Delamination, unfortunately, is a fairly common problem for modern circuit boards, especially high-density, multi-layer boards or those composed of advanced materials such as PTFE (which is inherently difficult to bond to other layers.) Your boards are probably delaminating because, during processing, they are thermally and/or mechanically stressed beyond their limits of adhesion.

One way to reduce or eliminate this problem would be to increase the bonding strength between all the layers of the board, so that they don't delaminate. This can be accomplished using a suitable plasma treatment on the layers of the board before they are bonded together. For example, increased adhesion (and decreased delamination) can be achieved by treating the individual layers of the board using an Argon-based plasma in a chamber in a low-vacuum environment, prior to the layers being bonded together to form the final board.

The plasma treatment can clean the layers and remove micro-contamination that interferes with bonding strength (this process is sometimes referred to as "plasma cleaning" the layers prior to bonding). The plasma treatment can also activate the surface of the layers before bonding together to make a finished board, so that each layer adheres more strongly to the other layers around it.

Plasma treatments are particularly useful and effective when trying to bond advanced materials like PTFE or various flexible materials together such as polyimide, in order to prevent delamination and increase board quality. Contact March Plasma Systems if you'd like to try some plasma cleaning solutions for your particular case.

There are no obvious answers I can point to here. Its time to start thinking about running a Design of Experiment to track down the problem. Most problems occur within a part, with in alot, or lot to lot. This will involve working closely with the people who are making the boards.

From you questions it is not clear if you are an assembly house, and boards are sent to you, or you are an OEM, and these are boards you built. So again its difficult to give you a reply to your question.

You can also approach the problem, from what causes delamination? Usually its a heat related so I would look for places where the boards may be getting hot. if you cannot use a thermocouple perhaps you could use temperature dots on the boards to see how hot they are getting in specific locations.

So look for hot spots, and try to see if its the boards from one manufacturer, one type of board material or one supplier.

Delamination is generally a result of moisture content, PCB lamination controll problems with Epoxy system and/or inherent stress that is evident in the PCB.

Because the lamination system is suspect, first place to start would be your humidity and moisture control systems.

From here, I would also check your preheating system and ensure you are not surpassing your PCB laminate system temperatures.

Please review your bake out time and temperature for your bare boards. Rule of thumb is 105C for 6 hours and the only caveat is to make sure we do not overtsack the pcb in the bake out chamber.

The pcb should be separated into small stacks to insure the boards reach temperature for atleast 4 hours to insure moisture is driven out of the laminate.

You should verify the temperature profiles of your pcb assy during the reflow process. If they are confirmed within tolerance, show them to the pcb supplier(s), together with the delaminated boards. It should then be the pcb laminates problem.

I'd be very suspicious of defects in either the adhesive or the surface it's being laminated to. The adhesive defects can be due to insufficient adhesive laydown or the existence of gels in the adhesive.

It is also possible that the adhesive composition is non-uniform, which might allow delamination zones to occur. Also, surface contamination on the substrate will result in insufficient wetting between the adhesive and the surface, resulting in delamination. I assume that the lamination pressure was uniform when adhesive was laminated to substrate?

The smallest speck of dust can also cause a focal point for delamination to occur. There are some analytical means available to troubleshoot these ideas, but you need access to a good chemical/analytical lab to do some thermal analysis, and some FTIR spectroscopic analysis.

In this case, before I can get more definitive, I absolutely will want to see samples to do comparisons between "good" and "bad" in the lab.

To find procedures for the proper stocking and moisture protection of electronic components, you can refer to IPC standards. (IPC/JEDEC J-Std-033B.1) This is unfortunately today not so if you are looking for guidance to handle printed circuit boards. And because there are not (yet) any published standards for stocking and moisture protection for printed boards, they're generally overlooked.

Even if you are paying attention to handling and floor time, the packaging quality of PCBs as received from the manufacturers can have a significant effect. Often a simple foil bag or an ESD bag is used instead of a well spec'd Moisture Barrier Bag (MBB) with desiccant and HIC (Humidity Indicator Card).

With such packing, boards are likely to arrive having already absorbed moisture, and stored like this they will be useless after a short time. Desorption process desiccant drying cabinets, such as those pioneered by Totech (http://www.superdry.info/) can be used for their careful drying.

The procedure of is simple and cost-effective. The previously absorbed moisture is removed from the printed board without temperature stress and with a controlled, revertive drying process in an atmosphere of less than 0.5 g/m 3 vapour content (in effect, a moisture vacuum).

In addition, oxidation is stopped by the removal of the electrolytic water molecules. Because these storage systems work at room temperature, boards need not be removed until ready for processing.

Dried and stored in this way even oxidation sensitive OSP coated printed boards are protected and can be used over long periods of time without the risk of delamination and with consistent wetability characteristics.