terça-feira, 31 de março de 2015
terça-feira, 10 de março de 2015
Problems with Insufficient Barrel Fill
 |
|
We
are seeing insufficient barrel fill (60-75%) during lead free wave
soldering. Can you point to some reasons why we may be seeing this
insufficient
condition and suggest a cure?
|
|
R. S.
|
|
Experts Comments
|
|
There are a number of factors that often combine to cause this problem.
First, there is usually a significant temperature gradient between the bottom of the board and the top side. Since solder likes heat, the solder is reluctant to flow up the barrel to a cold region. This can be determined by instrumenting the top of the barrel with a thermocouple during the wave soldering process. A top side heater can be added, but even then, it is not possible to eliminate the temperature difference completely. Second, internal ground planes connecting to the barrel act as a heat sink, preventing those pins from ever achieving a high enough temperature to reflow sufficiently. Thermal relief pad design for ground pins should be considered, as long as the electical performance of the grounding system is not compromised. The thicker the board, the more difficult it is to achieve adequate barrel fill sinc the issues mentioned previously are even more pronounced. Very thick boards, such as server boards, represent the ultimate challenge. Third, insufficient or inappropriate flux. Usually for through hole soldering, the barrel diameter is enlarged somewhat to facilitate flux spraying prior to wave soldering. Depending on the surface finish of the board, OSP or HASL, different fluxes with various levels of activity are available to address particular needs. One of the current trends occurring in the industry now is the elimination of wave soldering by use of pin in paste reflow techniques. In this approach, the connectors and other through hole devices are sourced as lead free reflow capable. Solder paste is printed on the through hole pad, and the through hole devices are reflow soldered at the same time as the SMT components. If the solder volume is insufficient to fill the hole, a solder preform from a tape and reel package can be automatically placed in the solder paste to augment the solder volume provided by the paste. Solder preforms in tape and reel packaging are available in standard sizes such as 0805, 0603 and 0402. All common alloys are supported, including SAC verions, and SnPb versions. Another trend is to solder the high density SMT devices on one side using typical SAC alloy solder paste, and the use a low temperature solder paste, i.e. SnBiAg to solder all the through hole components in a second reflow step on the other side of the board. The SnBiAg solder paste reflows at 138C, which is lower than SnPb, thus ensuring that all the existing through hole devices can go through the reflow oven. The low temperature of the second pass does not disturb the SAC solder. |
|
Paul J. Koep
Global Product Manager Alpha Mr. Koep is responsible for product planning and technical marketing for the Preform Products at Alpha. He is the co-author of several patents in the areas of soldering applications focusing on reflow and alternative methods. |
|
Possible causes of insufficient fillings are:
|
|
|
|
|
Insufficient barrel fill can be caused by many aspects in a wave soldering process.
Some of these can be machine parameters such as poor flux penetration through the barrel, insufficient preheat, excessive preheat, insufficient dwell time and solder temperature itself. This can also be caused or effected by assembly related issues such as tight hole to lead ratios, board thickness, pallet designs, and no thermal reliefs when needed. The above is just to name a few as the question is complex. In answering your question, let's look at the machine related parameters you have at hand that maybe able to help you. 1) Fluxer/flux. Make sure you are within the flux manufactures application specifications when using their flux. Beyond the recommended specifications for amount and thermal aspects is to make sure you are penetrating the barrel. A simple visual method to determine this is to use thermal fax paper. This type of fax paper can be used on alcohol based fluxes. If you are using VOC free you will need to acquire pH paper. The preferred method is to sandwich the paper between two bear boards to prevent lifting of the paper. You can however also tape the paper to the topside of a bear board. It is also preferred that you use the assembly you are having difficulty with. Run the assembly through the fluxer only, do not allow this into the preheat section. You may then quickly remove the paper from the board. You should see a clear defined imprint across all barrels of the assembly. The imprint should be very well defined. Also look at small via's as a process indicator. There are also test vehicles available in the industry. For example, ECD has a product called Flux-O-Meter. Care must also be taken to insure that excessive flux is not being applied to the assembly. Follow the flux manufactures recommendations for amount. most likely this will be between 850-1500 micrograms per square inch of the assembly. If the surface coating of the flux is too much the micro-droplets can not penetrate the barrel as excessive flux on the surface prevents this. This testing tool is just a simple and quick method. It's not intended to be anything more than a visual tool. 2) Considering you have completed section one, move on to the preheat. Again, use the recommended specifications from the manufacture. Most likely this will outline topside laminate temps to be between 210-240 deg F at the exit point of preheat. Consideration for any bottomside SMT being exposed to the wave should also be considered as to not thermally shock any components. Additional care should be taken to not over expose the flux to excessive time in the preheat section, a common mistake when soldering with lead free. Other process indicators of the flux being thermally over exposed will be solder webbing on the assembly mask, large globular icicles, excessive solder and bridging with what appears to be a rough surface. The best way to confirm you profile is to use a thermal profiler with thermocouples attached to both the top and bottomside of the assembly. 3) Wave contact time and immersion depth relates to the assembly thickness. On average you will want to achieve 4-7 seconds of contact time for your average .062" thick assembly. Depending on the type of system you are using care should be used with the use of a chipwave. Some older wave soldering systems had large gaps between the chipwave and the main wave. This gap can cause a lead free alloy to solidify between the two waves. If the gap is in excess of 3-4 inches and your process requires the use of a chipwave you will need to contact the equipment manufacture to see if they have a solution for this. When trying to trouble shoot contact time, some systems offer the user the ability to change the conveyor speed while the assembly is in the conveyor. Considering that the fluxer and preheat parameters are related to the conveyor speed you maybe able to hold the conveyor speed during fluxing and preheating and then just increase or reduce your conveyor speed at the wave. This might save you time in determining the best contact time needed. Of coarse you would then go back to the fluxer and the preheat and adjust those processes accordingly. 4) Alloy type being used. In historical controlled testing done in the past, some alloys have demonstrated a higher wetting ability than others in regards to barrel fill. This is especially true on thicker assemblies in excess of .093". I hope this is helpful, Good luck |
|
|
John Norton
Eastern Manager Vitronics Soltec John Norton started his soldering career in 1983 for Hollis Engineering. He has also worked with Electrovert as a technical training manager and Vitronics Soltec for the last ten years. He has held various technical development and sales positions. |
|
|
Lead free alloys wet more slowly than do tin led
alloys, and I suspect that this is the problem, the thicker the board
the greater the problem. The problem can also be exacerbated by the pad
finish especially if you are using a tin lead
OSP. The best thing to do is to increase the preheat temperature (the extent that you can do this will be dependent on the flux that you are using) and to increase the time on the wave (again the extent you can do this will depend on the flux). For 0.093 thick boards I have seen contact times of up to 9 seconds. |
|
|
Neil Poole
Senior Applications Chemist Henkel Electronics Dr. Poole is a Senior Applications Chemist in Henkel Technologies, electronics assembly materials application engineering group. He is responsible for all of Henkel's assembly products including soldering products, underfills, PCB protection materials, and thermally conductive adhesives. |
|
|
As with tin/lead wave soldering lack of barrel
fill is almost always heat related although poorhole solderability is a
possibility. More preheat and/or more time on the wave is probably the
answer.
Are the holes connected to ground / earth planes or heavy mass components? This would be another symptom of this being related to the thermal mass. |
|
Bryan Kerr
Principal Engineer - CMA Lab BAE Systems Bryan Kerr has 35 years experience in providing technical support to PEC assembly manufacturing. His experience ranges from analysis of materials and components to troubleshooting and optimizing, selecting reflow, cleaning, coating and other associated processes. |
|
|
Good soldering needs 3 things: Solder, a "wetable" surface(s), and heat. I suspect you have the solder, so that's not it. "Wetable" surfaces are those that will wet with solder and are free from contamination. I assume the component leads and PCB barrel are of wetable materials like tin or copper (there are others like gold, silver, nickel, etc). This leaves oxides and other contamination which your flux is designed to remove, provided the flux reaches in the barrel and to the top side of the board. You should confirm this. Finally, heat or temperature. Solder will not wet a surface that is not hot enough. You need to run a thermal profile of the area where the barrels do not fill, measuring the top side of the board to make sure it's hot enough for your solder alloy. If it is not, the wetable, flux cleaned, surfaces will not draw the solder up through the barrel because it's simply is not at the right temperature. |
|
Paul Austen
Senior Project Engineer Electronic Controls Design Inc Paul been with Electronic Controls Design Inc. (ECD) in Milwaukie, Oregon for over 34 years as a Senior Project Engineer. He has seen and worked with the electronic manufacturing industry from many points of view, including: technician, designer, manufacture, and customer. His focus has been the design and application of thermal process measurement tools used to improve manufacturing processes like: mass reflow and wave soldering, bread baking, paint and powder curing, metal heat treatment and more. |
|
|
There are several reasons for insufficient fill of
the PTH during wave soldering. The major reason is poor wetting that
can be due to oxidation on the leads, oxidation in the barrel, not
enough flux, etc.
One way to increase the barrel fill is to use an inert atmosphere which will increase the wetting of the solder into the barrel and in turn allow your current flux chemistry to work more efficiently. You can also use a more aggressive flux which will require intensive cleaning post wave. Your PTH plating maybe poor, however this can be reviewed with your board vendor to make sure that your specs are being adhered to. |
|
|
Gregory Arslanian
Global Segment Manager Air Products & Chemicals, Inc. Mr. Arslanian has been involved in electronics packaging processing and equipment since 1981 including flipchip, TAB, wirebonding and die attach. Current responsiblities include R&D, applications, marketing and customer interaction. |
|
|
George Oxx and I published the results of testing
we performed on Lead-Free Intrusive Reflow which showed good barrel
fill. This was published in SMT magazine Nov/Dec 2007 and titled "Intrusive
Reflow of Lead-Free Solder Paste". |
|
Bill Coleman
Vice President Technology Photo Stencil For over 18 years, Dr. Coleman has been the vice president of technology for Photo Stencil, working closely with customers to understand their printing requirements. His efforts have resulted in several new stencil products. |
|
|
Comment: Check solder bath
time and temperature and ensure you use a flow accelerator to improve
the solders ability to flow well. I would always use a Lead Free Solder
at 275C pot temperature and
generally have an immersion time of 2-3 seconds (yes the same as
LEADED). Often extended (over 4 seconds dwell time) will kill most
fluxes. Look for a typical top board temperature of 105 - 120C. If the
solder is a yellow/Gold discoloration and sluggish then
please contact us
|
|
|
Greg York
Technical Sales Manager BLT Circuit Services Ltd Greg York has twenty two years of service in Electronics industry. York has installed over 350 Lead Free Lines in Europe with Solder and flux systems as well as Technical Support on SMT lines and trouble shooting. |
|
|
One of CTQ aspect if not
covered are under initial DFM - PTH connected Layer count,weight (oz).
We have proved with derived formula with 90%+ success rate on initial
assessment of lower fillet height
on this.
Thermal value PTH= No of spokes X width of spokes X Layer Thickness if greater than 800mil.sqr will sure have problem of fillet height in wave soldering in wave optimized spec. >1500mil.sqr can trigger problem even in selective nozzle type wave soldering. |
|
Subrat Prajapati
Supplier Quality Leader Ge Healthcare Subrat has 10 year of extensive experience in PCB assembly process optimizing for quality, process includes screen printing, wave, reflow. He has a copyright in stencil design published in Apex Expo2010 at Las Vegas US. |
|
|
Reader Comment
There are two main causes for insufficient hole fill
Process
optimization will help only 10%-15% in hole fill. PCB PTH design
variables: P2H ratio, Thermal mass and thermal spokes design. If we
take care of above
mentioned three at DFM stage, will help us to reach >90% top side
Hole fill without any issue. (Required PCBA Manufacture friendly design
and manufacturing engineer should ask designer to follow those rule)
Mallikarjun, Jabil, India
|
|
|
Reader Comment
All
good comments. One more, check for solder mask in the thru-holes.
Masking can bleed into the holes preventing the solder from flowing up
the holes completely.
Jerry Wiatrowski, General Dynamics
|
|
|
Reader Comment
I
would to add two specific points to this great discussion. Did you
check by x-section analysis on PCB plated hole for some samples to see
if you have a proper
copper plating thickness (or copper hole wall for OSP) inside the hole
to be filled by solder?
I faced a similar situation times ago and, for my surprise, after do x-section in three or four samples of not used PCBs, the hole walls had not enough thickness or, in some cases, simply the hole copper wall did not exist! I do not know who your PCB supplier is, but you should consider this too. In the case you have a PCB quality issue, as it was supplied, you will not see the problem in an x-ray image, since most of times you cannot see the details of copper/plated walls in a x-ray image. Second, since 2006 when IPC released IPC-610D, including lead-free aspects for soldering inspection and solder acceptance, this subject is controversial. Now we have IPC-610F and, in my opinion, the IPC-610 committee was not still able to make barrel filling requirements fully clear so far.
Glayson Figueiredo, Philips Medical Systems, Brazil
|
Reliability of Automated Soldering vs. Hand Soldering
 |
|
On
occasion, we have had to place a skip in our SMT placement machine due
to unavailability of a surface mount component. The intention is to
build
the circuit card assembly and then hand place and solder the missing
components when available. I accept that there are times when you have
to rework/touch-up/replace components, but these should be kept to a
minimum.
Can you comment on the potential differences in reliability for SMT machine placement followed by reflow soldering vs. hand placement followed by hand soldering of surface mount components? |
|
P.K.
|
|
Experts Comments
|
|
There are three main
sources of differences in reliability: solder joint behavior, component
damage, and cleanliness. Hand soldering can involve tool tip
temperatures of 400C and ramp rates of more
than 100 C per second, and we need to consider the effect on the
component. Multi-layer chip capacitors, particularly larger ones, are a
high risk for damage during hand soldering, for instance. If you have
guidelines for rework that address the acceptable
processes for specific components, you can rely on these guidelines for
the hand-installation of the components.
The differences in solder joint reliability are hard to predict, and in most cases not your biggest concern, as long as you meet the applicable standards for soldering quality. Speaking of soldering quality, remember that your defect rate for hand-assembled product will almost always be higher than for automated assembly. The cleanliness risk is real, but as long as you have well-developed processes for hand soldering that include ensuring cleanliness of finished assemblies, you should be in good shape. In addition to the risk to the components being soldered, there are additional product risks, such as the risks of ESD or mechanical damage during the additional handling and processing necessary. It might not be easy, or even possible, to trace failures related to these additional risks back to the manual assembly process. In general, we try to avoid the "build short" practice as much as possible. The added cost of the manual labor and the reliability and quality risks are rarely worth it. |
|
Fritz Byle
Process Engineer Astronautics Fritz's career in electronics manufacturing has included diverse engineering roles including PWB fabrication, thick film print & fire, SMT and wave/selective solder process engineering, and electronics materials development and marketing. Fritz's educational background is in mechanical engineering with an emphasis on materials science. Design of Experiments (DoE) techniques have been an area of independent study. Fritz has published over a dozen papers at various industry conferences. |
|
|
Hand-soldering of SMD's
can be as reliable as an SMT reflow process but some care is required.
The important thing to remember is to avoid excessive temperatures and
contact times with the soldering
tip.
The area of concern is components which cannot tolerate fast ramp rates and are prone to cracking or delamination issues. Hand-soldering heats parts up very rapidly and within a second or two the temperature is well above the melting temperature of the alloy. Reflow soldering slowly increases temperature and reduces the effects of CTE mismatches of materials as well as thermal shock issues. It is important to be aware of the components ability to withstand higher temperatures and also ramp rate requirements as to avoid issues. The other point is to avoid heating the whole component but soldering the terminations only, this requires the use of the correct tip geometry. |
|
Peter Biocca
Senior Market Development Engineer Kester Mr. Biocca is a chemist with 24 years experience in soldering technologies. He has presented around the world in matters relating to process optimization and assembly. He has been working with lead-free for over 8 years. He is the author of many technical papers delivered globally. |
|
|
A process is only as good
as its repeatability. Hand soldering is not a repeatable process. Oxides
on the soldering iron tip or on the solder, pressure applied, heated
contact area, flux application
and operator skill can all have a significant impact on the finished
solder joint. One can have a very lovely solder joint on the outside but
a bad joint on the inside. The interface between the bulk solder and
the solder land or the soldered component lead
can have a dramatic effect on the reliability of the finished solder
joint.
Soldering results in an intermetallic compound (IMC) layer between the solder and the soldered object. The IMC layer in a SAC solder joint is composed of tin, silver and copper plus some of the solder wetted material. So if the contact surface is copper, excess copper will be part of the IMC along with tin and silver. If the solderable interface is nickel, nickel will be part of the IMC along with tin, copper and silver. IMCs are notably brittle and the thicker the IMC, the more brittle the solder joint. IMC thickness depends on two factors: temperature and time and the higher the temperature or the longer the time of heating, the thicker the IMC. The ultimate goal of any soldering operation is to minimize the IMC in addition to making smooth fillets of the proper configuration as per workmanship standard IPC-610. Hand soldering operations vary from station-to-station (equipment dependent) and operator-to-operator, therefore hand soldering is the least favorable of all assembly methods as it is not a reproducible process. Hand placement of components into wet paste is discouraged as that operation often results in solder bridging/shorts or even displacement of adjacent parts. If only loose parts are available, these parts can be re-taped onto a used reel. As an alternative, a matrix tray can be reused or created by simple machining and the loose parts nested in it for automated pick-up and placement. The ultimate goal is to minimize the number of soldering steps, only use reproducible processes (eliminate hand soldering) and minimize touches to the board to reduce board flexure and solder joint damage. Avoid hand soldering! |
|
Gary Freedman
President Colab Engineering A thirty year veteran of electronics assembly with major OEMs including Digital Equipment Corp., Compaq and Hewlett-Packard. President of Colab Engineering, LLC; a consulting agency specializing in electronics manufacturing, root-cause analysis and manufacturing improvement. Holder of six U.S. process patents. Authored several sections and chapters on circuit assembly for industry handbooks. Wrote a treatise on laser soldering for Laser Institute of America's LIA Handbook of Laser Materials Processing. Diverse background includes significant stints and contributions in electrochemistry, photovoltaics, silicon crystal growth and laser processing prior to entering the world of PCAs. Member of SMTA. Member of the Technical Journal Committee of the Surface Mount Technology Association. |
|
|
It is a well-known fact
that hand soldering is a much less controlled process than the SMT
process. If it handled correctly, hand soldering could introduce some
quality issues.
Here are some tips for hand soldering:
|
|
David Bao
Director New Product Development Metallic Resources, Inc David Bao has more than fifteen years of experience in developing new solder paste, wave soldering fluxes and other SMT consumables. He currently serves as the Director of New Product Development at Metallic Resources Inc. He received a Ph.D. in Chemistry at Oklahoma State University. |
|
|
This process is, of
course, one to avoid. When it happens though, try to keep it to a
minimum. Having SMT components soldered later in the process involves
several risks:
So if you do this,
remember to use a good, clean tip, flux pen (recommended instead of
bottle) and an operator that is very well trained for this type of
operations. Component terminations damage,
lifted pads and burnt boards are just few of the unwanted results
coming out of this practice. To summarize, you can get a good, reliable
solder connection as long as you can accept the potential risks,
increased costs and high probability of scrap.
|
|
Georgian Simion
Engineering and Operations Management Independent Consultant Georgian Simion is an independent consultant with 20+ years in electronics manufacturing engineering and operations. Contact me at georgiansimion@yahoo.com. |
BGA Joint Voids - Accept or Reject?
|
|
|
|
|
Per
IPC-610 acceptability standards, would you accept or reject this BGA
solder joint? Is this level of voids something we should try hard to
reduce?
|
I have heard that some level of voids within a BGA solder joint can actually improve reliability.
Is this true?
E.W.
Experts Comments
|
|
Too many voids at the interface. Single voids in the bulk of the solder play little role in time to failure.
Interfacial voiding can cause significant reductions.
|
|
Dr. Craig D. Hillman
CEO & Managing Partner DfR Solutions Dr. Hillman's specialties include best practices in Design for Reliability, strategies for transitioning to Pb-free, supplier qualification, passive component technology and printed board failure mechanisms. |
|
|
Reject the joint, the
level of voids and where they are would look to be too high, I would
x-ray a component that has not been reflowed to see if the balls are
already failed.
|
|
Richard Boyle
Global Product Champion Henkel Electronics Richard Boyle is a Global Product Champion at Henkel Electronics. He has over 25 years experience in the electronics assembly industry and is responsible for the global technical service of all of Henkel's solder materials. |
|
|
This looks like excessive
voiding. However, without a tool (like X-Ray voiding % check) that will
give you a quantitative result, this is just an impression.
Definitely something that you want to minimize. I do not know about increased reliability but this is obvious sign that the outgassing of the solder in the reflow process is not completed - there is a lot still trapped in the material. It is not uncommon to see voiding especially with lead-free solder that has more outgassing. Check your oven profile - usually a longer soak and a lower reflow temperature provides better results from the voids elimination point of view. |
|
Georgian Simion
Engineering and Operations Management Independent Consultant Georgian Simion is an independent consultant with 20+ years in electronics manufacturing engineering and operations. Contact me at georgiansimion@yahoo.com. |
|
|
Let's first address the
question of IPC-610 acceptability. The acceptability criterion for
voiding in IPC-610 is currently <25% void area on transmission x-ray.
This image is a cross section, which
is not directly comparable to an x-ray, because only the voids at the
plane of section are visible.
When viewing a transmission x-ray image, all voids, regardless of where they lie in the joint, are visible. This image contains 16.7% void area at the plane of section, as determined using a "threshold" image transformation in Adobe Photoshop. That's high. Based on this, I'd expect an x-ray image of the joint to show at least 25% voiding. If (and I emphasize if) that is so, the joint would be a reject per IPC-610 or J-STD-001. Now, let's look at the practical side. The voids in this joint are not localized at the component or PWB interface, but are evenly distributed throughout the joint. Current research seems to indicate that in this situation, the voiding is unlikely to have a negative effect on reliability. It is, however, a process indicator. Were this joint a product of my manufacturing process, I would be investigating the root cause(s) and optimizing the process to reduce the occurrence of the voids. For some additional reading on the matter, look at this article published by IPC, and consider reading the referenced paper. |
|
Fritz Byle
Process Engineer Astronautics Fritz's career in electronics manufacturing has included diverse engineering roles including PWB fabrication, thick film print & fire, SMT and wave/selective solder process engineering, and electronics materials development and marketing. Fritz's educational background is in mechanical engineering with an emphasis on materials science. Design of Experiments (DoE) techniques have been an area of independent study. Fritz has published over a dozen papers at various industry conferences. |
|
|
If this was mine I would
reject as it doesn't appear to have reflowed correctly nor does it
appear to have paste printed on the pad at all.
So my guess looking at the picture is
Hope it helps, I would be
reflowing on a populated PCB at 245C, if you have a bare PCB then
typically set up the profile to 255C to allow for component losses and
density, so this should give you
a safety window. Lastly use T4 powder for dense prints = less voids.
|
|
|
Greg York
Technical Sales Manager BLT Circuit Services Ltd Greg York has twenty two years of service in Electronics industry. York has installed over 350 Lead Free Lines in Europe with Solder and flux systems as well as Technical Support on SMT lines and trouble shooting. |
|
|
Here's my thoughts:
|
|
|
Mike Green
Design Engineering Lockheed Martin Space Systems Mike Green is co-chairman of the IPC Terms and Definitions Committee. He has been working with board design and manufacturing for 33 years. |
|
|
Of all the voids I've
looked at, none have shown this type of characteristics. I would ask how
this solder ball was made and conduct an evaluation of the
solderability of the material to verify the
goodness of the paste and the flux.
Secondly, since this does not look like anything we've ever looked at, it would create some questions in my mind as to its goodness and reliability. Therefore since this is unknown, I would reject this condition, from the perspective that you don't have a homogeneous solder joints, and the joint interface is loaded with planar voids. |
|
Leo Lambert
Vice President, Technical Director EPTAC Corporation At EPTAC Corporation, Mr. Lambert oversees content of course offerings, IPC Certification programs and provides customers with expert consultation in electronics manufacturing, including RoHS/WEEE and lead free issues. Leo is also the IPC General Chairman for the Assembly/Joining Process Committee. |
|
|
This is a difficult call
on a generalization of voids, but I would remind you this is a thin
section of the bigger ball and the amount of voids that can be seen on
the ball to pad interface would
concern me and suggest that the boards with these voids be temperature
cycled to assess the joint integrity and determine if this amount of
voiding has a negative or positive affect on the balls.
|
|
Terry Munson
President/Senior Technical Consultant Foresite Mr. Munson, President and Founder of Foresite, has extensive electronics industry experience applying Ion Chromatography analytical techniques to a wide spectrum of manufacturing applications. |
|
|
What you have heard about voids increasing mechanical reliability in BGA solder joints is not necessarily untrue.
I am familiar with research conducted by Glen Dody of Motorola in the 90's that found voids in the bulk solder would tend to terminate cracks that were propagating. In the section you have shown I would be concerned about the amount of voiding at the inter-metallic compound (IMC) interface. While voids in the bulk solder might stop a crack, the IMC is the weakest mechanical location in the solder joint. Voids along this interface will tend to promote propagation as a separation occurs at the IMC. I would work to reduce voiding in your process. |
|
Stephen Schoppe
President Process Sciences, Inc. Stephen Schoppe is President of Process Sciences, Inc., and has 19 years experience providing SMT services to electronics manufacturers. Stephen provides consulting to several Fortune 500 clients on solder and SMT processes, and is a frequent guest speaker at SMT industry events. |
|
|
This BGA definitely has
too much voiding and needs to be corrected. Whenever this much voiding
forms in a single BGA, it is an indication of a poor solder joint
formation. This could be due to either
weak flux activity or excessive solder powder oxidation. Mixing two
different solder powders also causes this kind of phenomenon.
|
|
David Bao
Director New Product Development Metallic Resources, Inc David Bao has more than fifteen years of experience in developing new solder paste, wave soldering fluxes and other SMT consumables. He currently serves as the Director of New Product Development at Metallic Resources Inc. He received a Ph.D. in Chemistry at Oklahoma State University. |
|
|
For specific BGA void
acceptability recommend to refer IPC 7095 and incorporate same in Shop
floor X-ray machine. This std has specific criteria based on size of
void and numbers to Pass or Fail.
As per picture it looks not meeting IPC 7095 void criteria or having
very low reliability BGA Balls.
Beyond that, can see the ball has not being melted sufficiently and collapsed with a BGA placement shift. |
|
Subrat Prajapati
Supplier Quality Leader Ge Healthcare Subrat has 10 year of extensive experience in PCB assembly process optimizing for quality, process includes screen printing, wave, reflow. He has a copyright in stencil design published in Apex Expo2010 at Las Vegas US. |
|
|
Reader Comment
As
Mike Green noted in his point 5, there seems to be something seriously
strange with the initial solder ball. Have you considered this might be a
"counterfeit
part?" And as noted by others, it appears there was a very low amount
of solder paste available to make good joint fillets (both at the PCB
& Component side). Along with the previously noted lack of
registration and small PCB pad size.
I would also be concerned about such large "notch voids" open to the outer surface of the ball. These would be "stress concentration points" under shear forces and could become crack initiation sites. I am also concerned about the relatively high level of voids along the PCB side interface, especially if this was an ENIG surface finish with it's much weaker and brittle Tin/Nickel IMC interface vs a Tin/Copper based IMC interface (when using HASL, Pb-HASL, ImSn, ImAg or OSP). In regards to the question of some level of voids leading to higher reliability, there is indeed some level of reasonable and potentially related scientific extrapolated truth to a possible answer. There have been test results in the past on IGBT Chip to DBC solder joints that clearly showed that such joints, with a evenly distributed pattern of small voids, had noticeably better power cycle life. The reasonable theory being that the small voids allowed an increased level of, shall we say "flexibility", in the joint and absorbed some of the shear strain caused by the CTE mismatch of the materials during the Power cycle induced temp cycling. Therefore, I believe it would be very reasonable to postulate that an evenly distributed array of small voids in a BGA Solder ball would also result in greater flexibility of the ball, and thus result in greater temp cycling reliability (in say an IPC-9701 2nd level interconnect test regime). Taller balls and solder columns have the same positive result due to increased flexibility (as does the Z-axis thickness of any solder joint). The challenge/problem, of course, is trying to get such evenly distributed and controlled small size voids in a Solder ball (and without getting to many voids at the interface), and preferably with none of the voids exposed on the outside surface. Basically a "closed cell foam solder ball with a full skin". Something interesting to think about.
Steven McLaughlin, ABB, Switzerland
|
BGA Replacement Limit
 | ||||
How many times can a BGA component be replaced at the same location on the same PCB and retain reliability?
We are using BGA components with polymer balls on gold plated multilayer PCB's that are 2 millimeters thick. |
||||
|
M.B.G.
|
|
Experts Comments
|
|
1. A lot of our customers will count the thermal
cycles of the PCB. And based on the PCB material integrity, they
determine at what point the PCB become questionable.
a. Placing and soldering a BGA equates one cycle. b. Removing the BGA is the second cycle c. Cleaning the site of is the third cycle. d. Soldering a NEW BGA is cycle #4 So the question becomes, do you want to go thru these cycles again? Is the thermal cycle for EACH step consistent? The answers may not be a simple 'Yes" or "No", and may include data that supports reliability test data that provides proof that a PCB can withstand X amount of thermal cycles before the material degrades and becomes a liability. 2. Recycling a BGA (reballing) is similar to above. BGA manufactures will normally specify how many times a component can be thermally cycled before the chip within the BGA package is no longer functional, or questionable. Although the solder is a key aspect of successfully soldering a component multiple times, that is only one of three critical areas that need to be considered. The PCB and the component themselves need to be fully understood and should be tested throughout the process in deciding how many times a rework cycle can and should be used. |
|
Neil O'Brien
Sales Director Finetech Neil O'Brien has worked in the field of electronic manufacturing equipment for over fifteen years and is currently Sales Director for Finetech, a manufacturer of precision rework systems and die bonders. |
|
|
There is no hard industry rule but the rule of
thumb for most companies is no more than 5 to 6 thermal cycles at
reflow. The adheasive system for the board pads continues to breakdown
after 3 cycles and the bga part itself will also start
to weaken at 3 cycles so most companies err on the cautious side and
only replace twice at the same location after the initial build which is
normally 2 thermal cycles for top and bottomside reflow thermal cycles.
Then one would have a remove and replace which is 2 more thermal cycles and then possibly another remove and replace which is 2 more for a total of 6 cycles. High reliability hardware normally does not like seeing more than 4. But again this normally customer generated and it differs between companies and the degree of long term reliabity expected. Based on the information given this is a general comment for the number of rework cycles. |
|
Mark McMeen
VP Engineering Services STI Electronics Inc. Mark T. McMeen is STI Electronics Inc.ʼs Vice President of Engineering Services. He oversees the daily operations of the Engineering Services division of STI. He has over 18 years experience in the manufacturing and engineering of PCBs. |
|
|
It depends on many factors; leaded solder will
provide you with additional cycles where a lead-free application will
reduce the number of possible cycles. If you are adding solder paste to
the operation it will help with the reliability
of the finished product. Testing and evaluation of your particular application must be performed to confirm the maximum number of cycles for your particular PCB. |
|
Edward Zamborsky
Regional Sales Manager OK International Inc. Mr. Zamborsky serves as one of OK's technology advisers to the Product Development group. Ed has authored articles and papers on topics such as; Low Volume SMT Assembly, Solder Fume Extraction, SMT Rework, BGA Rework, Lead Free Hand Soldering, Lead Free Visual Inspection and Lead Free Array Rework. |
|
|
I totally agree with Mark McMeens' assertion below
that often manufacturers only get as few as two attempts to get it
right, but this begs a bigger question, "Why am I reworking my BGAs in
the first place?"
I believe the answer more often than not has to do with a poor profile. This year I have seen tremendous interest in this area as manufacturers continue to struggle with micro BGA on more complex boards, which is even further challenging for CMs who get 50 boards from a customer who wants 50 boards back. No drilling holes for thermocouple readings under the BGA! I added some additional comments on this subject along with useful information from TC attachment (non destructive) to how to profile BGAs without destroying other more heat sensitive components. http://profilingguru.com/ |
|
Brian O'Leary
Global Account Manager Indium Corporation Mr. O'Leary is the Global Account Manager for Indium. He has and extensive global network of contacts in the electronics industry with expertise in SMT equipment and processes. |
|
|
Reader Comment
All
depend of the board finished. We are using AgIm and ENIG-Au/Ni we
reworked Ceramic BGA's three times on AgIm boards and five times on
ENIG-Au/Ni finished.
The BGA removal profile is crucial in order to not damage or burn the
PCB lands.
Sergio Ilescas Hernandez, Arris Group de Mexico
|
|
|
Reader Comment
We
have tried the BGA replacement two to three times on the same PCB. And
we found it reliable. The only thing that is to be considered to have
proper removal
of BGA in order not to damage the PCB solder pads.
Maninder Singh, Deltron (A Division of CDIL), India
|
|
|
Reader Comment
At
my company we qualify PCB suppliers in part based on HATS or IST via
reliability results. We specify 6 preconditioning cycles to simulate
assembly processes.
Our reliability test results, then, are only valid up to 6 thermal
cycles.
As Mr. O'Brien points out, a board with one reflow cycle has margin for only one BGA replacement. So we limit our assembly suppliers to 6 thermal cycles at any one site, but only one BGA replacement.
Jimmie King, GE Healthcare
|
Assinar:
Postagens (Atom)