What Are Realistic Guidelines for PCB Flexing?

IPC 600 indicates 0.75% or less bow or twist for PCB assemblies with surface mount components. 0.75% is practically no flex at all and very difficult to achieve. What are the experts views are on board flexing?

F.S.

Expert Panel Responses

Flexing is a very different concept from Bow and Twist.

Bow and twist is essentially the flatness of a board as received. When you place it on a flat surface, how much bow and twist do you see? Do the corners turn upward? Is the board distorted in anyway? This can be caused by material selection in the stack up as well as lamination processing and cool down.

The 0.75% requirement for bow and twist can be difficult to achieve. It requires the design stack up to be balanced from top to bottom and to make sure copper is evenly distributed throughout the board. From a processing standpoint, the manufacturer much be sure to have a lamination stack that keeps the board flat and follows proper cool down in the lamination cycle so that no stresses are induced during lamination. Warped boards can be a real problem to populate, especially if they do not have any flexibility to lay flat.

Flexing is the tendency of a board to deflect when a load is put on it. There is no requirement for flexing. The thicker the board,the less flexing you will typically see. Flexing is also dependent upon the amount of reinforcement in a board. For example, a pure flex circuit will be very flexible as it has no glass reinforcement.

Another variable can be if the PCB is a rigid flex. This will create regions for flex in the middle of a pallet. The bow and twist requirement is limited to individual rigid sections. However, having rigid flex circuits in a pallet may introduce additional difficulties for those who have not had experience populating rigid flex boards.

When flexing is a problem, it is necessary to use fixturing during SMT population to support the parts to ensure successful population. This is common practice for flex circuits and is often used on thin rigid boards.

In the end, many suppliers have learned how to manage board flexibility during population. It is something that usually requires fixtures when the board deflects.

Nick Koop
Product Manager
Minco Products Inc

Nick has been in the circuit industry since 1985, in technical and management roles. Nick is the Vice Chair of the IPC Flexible Circuits Committee, and Chair of IPC 6013 Flexible Circuits Perfomance Subcommittee. Nick holds a Bachelors Degree in Mechanical Engineering and a Masters Degree in the Management of Technology from the University of Minnesota.

Flatness of the PCBs is critical not only for processing but from a reliability point of view. I do not consider this standard is impossible to achieve. If you are not getting this level of flatness than the processing becomes an issue starting with the SMT solder paste deposition, pick and place and reflow.

Due to specific design constraints, some of your boards will not be perfectly flat. However, keeping them flat through the assembly process is a must. Depending on the components populating the board, the flatness % can make the difference between a good functional board and a non-conforming one. And I am talking here about the processes involving dramatic temperature fluctuations (SMT reflow, wave soldering, selective soldering).

Fixtures/pallets can be used for these processes. At the end of the process I recommend a cool off time before removing the assembly from the pallet and move it to the next process.

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.

IPC A-600 applies only to unassembled PWBs. The reason for the reference to SMT boards is that raw PWBs that will have SMT components assembled on them need to be flatter than boards that only get through-hole components inserted. Now, today there are precious few boards that use through-hole only technology, so as a rule, unassembled PWBs need to meet the 0.75% criterion.

During assembly, bow and twist may increase, or in some cases they may actually decrease. What is acceptable is entirely dependent on the design,including the design of the circuit card assembly and the overall product design.

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.

I would like to point out that the amount of warpage given in IPC-A-600 applies at the assembly (CCA) level also. IPC-A-610 section 10.2.6 (Laminate Condition) also references the same maximum warpage of .75% as measured per IPC-TM-650 Method 2.4.22. And just to clarify, the .75% maximum warpage applies from the highest point of deviation from the plane on the bare board or the assembly, and is a measurement of warpage from that point to the nearest edge, not a sum total of all warpage across the entire PWB or CCA. If you read and understand this Test Method (a free download), you will see it is not as stringent as it seems.

Very seldom do I see warpage exceeding the .75% limit as defined in the Test Method. However, anything OVER that limit is definitely an issue due to the fact that if the CCA is clamped in a fixture or cabinet after any soldering has occurred, the flexure stress almost certainly will lead to some level of premature solder joint failure, especially in lead-free assemblies due to the more brittle and less forgiving (lower modulus of elasticity) properties of solder containing little or no lead. The one-time relaxation factor of Sn63Pb37 solder is much greater (and faster) than lead-free alloys. I am not sure if any studies have been done to support the continued use of the .75% maximum for lead-free CCAs, but I would stay well into the safe side of that specification.

Richard D. Stadem
Advanced Engineer/Scientist
General Dynamics

Richard D. Stadem is an advanced engineer/scientist for General Dynamics and is also a consulting engineer for other companies. He has 38 years of engineering experience having worked for Honeywell, ADC, Pemstar (now Benchmark), Analog Technologies, and General Dynamics.

Reader Comment

I would agree with other posters that the 0.75% bow and twist number is a reasonable criteria to pass and most boards we see meet this criteria. If anything, I hear more industry push that this number is too lenient. Of course, this will depend on who you ask. However, at the end of the day it really comes down to the local shape and especially shape change over reliability or reflow temperatures in those surface mount attach areas. Thus the reason for tighter thresholds when having surface mount. While the bow/twist is a good general rule, more practically having an idea of what is happening in your surface mount areas will give a better idea of the board shape will cause you yield or reliability issues.

Neil Hubble, Akrometrix

Submit A Comment

Comments are reviewed prior to posting. You must include your full name to have your comments posted. We will not post your email address.

Free Newsletter Subscription
Circuitnet is built for professionals who bear the responsibility of looking ahead, imagining the future, and preparing for it.

Insert Your Email Address