Quoted from D-Gottlieb:The lightweight nature of the parts susceptible to the lead fracture have vibration moments in tune with the shocks transmitted through the playfield during play. While the boards in my spike machine have the zip tie securement from the factory, I took an extra step that I feel will prevent any potential vibration damage to the other transistors and components.
D-G,
The lightweight nature of SMD parts is actually what helps prevent problems vs leaded parts. Also, thermally - IF (capitalized on purpose) done correctly- SMD actually helps because of a typically improved thermal path of the part. I know that wasn't part of the discussion, but it is a side benefit. If the mass is nil, then that helps eliminates the issue since it raises the resonant frequency. They will not be susceptible as much to low frequency vibration and shock. But it's always a good idea to isolate resonances. In other words - if rubber grommets isolate the the boards, then the grommets should eliminate the resonant frequencies (i.e. a coil firing) that cause problems (i.e. you are preventing the parts from vibration in the first place). However, and as Wolf correctly pointed out, the EE and ME designers need to be aware of the frequency that they are trying to eliminate. Physical placement of the grommets is important (are they far apart or close together?). Mounting holes closer together will raise the resonant frequency vs. ones that are spaced far apart which lowers the resonant frequency. If Stern or others are not isolating the PC boards, then I would be worried over the long term. A zip tie can can actually induce a problem vs. shock mounting ony w/rubber grommets since it changes the resonant frequency. Generically speaking, I'd stick with grommets alone, nothing else.
There is absolutely no doubt that SMD parts are fine. But (and that's a big 'BUT'), you need to design the PCB board and mounting hardware to prevent issues. I deal with SMD parts and vibration to an extreme (think in the context of engine vibration related on an airborne platform) without an issue. Far more severe than a pinball machine. But I doubt Stern or other vendors have the capabilities to analyze the scenario correctly. I would think it's more of a 'seat of the pants' design (i.e. it worked before so it must be OK) when it comes to pinball machines.
In my line of work - SMD provides better performance, both electrically and mechanically, than thru hole almost w/o exception. I just can't use thru hole parts and get the same performance...period. But that isn't as much of an issue with a pinball machine.
Apologies for getting wordy - I tried to keep it simple, almost to a fault. While I don't design pinball machines - the rules still apply. In the end it's just frequency vs. isolation at the frequency that can cause an issue. Your comment "The lightweight nature of the parts susceptible to the lead fracture have vibration moments in tune with the shocks transmitted through the playfield during play." is a perfectly worded assessment.
On the solder topic mentioned by others: I still use lead based solder simply due to the fact it is not susceptible to the tin whisker issue over time (i.e think a decade or longer). But I believe consumer things live by different design rules than I deal with, especially with solder.
If I wanted to word the long winded statement above to the bare bones it would be this: If a board resonates at xxxHz, then you design the isolation hardware to eliminate xxxHz - issue solved.