Crystal resonator or an oscillator? How to choose the most effective technology for your design

Although the unit cost of crystals is generally lower than oscillators, once the total cost is calculated, the picture looks much different.

This article originally appeared in the August'24 magazine issue of Electronic Specifier Design – see ES's Magazine Archives for more featured publications.

By Jim Holbrook, Director, Customer Engineering at SiTime

Three common crystal design issues

At SiTime, we often hear from many customers when they have crystal design issues such as cold startup failures, oscillator circuit problems from mismatched crystals, or failure to pass EMI tests. These problems cause engineering cost overruns during development and can create costly quality issues. In addition, delaying the product release date leads to lost revenue. Let’s examine three situations when customers came to SiTime to help reduce their overall cost of ownership when facing crystal design concerns.

The difference between a crystal resonator an oscillator

First let’s cover the basics – what is the difference between a quartz crystal resonator (XTAL) and a crystal oscillator (XO)? In simplified terms, a crystal (sometimes called a resonator) is a moving/ resonating passive component that connects to an external oscillator circuit in the chip that it is timing, such as an SoC, microcontroller or processor, as shown in Figure 1.

An oscillator, shown on the right, is an integrated timing solution that contains both a resonator and an oscillator IC in one active device. In the case of SiTime oscillators, the resonator is based on silicon microelectromechanical systems (MEMS) technology instead of a traditional quartz crystal. This integrated architecture enables robust ‘plug-and-play’ oscillators that are flexible and very easy to design into a system.

Total cost of ownership

Oscillators are easier to design into a system since they include functionality and features that solve common and often difficult timing design problems. We’ve laid out some total cost examples, which are based on pricing from DigiKey and SiTimeDirect for XTALs and XOs with the same output frequency, stability, and package size. Then we add the cost of the engineering workhours (based on $100 per hour) required to remedy the problem.

Each use case has a different breakpoint based on production volume and engineering time. In general, the cost of designing with a crystal is lower when quantities are very high and design costs are amortized over large volumes. Conversely, the cost of using an oscillator is lower when quantities aren’t in the tens of thousands.

But there’s more to the story.

What’s NOT factored into the following examples is the opportunity cost (lost revenue) due to project design delays, which can be tremendous in some markets. In some cases, there are additional costs for outside services and testing, which can also be significant. There are also other penalties that add up. These include the expenses for additional materials and components needed for board re-spins, the cost of load capacitors that are required with crystals (and not with oscillators), and the additional board space consumed by the capacitors – all of which further tilts the equation toward using an oscillator.

For simplicity’s sake, in the following examples we’ve included ONLY the cost of the timing component and the engineering time to correct the crystal problem.

Cost of crystal vs oscillator – crystal matching requirement and potential for oscillator failure

Getting a crystal resonator and oscillator circuit to work together can require some careful engineering to deal with issues such as crystal motional impedance and oscillator negative resistance, as well as optimising the drive level, and ensuring the crystal will operate consistently in resonant mode. Because the oscillator is an integrated solution that combines the resonator and oscillator IC in one package, designers don’t need to worry about pairing considerations matching errors are eliminated. In this customer case, 40 hours of engineering work was required to correctly match the crystal and oscillator.

Factoring in the cost of this work, it would have been less expensive to use an oscillator for a production volume up to about 17,000 units.

Cost of crystal vs oscillator – cold startup failure

A lack of diligence in matching the quartz resonator to the oscillator circuit means the oscillator may not start-up properly.

Note also that start-up problems may not be limited to cold conditions. Crystal datasheets, unlike oscillators, offer no start-up guarantee because the crystal is just one piece of the oscillator circuit. To ensure reliable starting under all conditions, the user must take care to match the crystal to the oscillator circuit as described in the previous section. Unlike crystal resonators, MEMS oscillators do not have startup problems. In this customer use case, 15 hours of engineering work was required to correct a crystal cold-startup failure. With a relatively quick fix, the cost benefit of using a MEMS oscillator is realised when production volume is around 6,500 units or less. In other words, unless you are manufacturing in very large volumes, you could be paying more for a crystal-based timing solution in the long run.

Cost of crystal vs oscillator – EMI compliance failure

Crystals can be susceptible to EMI (electromagnetic interference) from external sources. The traces that connect the crystal to the oscillator IC act like antennas and the sine waveform going into the IC has a relatively slow rise/fall time compared to oscillators. If designers need to use shielding, modify the board layout, or introduce extra components to work with the crystal, this adds expense and board space. Plus renting an anechoic chamber for additional testing could incur another $3,000 or more. Redesigning the board and retesting can require 50 hours of engineering work, making it more beneficial to use a MEMS oscillator at volumes of around 23,500 or less. And this doesn’t include the additional materials and test facility costs mentioned above.

Thanks to their high level of integration, SiTime MEMS oscillators are less vulnerable to external interference. On the other hand, the clock is often the largest contributor to EMI in a system and can cause a prototype to fail compliance testing, so SiTime oscillators also provide features that allow quick and easy reduction of EMI emissions. These include support for spread-spectrum clocking, as well as FlexEdge, a proprietary, programmable feature that lets designers adjust the rise/ fall time (slew rate) of the clock signal to lower EMI.

Bottom line – savings across the board

In addition to direct costs, there are other factors that affect the cost of designing with crystal resonators. For example, oscillators can drive multiple loads. That means one oscillator can replace multiple resonators, which can provide a timing signal for only one device.

Additionally, SiTime MEMS oscillators are based on a programmable architecture that makes them readily available in any frequency, stability, and voltage within a very wide range. This provides greater flexibility for designers to optimise their design. In fact, SiTime oscillators can be programmed by key distributors or even by customers in their own labs using the Time Machine II.

Programmability can also reduce the cost of qualification efforts if specification changes are needed. This cost- and time-saving benefit is possible because a MEMS oscillator (before programming) can generate millions of part numbers. Once the base part is qualified, it can be configured to support a huge variation of specifications.

Perhaps one of the biggest indirect savings comes in the form of higher quality and reliability. SiTime MEMS oscillators have less than 1 DPPM (defective parts per million) and more than2 billion hours MTBF (mean time between failure), which is up to 50 times better than quartz devices. In addition, SiTime MEMS oscillators have much better survival rates against shock and vibration compared to quartz crystals.

The higher failure rates of quartz crystals can increase costs in many ways, such as the added resource costs for root-cause analysis or extra service and replacement costs.

Furthermore, the damage that quality issues can do to a brand’s reputation can have a long-lasting negative effect on a company’s bottom line.

Using an oscillator instead of a crystal resonator can lower costs in many ways. Why not skip all the headaches and extra expenses, and use an oscillator? When procurement is focused on reducing component costs, looking at the big picture ultimately will help your company save in the long run. To learn more about the benefits of oscillators beyond cost, read our white paper: The top 8 reasons to use an oscillator instead of a crystal resonator.