds. Or, Graham, are you talking about non-ionic materials that break down in the heat of a solder wave or if it is exposed to the heat of a reflow profile?

OK - non-ionic surfactant additives are within almost all current process chemistry’s. They are commonly ultra-low surface energy material. I am not claiming that they alone can cause ECM, rather that they can “hide” ionics. However, as you say, their behaviour during the soldering process may change.

I am looking forward to catching up at APEX….

Kindest regards

Graham Naisbitt

> On 20 Dec 2019, at 21:46, Bev Christian <[log in to unmask]> wrote:
> 
> TechNetters, 
> A few comments on what Graham wrote. I guess we can consider what he wrote his Christmas present to us all.  😉
> 
> 1) Porridge?  Yum.  Yes, an apt description of what might be under a partially cleaned LGA/QFN.
> 2) FTIR's lower limit of detection is around 2% in a mixture, assuming at least some significant vibration bands are not obscured by those of other species.
> 3) How can a non-ionic species cause an ECM problem?  I do not understand how one can detect a non-ionic species by SIR or ECM testing.  You need high pressure liquid chromatography for non-ionic compounds. Or, Graham, are you talking about non-ionic materials that break down in the heat of a solder wave or if it is exposed to the heat of a reflow profile?
> 4) "your own supply chain should be providing you with the actual, measured SIR data of their product."  Ha!  Good luck with that!  Other than from solder paste/flux manufacturers, the data is slim to none.  Oh, and by the way, using the set: copper mirror, halide ion spot test, SIR and ECM testing over the 25 year portion of my career in electronics manufacturing quality/materials labs my people and I FAILED ~30% of what we tested and we tested ALL chemicals that were going to come in contact with a print circuit board.  So that included tapes, labels, adhesives, solder masks, temporary solder masks, cleaning agents, fluxes, tacky fluxes, solder pastes, cored wires, flux coated preforms, inks, connector lubricating oils, dusters, freezers, conformal coatings, corrosion inhibitors, underfills, solvents and thermal compounds.
> 5) Yes 2020 will be an interesting year.
> 
> Merry Christmas and Happy New Year
> 
> Regards,
> Bev
> 
> -----Original Message-----
> From: TechNet <[log in to unmask]> On Behalf Of Graham Naisbitt
> Sent: Friday, December 20, 2019 12:57 PM
> To: [log in to unmask]
> Subject: [TN] SIR test coupons
> 
> Hi Ben and indeed Doug,
> 
> It is important to keep in mind that the original design of what became the IPC B52, was the TB57 developed by NPL. The component selection was ONLY to be those that presented the greatest challenge for residue build-up in and from the overall assembly process.
> 
> What is going on now, as Doug has amply described, is the adoption of ultra-low stand-off components, such as QFN’s that certainly present a huge challenge to those who continue to clean. It is not an issue as to whether the cleaning material can penetrate effectively, rather it is whether the resultant porridge underneath has been adequately removed.
> 
> You might keep in mind that there are many ultra high reliability, safety critical manufacturers who don’t clean, hence do not experience this problem. Of course their production is likely to be very high volume, low mix, unlike Aerospace and Defense that is low, or very low, volume and high mix. Each have their own challenges.
> 
> Also keep in mind that all process chemistries in use today (e.g. solder resist, surface finish, flux, paste, underfills, adhesives, coatings even cleaning solvents etc..) include non-ionic surfactant additives employed for either wetting or de-wetting requirements. Being non-ionic, the old ROSE or indeed IC even FTIR, are incapable, or at least less able, to detect non-ionic species. Yet these do present ECM challenges. How do you know if they are there or not?
> 
> For this very reason, Insulation Resistance measurements are the only practical test method.
> 
> The Insulation Resistance measurement instrument, must be capable of accurately measuring what it sees, no matter whether it is ionic or non-ionic material. The instrument therefore informs you whether you have an ECM reliable example of your intended end product. If it does show unacceptable results, then IC and/or FTIR or more, should be used to try and identify what is present causing you a problem, always presuming that you really want to know!?
> 
> All IR (SIR or CAF) measurement instruments can “do the job”. The B52 is today, our best and most frequently used test coupon employed to determine an ECM reliable end product. You are obligated to employ a representative example of your intended end product, don’t use test coupons produced by anyone other than your intended bare board supplier!
> 
> So, to gather a sufficient amount of usable and reliable data, the test coupon will have multiple test channels, B52 = 16 per coupon. Correct analysis will surely involved at least 16 coupons from each “wet process” assembly stage to help identify where a detected problem might be, or is, occurring. It isn’t the individual process material you will be looking at, it will be the synergistic ECM influences. A measurement instrument must therefore be able to measure a minimum 256 individual test channels: 16 coupons x 16 test channels = 256.
> 
> Fewer channels = inadequate data for examination and comparison. Daisy-chained components = potentially misleading data. 
> 
> Of course, your own supply chain should be providing you with the actual, measured SIR data of their product. It is only when you put your selected material set together that synergistic influences come into play.
> 
> Also, of course you would try to reduce the cost of “screening” intended process chemistries, and this is why the new IPC B53 test coupon will be so helpful to you. It is less expensive!! There will, of course, be options to use even lower cost test coupons such the TB02, although there are many more. Refer to IPC 9201 where Doug and I, plus a cast of thousands, include all the test coupons we both found at that time…2007 - Yikes, 12 years ago!
> 
> As many of you will say, “we always used the B24”. OK, but how representative is that to your current product? The B24 has 400µm track width and 500µm track spacing. I wouldn’t mind guessing that many, if not most of you, are using far narrower track spacings than that, which is but one reason why I conducted a RR test looking at patterns that are: 400µm x 500µm B24; 400µm x 200µm B53; 318µm x 318µm for those insistent upon retaining the old Bellcore patterns, but interdigitated. These are all included on the IPC B53 artwork which, is available from your local IPC store!
> 
> The choice of the test coupon is something that we shall talk about at the SIR 5-32b meeting in APEX. Do come along and help develop the next generation of:
> 
> IPC-TM-650 Method 2.6.3.7
> IPC 9201
> IPC 9202
> IPC 9203
> IPC 9205….
> 
> Oh! By the way, we then have to review CAF testing that is the sub-surface ECM problem in bare board production. That is the B-32e meeting at APEX. 
> 
> Have a Merry Christmas everyone, 2020 is going to be a really exciting year…
> 
> Kindest regards
> Graham Naisbitt
> Gen3 Systems
> Chairman and CEO
> 
> P: +44 (0)12 5252 1500
> E: [log in to unmask]
> W: gen3systems.com
> 
> 
> IPC:
> TAEC Technical Assessment Executive Committee
> 5-30 Cleaning & Coating Committee – Vice Chair
> 5-31 Cleaning & Alternatives – Subcommittee
> 5-32 Cleanliness Assessment Subcommittee
> 5-33 Coating Subcommittee
> 5-32b SIR & ECM – Chair
> 5-32e Conductive Anodic Filament (CAF) – Vice Chair
> 
> Task Group Member:
> 5-22a J-STD-001 Task Group
> 5-23a Circuit Board Solderability
> 5-23b Component Solderability
> 5-24a Flux Specifications
> 5-32a Ion Chromatography / Ionic Contamination 5-32c Bare Board Cleanliness 5-22arr Conformal Coating Application
> 
> IEC TC91 WG2, WG3 & WG10
> Maintenance Leader for
> 
> 61189-5-501 SIR for fluxes
> 61189-5-502 SIR for process characterisation
> 61189-5-504 PICT Process Ionic Contamination Testing
> 61189-5-506 Technical Report on SIR RR testing and the new IPC B53 coupon
> 60068-2-69 Solderability Testing
> 
> ISO – 9455-17 SIR Testing
> 
> 
> 
>> On 20 Dec 2019, at 14:48, Gumpert, Ben <[log in to unmask]> wrote:
>> 
>> Guy,
>> 
>> A couple of points about coupon selection:
>> 
>> 1. The standard coupons are useful for collecting data that you can 
>> compare to industry accepted values. You are essentially demonstrating 
>> the capability of your cleaning process. But you also want the coupon 
>> to be representative of your hardware, and that's where newer coupons 
>> containing those parts that Doug mentioned can be a good choice. 
>> (Keeping in mind that some old parts are still challenging as well!)
>> 
>> 2. When you look for a coupon that is representative of your hardware, the obvious thing to look at are the components included. But you'll want to keep some finer details in mind, such as solder mask configuration. Does the coupon have the same solder mask thickness or pattern as your hardware? On low standoff parts, this can have a large impact.
>> 
>> Ben
>> 
>> -----Original Message-----
>> From: TechNet <[log in to unmask]> On Behalf Of Douglas Pauls
>> Sent: Friday, December 20, 2019 9:36 AM
>> To: [log in to unmask]
>> Subject: EXTERNAL: Re: [TN] [External] [TN] SIR test coupons, 
>> Magnalytix
>> 
>> Good morning Guy,
>> Joe Russeau gave a pretty good summary on the IPC-B-52 test assembly 
>> and
>> 9202 protocols.  The 9202 is wrapping up some of the Rev A work (I hope) but is still focused on the B-52 assembly.  It is a good test vehicle and continues to provide valuable insights into manufacturing processes. The primary criticism is that it was designed in the early 2000s with components common to that time frame.  It does not contain some of the more challenging components like QFNs or mammoth BGAs.  While the B52 does have areas where those components can be designed in, and some people have, but I see a whole new vista of SIR tools coming to the market.
>> 
>> I have worked with Mark McMeen and Mike Bixenman of Magnalytics for about the last 2 years on developing their SIR test system.  I am excited about the system and plan on getting one myself once the development phase is done (which is pretty much now).  We are presenting a paper at Apex on some of  our work to correlate their test card with the B-52.  The Magnalytix card has QFNs, BGA and the B52 QFP80 patterns.  Mark has also developed a series of test cards to look at other aspects of residues and manufacturing processes.  I haven't had this much fun in years.
>> 
>> 
>> *Douglas Pauls *| Principal Materials and Process Engr | Advanced 
>> Operations Engineering
>> 
>> *COLLINS AEROSPACE*
>> 
>> 400 Collins Road NE, MS 108-101, Cedar Rapids, IA  52498  USA
>> 
>> *Tel:* +1 319 295 2109 | *Mobile: *+1 319 431 3773
>> 
>> [log in to unmask]
>> 
>> [log in to unmask] for all Export Compliant Items
>> 
>> 
>> On Thu, Dec 19, 2019 at 2:18 PM Guy Ramsey <[log in to unmask]> wrote:
>> 
>>> We have been exploring the J-STD-001G amendment.
>>> We think we have a "qualified" process because:
>>> 1. We have monitored our process for several years and established an 
>>> action threshold at a level that is much lower than the old IPC limit.
>>> Because with know the "normal" ROSE test result from or Ionograph.
>>> 2. We have never seen evidence of corrosion on assemblies returned 
>>> from the field. And have no reports of corrosion from customers.
>>> 
>>> But, we have not done any SIR testing in years. The mix of parts and 
>>> part densities have changed dramatically. We have no reason to trust 
>>> the old SIR test results.
>>> We would like to gather objective evidence that our cleaning process 
>>> is compliant.
>>> 
>>> Any experience out there with the Magnalytix system?
>>> Any advice about test coupon selection?
>>> Anyone?
>>>