Based on work I have been doing recently the most common reasons for component loss is board flexture and vibration on the reflow process. If parts do not reflow then weight is not an issue. There is a video on double sided reflow and PIHR from the SMART Group. There will shortly be a report on both subjects available from the SMTA in the USA. | David, | There was a thread going on the IPC TechNet about exactly the same thing... | The ratio that was posted is: | Less than, or equal to 30-grams per square inch | Which means if you weighed a part and it is 30-grams, you'll need .033" of surface/pad area for each gram. There were a few comments that said that the 30-gram vs. 1-inch was a little conservative, as well as pointing out that this rule doesn't apply to ceramic or metal packaged devices. | In my experience, almost every standard packaged device that's out on the market will "hang on" during a double sided reflow. Back when I used to work at a memory module company, everything we built was double sided reflow...we didn't own a wave solder machine. | Almost every gull winged device is no sweat, and most "J-leaded" devices up to a plcc 52 can stay put. The only parts I've ever had trouble with are MELF's, some types of SMT crystals and oscillators, and all of those weird RF components with the strange leads and stuff.... | Big thing about double sided reflow is: | 1. EVERYTHING has to be absolutely, perfectly solderable...that PCB's and parts. | 2. You GOT to know what your oven is doing and be intimately familiar with your reflow profile. If you're the type that doesn't like to do plots, then better get ready to change...that is if you want to do double sided reflow successfully. | Good luck! | -Steve Gregory-
reply »