TECHNET Archives

April 2009

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 <[log in to unmask]>, [log in to unmask]
Date:
Fri, 24 Apr 2009 17:32:43 -0400
Content-Type:
text/plain
Parts/Attachments:
text/plain (377 lines)
Kevin,
The answer is obvious---take my workshops (see www.engelmaier.com); a less preferred option (from my perspective) is to get a copy of IPC-D-279.

Werner


-----Original Message-----
From: Kevin Glidden <[log in to unmask]>
To: [log in to unmask]
Sent: Fri, 24 Apr 2009 9:12 pm
Subject: Re: [TN] solder fatigue analysis for slowly changing thermal environment



I apologize to all for the thread hi-jack, but it seem's pertinent: So where
ould someone less experienced look to learn more about
redicting/calculating SJ life due to fatigue/creep etc via thermal cycling?
Thanks,
Kevin Glidden
uminescent Systems Inc.

----Original Message-----
rom: Werner Engelmaier /* [mailto:[log in to unmask]] 
ent: Friday, April 24, 2009 9:50 AM
o: [log in to unmask]
ubject: Re: [TN] solder fatigue analysis for slowly changing thermal
nvironment
Hi John,
ou need to consider that solder creep-fatigue is not a stress issue but a
train issue; and that creep is a time-dependent process. Thus, the slower
he transition the more complete the creep process the larger the fatigue
amage.
our worst case scenario are cycles of -50 to +50 giving you a maximum 100C
elta-T; anything else will give you longer life/higher reliability.
he easiest way to deal with all of this is calculate the the mean fatigue
ives for all the cycles involved; that tells you if you have a problem
eliability-wise in 50 cycles of any of them---you would realy have to have
 very large CTE-mismatch for 50 cycles to be a probem.
Werner
 
-----Original Message-----
rom: Nieznansk
i, John A - SSD <[log in to unmask]>
o: [log in to unmask] <[log in to unmask]>; [log in to unmask]
[log in to unmask]>
ent: Thu, 23 Apr 2009 6:51 pm
ubject: RE: [TN] solder fatigue analysis for slowly changing thermal
nvironment











i Werner,



hanks for responding. Sorry for the confusion. I’ll try to clarify further.
1] and [2] are the limiting cases: i.e., [1] = HOT case  = +50C Heat Sink,
2] = COLD case = -50C Heat Sink.. They do not occur sequentially as implied
y [4]. The daily heat sink temperature drift [3] = 10C is a worst case
alue.

 

ssume when the part is OFF it is at the heat sink temperature and when ON
t is 20C above the heat sink temperature.



or the example20given, there are only 50 operational cycles in total. An
perational cycle is defined as 30 minutes ON + 30 minutes OFF  at some
teady-state heat sink temperature between [1] = HOT case and [2] = COLD
ase.



he 50 operational cycles do not occur periodically or at predictable
emperatures or time intervals, but can be separated in both time  and
emperature. In between operational cycles, the system is OFF but can slowly
rift in temperature <=10C per day.  So going from the HOT case heat sink to
he COLD case heat sink will take a minimum of 100C delta / 10C per day = 10
ays.



ith this much variability, I want to try to bound the upper/lower fatigue
amage calculations and perhaps estimate the expected fatigue  (if
ossible). 



t seems a 
reasonable high damage estimate can be calculated assuming 50
onsecutive operational cycles in the HOT case: hold heat sink  steady at
50C, complete 50 operational cycles (each operational cycle = 30 minutes ON
 30 minutes OFF), and ignore the 10C daily temperature drifts. Do you
gree? A variant here which will be more damaging yet is to allow more OFF
ime between operational  cycles while in the HOT case (perhaps days). Do
ou agree?



t seems a reasonable low damage estimate can be calculated assuming 50
onsecutive operational cycles in the COLD case: hold heat sink  steady at
50C, complete 50 operational cycles (each operational cycle = 30 minutes ON
 30 minutes OFF) , and ignore the 10C daily temperature drifts. Do you
gree? A variant here which will be more damaging yet is to allow more OFF
ime between operational  cycles while in the COLD case (perhaps days). Do
ou agree?



 think the other case of interest from a fatigue analysis standpoint is as
ollows: The system starts out at the upper temperature limit,  runs ONE
perational cycle (30 minutes ON + 30 minutes OFF) while heat sink is
OT=+50C, shuts off, cools down over 10 days to the lower temperature limit,
uns ONE operational cycle (30 minutes ON + 30 minutes OFF) while heat sink
s COLD=-50C, heats up over  10 days to the upper temperature limit where
he cycle repeats. This would give me a total of 25 cycles while heat sink
s HOT and 25 cycles while heat sink is COLD.  



o there are quite a20few variables here to consider. Any confirmation or
epudiation of these ideas or other insights you can offer up  are greatly
ppreciated. I recognize that the +50C/-50C range is in the realm of “Large
emperature Excursions” where additional damage mechanisms may apply.

 

hanks again.



ohn Nieznanski













From:
[log in to unmask] [mailto:[log in to unmask]]

ent: Thursday, April 23, 2009 2:44 PM
To: [log in to unmask];
ieznanski, John A - SSD
Subject: Re: [TN] solder fatigue analysis for slowly changing thermal
nvironment





i John,
0
Form your email it is not clear [at least to me] what is happening.
How many cycles [1.]? How many cycles[2.]? How does your assembly go from
1.] to [2.]?



ycle [1.] is much more damaging than cycle [2.] because of the higher mean
yclic temperature.
The cycle described in [4.] is even more damaging because of the much larger
elta-T.
Fatigue analysis is based on the assumption [and there is very strong
vidence that it is correct—Palmgren-Miner's Rule] that fatigue damage
ccurring at a given cycles is additive to any damage occurring at other
ycles.
So, in order to analyze the fatigue life of anything including solder
oints, you first must determine the cyclic loading history to be expected,
alculate the fraction of life consumed for each cycle type, and add it up.

Regards,
Werner Engelmaier
Engelmaier Associates, L.C.
Electronic Packaging, Interconnection and Reliability Consulting
7 Jasmine Run
Ormond Beach,
L 32174
SA
Phone:=2
0386-437-8747, Cell: 386-316-5904
E-mail: [log in to unmask], Website: www.engelmaier.com






----Original Message-----
From: Nieznanski, John A - SSD <[log in to unmask]>
To: [log in to unmask]
Sent: Wed, 22 Apr 2009 2:11 pm
Subject: [TN] solder fatigue analysis for slowly changing thermal
nvironment


ello TechNet Gurus,






Using the classical IPC-D-279 methods, how would you determine SMT solder
oint 



atigue for the following situation?







1.  SMT part turns on at upper system temperature limit (Thi= +50C), runs
or 



0 minutes at Thi+ = +70C, shuts off, part stabilizes at Thi= +50C for 30 



inutes.



2.  SMT part turns on at lower system temperature limit (Tlo= -50C), runs
or 



0 minutes at Tlo+ = -30C, shuts off, part stabilizes at Tlo= -50C for 30 



inutes.



3.  And now the twist, the system temperature can slowly vary between these



limits as much as 10C in a 24-hour period.



4.  How much fatigue develops after 50 thermal cycles between T-hi and
-lo?






Can I claim that the upper limit on fatigue can be calculated simply by
unning 



0 operational cycles from the upper temperature limit (Thi= +50C to Thi+ = 



70C)?






Can I claim that the lower limit on fatigue can be calculated simply by
unning 



0 operational cycles from the lower temperature limit (Tlo= -50C to Tlo+ = 



30C)?






Can I claim that the actual fatigue is somewher
e between these two limiting 



ases (worst case, best case)?



ohn










This e-mail and any files transmitted with it may be proprietary and are
ntended solely for the use of the individual or entity to whom they are
ddressed. If you  have received this e-mail in error please notify the
ender.
Please note that any views or opinions presented in this e-mail are solely
hose of the author and do not necessarily represent those of ITT
orporation. The recipient should check this e-mail and any attachments for
he presence of viruses. ITT accepts no  liability for any damage caused by
ny virus transmitted by this e-mail.

 

--------------------------------------------------
echnet Mail List provided as a service by IPC using LISTSERV 15.0 To
nsubscribe, send a message to [log in to unmask] with following text in the
ODY (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
OMAIL or (MAIL) To receive ONE mailing per day of all the posts: send
-mail to [log in to unmask]: SET Technet Digest Search the archives of
revious posts at: http://listserv.ipc.org/archives Please visit IPC web
ite http://www.ipc.org/contentpage.asp?Pageid=4.3.16 for additional
nformation, or contact Keach Sasamori at [log in to unmask] or 847-615-7100
xt.2815
----------------------------------------------------
---------------------------------------------------
echnet Mail List provided as a service by IPC using LISTSERV 15.0
o unsubscribe, send a message to [log in to unmask] with fol
lowing text in
he BODY (NOT the subject field): SIGNOFF Technet
o temporarily halt or (re-start) delivery of Technet send e-mail to 
[log in to unmask]: SET Technet NOMAIL or (MAIL)
o receive ONE mailing per day of all the posts: send e-mail to 
[log in to unmask]: SET Technet Digest
earch the archives of previous posts at: http://listserv.ipc.org/archives
lease visit IPC web site http://www.ipc.org/contentpage.asp?Pageid=4.3.16 for 
dditional information, or contact Keach Sasamori at [log in to unmask] or 
47-615-7100 ext.2815
----------------------------------------------------


---------------------------------------------------
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