TECHNET Archives

February 1998

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:
Werner Engelmaier <[log in to unmask]>
Reply To:
TechNet E-Mail Forum.
Date:
Mon, 2 Feb 1998 16:59:06 -0800
Content-Type:
text/plain
Parts/Attachments:
text/plain (179 lines)
ems with the=0APTVs, but it is not a given. A higher Tg certainly helps=
=97and I have been told=0Athat higher Tg materials do NOT carry a cost pr=
emium. A lot will depend on the=0Aquality of your PTVs in terms of platin=
g thickness, plating uniformity,=0Atensile strength and ductility of the =
Cu deposit inside the PTV. You also did=0Anot say whether you have any ma=
nual soldering operations (they are worse than=0AHASL or reflow) as well =
as the severity of the use environment of your=0Aproduct.=0AYou may want =
to take a look at Appendix B in IPC-D-279, 'Design Guidelines for=0ARelia=
ble Surface Mount Technology Printed Board Assemblies' for more=0Ainforma=
tion. =0AOn hole breakout the necessary information is not yet in. some t=
ime ago, there=0Awas an ITRI report and significant discussion on the Tec=
hNet on the subject of=0Anon-functional landswhich is a related subjest m=
atter and which I am sending=0Ayou below:=0AThis is from the IPC News Rel=
eases dated 6/26/97:=0AITRI STUDY: REQUIREMENT FOR ANNULAR RING WASTES MI=
LLIONS=0ANorthbrook, IL 6/26/97=0AThe printed wiring board (PWB) industry=
 loses more than $70 million annually=0Ain boards discarded due to hole-t=
o-land misregistration. The Interconnect=0ATechnology Research Institute =
(ITRI) created an Improved Registration Project=0ATeam to investigate the=
 problem, and has just made the report on its findings=0Aavailable for pu=
rchase.=0AIn its report, PWB Hole-To-Land Misregistration: Causes and Rel=
iability,=0AProject Final Report, the team concluded The requirement for =
an intact annular=0Aring at every innerlayer pad is excessively conservat=
ive. "This study=0Ademonstrates that the requirement prohibiting breakout=
 should be changed to=0Aone that maintains a minimum spacing to adjacent =
circuits," says Marshall=0AAndrews, Chief Executive Officer of ITRI. "Thi=
s change will allow PWB=0Afabricators and their customers to save million=
s of dollars while improving=0Aboard routability."=0AThis conclusion culm=
inates two years of research by ITRI. The project team=0Acreated a PWB te=
st vehicle that was designed to have different degrees of pad=0Abreakout =
from none to just a plated hole in a circuit line with no pad. Hole=0Arel=
iability tests were conducted by thermal cycling the boards from -65 to=
=0A+125 degrees centigrade. No failures due to the line to barrel connect=
ion were=0Afound even after 1000 cycles, 10 times the specification requi=
rement.=0AFor more information on the report, or ordering instructions, c=
ontact Diana=0AHempstead, Administrator, at ITRI at (512) 833-9930, fax q=
uestions to (512)=0A833-9925, or e-mail [log in to unmask] is an=
 industry-led consortium for collaboration among the electronic=0Ainterco=
nnection industry, government and academia. Its purpose is to enhance=0At=
he global competitiveness of the North America electronics interconnectio=
n=0Aindustry by enabling collaboration among its members to accelerate de=
velopment=0Aand application of revolutionary and evolutionary technology.=
=0ASeeking data/reports to address questions on the risks associated with=
 Annular=0ARing size.  In particular, as the mis-registration of the of t=
he mechanical=0Adrill to the PTH Pad moves towards a trace leading into t=
he pad the potential=0Afor an 'open' to occur increases.  Concerns center=
 around 'thin' ring passing=0AET, but in field stresses cause fracture an=
d crack propagation to 'open'=0Astate.=0AThanks, Chris [Chris_Gardini-ECG=
[log in to unmask] (Chris Gardini-ECG005)]=0A=0AChris,=0908/22/97=0AITRI (=
The Interconnect Technology Research Institute) recently published=0Arepo=
rt that addresses your concerns. In essence they found that a PTH without=
=0Aannular ring is as reliable as a pad with a large annular ring.  Sampl=
es were=0Asubmitted from several PWB manufacturers and the thermal cycle =
testing was=0Aperformed by NIST or Crane. Any way I'm sure you can get th=
e report either=0Afrom ITRI in Austin, Texas or from the IPC=0ADavid Ariv=
ett [[log in to unmask] (David Arivett)]=0ACuplex Inc.=0A=0AHi=
 Chris,=0AIf you get the ITRI Report 'PWB Hole-To-Land Misregistration: C=
auses and=0AReliability, Project Final Report", you may be a little caref=
ul in applying=0Athe conclusions. I am not saying that the conclusions ar=
e wrong, but that=0Abased on the test performed, the conclusions may be a=
 stretch. From a=0Areliability point-of-view of the PTH barrel-to-land in=
terconnect, the most=0Asevere loading occurs during excursions to solderi=
ng temperatures not thermal=0Acycling, even from -65 to +125C. That is be=
cause the largest thermal expansion=0Aof the epoxy that causes the radial=
 barrel compression and land rotations=0Aoccurs only at temperatures abov=
e Tg.=0AWerner Engelmaier=0A=0AHi Chris -- I agree with Werner.  The Z-ax=
is stress is the concern which=0Anormally occurs, above the 125C tg, at s=
oldering temperatures, around 260C. A=0Asecond point to consider is the s=
olderability issue. The annular ring aids in=0Athe transfer of heat to th=
e solder joint especially for hand soldering during=0Aassembly or field r=
epairs.=0Aregards,=0AArny Andrade [[log in to unmask]]=0ASandia Labs/=
ret.=0908/25/97=0A=0AHi Werner,=0908/27/97=0AYour comments to Chris were =
interesting and I agree for the most part, but I=0Ahave a question with r=
egard to the comment about radial barrel compression=0Aduring elevated te=
mperatures. It is my understanding that as the board expands=0Ain X & Y t=
hat the hole will also increase in diameter. Due to the=0Adisproportionat=
e difference in CTE I would expect that the barrel would be in=0Atension =
and not in compression (a hoop stress). This phenomena has been=0Ademonst=
rated in the Physics classroom on many occasions with Ring and Ball=0Aapp=
aratus. It's good to see you responding to some of these interesting=0Apr=
oblems showing up on Technet. So what do you think about this analysis?=
=0ALes Connally [[log in to unmask]]=0A=0AHi Les,=0AI don't know=
 about the 'Ring and Ball' experiment, but I do know what happens=0Ainsid=
e a PWB around the PTHs during heating to solder temperatures.=0A(1) At t=
hose temperatures the epoxy is way above its Tg, is very soft, and=0Ashou=
ld be considered a high-viscosity liquid as far as behavior is concerned.=
=0A(2) For analysis as to what is going on around a PTH, the PWB can be=
=0Aconsidered an infinite plane in the x- and y-directions.=0A(3) While t=
he epoxy tries to expand in all directions, it is constrained in=0Athe x,=
y-directions by the glass reinforcements.=0A(4) Thus, the primary expansi=
on is in the z-direction at roughly 3 times the=0Alinear thermal expansio=
n, because all the volumetric expansion is funneled=0Ainto the z-directio=
n.=0A(5) The PTH hole, however, allows the epoxy to also expand into the =
space=0Aoccupied by the PTH if the PTH barrel wall is thin enough to by c=
ompressed.=0AThat is why in cross-sections of PTHs with 1 mil of copper p=
lating and less,=0Aone sees the PTH barrel bowed inward between innerlaye=
r lands, particularly in=0Aprepreg layers (higher resin content, less cur=
e) and when there are larger=0Adistances between lands (no non-functional=
 lands), and sometimes a space=0Abetween the barrel wall and the receded =
epoxy. If there were no copper=0Aplating, the hole diameter would decreas=
e on heating.=0A(6) Thus, the PTH barrel in tension in the z-direction, w=
ith the largest=0Astresses near the PWB center, the PTH barrel experience=
s compressive hoop=0Astresses, innerlayer lands experience tensile stress=
es at their attachment to=0Athe PTH barrel, and innerlayer lands near the=
 PWB surface experience bending=0Amoments due to land rotation resulting =
from the z-direction expansion of the=0Aepoxy surrounding the PTH barrel.=
=0AWerner Engelmaier=0A=0AHi Ken & Richard,=0908/29/97=0AThis has been a =
long ongoing debate. =0AThe work done for IPC-TR-579, 'Round Robin Reliab=
ility Evaluation of Small=0ADiameter Plated Through Holes in Printed Wiri=
ng Boards', indicated a small=0Aimprovement in reliability with the prese=
nce of functional and non-functional=0Ainnerlayer lands. This was NOT a m=
ajor finding of this study, nor was it a=0Arecommendation for the use of =
non-functional innerlayer lands. =0AThe results of the just released ITRI=
 Report, 'PWB Hole-to-Land=0AMiregistration: Causes and Reliability', giv=
e the same indication (Note:=0Aessentially all the failures come from one=
 PWB manufacturer, primarily from=0APTV barrel cracking, before the test =
was stopped at about 1100 cycles; the=0Aother manufacturers show some int=
erconnect separations of the innerlayer trace=0Afrom the barrel). The 199=
7 ITRI report also shows the vast improvement in the=0Aquality of PTV str=
uctures since the 1988 IPC-TR-579 study. =0AThe ITRI study was done with =
temperature cycling from -65<->+125C; thus, the=0Areliability results dir=
ectly, on an accelerated basis, relate to the thermal=0Acycling PWBs see =
in product operation. However, because no excursions were=0Ataken to sold=
ering temperatures, where because of being above Tg of the epoxy=0Asignif=
icantly more thermal expansion mismatch occurs and the epoxy is much=0Aso=
fter, the results can not be assumed to be representative of what might=
=0Ahappen during PWB assembly. It is during solder temperature excursions=
 that=0Athe compressive hydro-static hoop stresses act on the PTV barrel =
and that=0Asurface and innerlayer land rotations occur. These represent q=
uite different=0Aloading conditions than -65<->+125C temperature cycling,=
 and may produce=0Adifferent conclusions.=0ACONCLUSIONS: Non-functional l=
ands are not likely worth having, because the=0Asmall potential reliabili=
ty gain for high-reliability PTV structures for most=0Aapplications is no=
t worth the manufacturing difficulties and costs that have=0Abeen encount=
ered. Misregistration and the absence of annular rings are no=0Asignifica=
nt reliability threat to PTV barrels. Assuming, that the behavior=0Adurin=
g solder temperature excursions is not dramatically different and the PTV=
=0Ainterconnect structure survives assembly, misregistration and the abse=
nce of=0Aannular rings are no significant reliability threat to PTV inter=
connects=0A(Interconnects can be protected from large loads by having a 9=
0 degree trace=0Ajog just before the PTV, i.e. do not approach the PTV wi=
th a long straight=0Atrace).=0AWerner Engelmaier=0A=0AWerner Engelmaier=
=0AEngelmaier Associates, L.C.=0AElectronic Packaging, Interconnection an=
d Reliability Consulting=0A7 Jasmine Run=0AOrmond Beach, FL  32174  USA=
=0APhone: 904-437-8747, Fax: 904-437-8737=0AE-mail: [log in to unmask]

################################################################
TechNet E-Mail Forum 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://jefry.ipc.org/forum.htm) for additional information.
For the technical support contact Dmitriy Sklyar at [log in to unmask] or 847-509-9700 ext.311
################################################################

################################################################
TechNet E-Mail Forum 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://jefry.ipc.org/forum.htm) for additional information.
For the technical support contact Dmitriy Sklyar at [log in to unmask] or 847-509-9700 ext.311
################################################################

################################################################
TechNet E-Mail Forum 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://jefry.ipc.org/forum.htm) for additional information.
For the technical support contact Dmitriy Sklyar at [log in to unmask] or 847-509-9700 ext.311
################################################################


ATOM RSS1 RSS2