Sand 9's interdigitated piezo-electric resonator (center) is the world's first MEMS timing solution for cellular phones, due to its electro-mechanical coupling that is 100 times better than traditional MEMS.
(Source: Sand 9)
Another advantage of piezo-electric resonators, when compared to quartz crystals, is that they do not experience what are called "activity dips" according to Sand 9. Quartz crystals have random defects which cause occational glitches in their timing signals that can cause calls to dropped, data packet loses and GPS lock-up. Sand 9 claims their piezo-electric design is immune to activity dips.
I don't understand why this is any better than the *INTEGRATED* silicon oscillators (SiLabs, IDT, et. al.) ... that also do not experience "activity dips" or shock problems. Seems like adding these extra non-standard material combinations (i.e., piezo materials) for the same type of performance is a lot of uncertainty for no real advantage.
I see your point, but Sand 9 claims that their resonator is the only one that meets the stringent performance requirements of cell phone, GPS, etc. You can download white papers that makes their case here:
That's pretty cool. Hope they start offering these devices in frequencies actually used by the standards with the difficult requirements (GSM, GPS, LTE, etc.) so that they can sell into those markets. Also, the integrated CMOS ASIC that they were talking about in 2009: http://www.eetimes.com/document.asp?doc_id=1172309 would be really helpful. These devices leave a lot of implementation to the IC designer with this 'naked' approach.
Sand 9 says they went with the bare resonator first because cell phones mostly use bare quartz crystals--and they have customers that wanted to keep using their exisiting oscillator circuitry. However, they also said they already have customers asking for the integrated ASIC with oscillator circuitry in the cap, and plan to release those devicesw next on their road map Also more frequencies are on the way.