TECHNET Archives

August 2008

TechNet@IPC.ORG

Options: Use Monospaced Font
Show Text Part by Default
Show All Mail Headers

Message: [<< First] [< Prev] [Next >] [Last >>]
Topic: [<< First] [< Prev] [Next >] [Last >>]
Author: [<< First] [< Prev] [Next >] [Last >>]

Print Reply
Subject:
From:
Reply To:
Date:
Fri, 8 Aug 2008 13:09:11 +0530
Content-Type:
text/plain
Parts/Attachments:
text/plain (235 lines)
Thanks a lot Ian for the inputs. We already did a few of them . Horizontal section for a board per lot / Repeated HAL till it fails / This with baking and without baking. But the issue is we are unable to concretely identify the root cause and to convince my customer.

Typical process characteristics that we follow are as below.
1. High Tg 175 Material : Nelco 4000-6
2. Stress releiving after sheet cutting by baking under weight
3. Press parameters as per Supplier specs.
4. Only Undercut drill bits upto 300 hits used
5. Clad entry instead of Aluminium entry
6. Acid etchback followed by baking and then desmear and glass etch
7. Rate Panel check to ensure the effectiveness of desmear / electroless copper plating
8. Flash plating before photoprinting
9. Electroless - Flash - Electroplating within 24 hours
10. Electroless thickness of only 1-1.2 microns
11. Minimum PTH copper thickness of 30 microns

Can you suggest anything further or any process modifications which you feel will improve the process and the reliability of the board

Rgds

Pradeep


Ian Hanna wrote at Thu Aug 07 23:33:24 IST 2008:
>>A very good question indeed.  Micro-etch is a very very sensitive art.
>>The best bet would be to get a good view of a known good interconnect
>>and compare that to a suspect one.  'As received' meaning after polish
>>and before etch -- I would be suspicious if I saw much of a demarcation
>>line -- but it all depends on how well you polish.  Usually I find this
>>is hidden by the polishing and only brought out during the micro-etch.
>>If you're really lucky you can see both a good connection and smeared
>>connection in the same interconnect.  I have attached a sample that I
>>got.  Incidentally this is one that passed a 10ohm probe test with no
>>problems after...The IPC A-600 really is no help here.  If you have this
>>document, and look the difference is very subtle between interconnect
>>separation and differential micro-etch.  If you look close you will
>>notice that both photos are the same sample - the fail example simply
>>being re-touched to emphasis the thickness of the line.  I pushed IPC on
>>this and got nothing but a vague answer...
>>
>>The rework issue is a little chicken/egg in nature -- which came first?
>>
>>It could be that an underlaying interconnect weakness is being detected
>>by the extra rework cycles -- or it could be that perfectly good boards
>>are being fried.  Again good failure analysis, and an eye for
>>discriminating detail.   Any unusual damage to the reworked area,
>>material distortion, discoloration, etc, and again, some good views of
>>known good product for control samples.
>>
>>  If you have a couple unpopulated boards to spare -- take a few and
>>cycle them through HASL and ET until you get a failure.  One or two
>>passes HASL, Test, Repeat...cut the hole that fails and that should show
>>you your weakest link, and give you an indication of how weak it is.
>>(did it fail in 3 passes or 30 passes?).  If you see a couple of
>>failures due to smear after short cycles, then you've found your
>>problem.  If you don't' see failure after half a dozen passes then it
>>suggests that the cause was something unusal at the repair side of
>>things (or one very bad board...)
>>
>>Personally I would be suspicious of any panels in a lot that showed some
>>smear -- it only takes one bad interconnect to bring a board down.  I
>>always worry about the 'good' boards in a lot that has unusual fall out.
>>
>>Microsection is a very very powerful tool, but the sample size of 1 to 6
>>holes is statistically insignificant when you're looking at 10 to 50,000
>>vias on a board...so be careful.  It's what you don't see that really
>>hurts.
>>
>>(I'm board-shop quality guy -- so I'm pretty used to seeing this from
>>your side of things)
>>
>>hope this helps
>>
>>Ian
>>
>>
>>
>>
>>-----Original Message-----
>>From: mp3 [mailto:[log in to unmask]] 
>>Sent: Thursday, August 07, 2008 1:25 PM
>>To: Ian Hanna
>>Cc: ks
>>Subject: Re: RE: [TN] Smear and rework failure
>>
>>Hi Ian,
>>
>>Thanks for the inputs and the microsection photograph. Seeing your
>>photograph, I have couple of more queries. You have microetched the
>>section as is clear from the picture. How do we differentiate smear and
>>the separation between electroless and plated copper? If the 'as
>>received' mould also shows a uniform demarkation as if it is
>>microetched, does it indicate presence of smear?
>>
>>Our issue is that the failure is found by the customer only in boards
>>which have been reworked. The boards sent to the customer have all been
>>checked for microsection acceptance before despatch. Since of the whole
>>lot, 1 PCB has been rejected inhouse for presence of smear in
>>microsection, the customer concludes that the cause is smear. We do
>>microsection in both horizontal and vertical coupons for microsection
>>analysis. What else can be done to carry out a good failure analysis in
>>this case?
>>
>>Rgds
>>
>>Pradeep
>>
>>Ian Hanna wrote at Thu Aug 07 20:00:23 IST 2008:
>>>Hello Pradeep,
>>>
>>>Unless you have grossly inappropriate drilling parameters, the smear
>>>will show up as a long-term reliability issue and not as an
>>>infant/assembly mortality.  You will need techniques with finer
>>>resolution than 1x solder shock and bbt coninutity and isolation.  I
>>>found that even with very high sensitivity flying probe equipement, I
>>>could not reliability detect interconnects with smear around 270 degree
>>>of the pad.  You will need a method with finer resolution such as IST,
>>>to get a really accurate feel for the interactions of desmear and
>>>etchback.  
>>>
>>>To answer your questions, smear is not usually deposited evenly around
>>>the entire circumference of the hole.
>>>
>>>The presence of smear will reduce the long-term reliability of your
>>>interconnects.  The actually time to failure depends on many factors.
>>>Hand re-work is notorious for inducing stress on local areas of the
>>>board and may case latent failures to surface.  It may also cause
>>>perfectly good boards to blow.  All depends on the nature of the hand
>>>rework and skill of the operator.
>>>
>>>Best plan is to run a good failure analysis of the blown via as well as
>>>some typical vias from the same board.
>>>
>>>Horizontal sectioning would be a good tool...i have attached one of
>>>mine.
>>>
>>>Cheers
>>>
>>>Ian
>>>
>>>
>>>
>>>
>>>
>>>
>>>-----Original Message-----
>>>From: TechNet [mailto:[log in to unmask]] On Behalf Of mp3
>>>Sent: Thursday, August 07, 2008 9:44 AM
>>>To: [log in to unmask]
>>>Subject: [TN] Smear and rework failure
>>>
>>>Dear technetters,
>>>
>>>We are a military and space grade multilayer PCB manufacturing company.
>>>We have been following a comparatively longer desmear for High Tg
>>>materials and also acid etchback to achieve a 3 point contact.
>>>
>>>As a part of an analysis, we moved a 14 layer board without desmear /
>>>etchback directly to electroless. We found a couple of opens in BBT
>>>immediately after etching. Then we did an HAL cycle to give a sort of
>>>thermal shock. We did not find any increase in the number of
>>>discontinuities. This was contrary to our expectations. We believe that
>>>desmear is critical for the reliability. We also found during
>>>microsection that copper layers to be quite shiny but we found a spec
>>in
>>>the inner layer area too. A few queries we have are 
>>>
>>>1. Will the smear be uniformly formed around the hole.
>>>2. Will the presence of smear affect the board while rework
>>>
>>>One of our customers have observed opens in reworked boards at the same
>>>points where it has been reworked. On going thru the traveller card of
>>>the lot, we find 1 PCB to have been rejected at microsection stage for
>>>smear. This was a high reliability board and we have done microsection
>>>analysis on the coupons of every panel and all other panels were found
>>>to have passed the requirements of microsection analysis. The customer
>>>feels that the reworked panels have failed due to the presence of
>>smear.
>>>He is insisting on horizontal microsection analysis. The boards are
>>hand
>>>soldered. Can technetters help us in analysing this problem further. 
>>>
>>>The main question is will a presence of smear result in a failure
>>during
>>>rework. It passes a thermal stress and even 1st or second rework too. 
>>>
>>>Looking forward for your valuable inputs
>>>
>>>Rgds
>>>
>>>Pradeep Menon
>>>
>>>
>>>
>>>-------------------------------
>>>Micropack Ltd, Bangalore, India
>>>
>>>---------------------------------------------------
>>>Technet Mail List provided as a service by IPC using LISTSERV 15.0
>>>To unsubscribe, send a message to [log in to unmask] with following text
>>>in
>>>the BODY (NOT the subject field): SIGNOFF Technet
>>>To temporarily halt or (re-start) delivery of Technet send e-mail to
>>>[log in to unmask]: SET Technet NOMAIL or (MAIL)
>>>To receive ONE mailing per day of all the posts: send e-mail to
>>>[log in to unmask]: SET Technet Digest
>>>Search the archives of previous posts at:
>>>http://listserv.ipc.org/archives
>>>Please visit IPC web site
>>>http://www.ipc.org/contentpage.asp?Pageid=4.3.16 for additional
>>>information, or contact Keach Sasamori at [log in to unmask] or
>>847-615-7100
>>>ext.2815
>>>-----------------------------------------------------
>>>
>>
>>-------------------------------
>>Micropack Ltd, Bangalore, India
>>
>>
>>
>

-------------------------------
Micropack Ltd, Bangalore, India

---------------------------------------------------
Technet Mail List provided as a service by IPC using LISTSERV 15.0
To unsubscribe, send a message to [log in to unmask] with following text in
the BODY (NOT the subject field): SIGNOFF Technet
To temporarily halt or (re-start) delivery of Technet send e-mail to [log in to unmask]: SET Technet NOMAIL or (MAIL)
To receive ONE mailing per day of all the posts: send e-mail to [log in to unmask]: SET Technet Digest
Search the archives of previous posts at: http://listserv.ipc.org/archives
Please visit IPC web site http://www.ipc.org/contentpage.asp?Pageid=4.3.16 for additional information, or contact Keach Sasamori at [log in to unmask] or 847-615-7100 ext.2815
-----------------------------------------------------

ATOM RSS1 RSS2