Many engineers underestimate the importance of having a solid plan for addressing EMI compliance failures. The absence of a plan B can lead to expensive and/or time-consuming redesigns and delays, and these, in turn, could cause a critical time-to-market window to be lost. SiTime explains that their new SiT9005 series of spread spectrum oscillators, referred to as EMI reduction oscillators, can help overcome EMI compliance issues and failures.
SiTime’s SiT9005 series of EMI reduction oscillators. Image courtesy of Digi-Key.
A Complicated, Yet Necessary, Part Numbering System
SiTime has been generous in providing technical information for these new components. This series of spread spectrum oscillators (SSXOs) offers a host of features, all of which are called out in the seemingly complicated, but necessary, part numbering scheme. However, SiTime clearly explains what each segment of the part number represents (see image below).
SiT9005’s rather long and complicated, yet necessary, part numbering system. Image taken from the datasheet.
Apparently intending to make the process of creating a part number a little (or a lot) easier, SiTime has embedded within the datasheet a hyperlink to their Part Number Generator (see image below). Unfortunately, it doesn’t appear that a part number is actually generated. Bummer.
Part number generator link embedded in the datasheet (page 8).
Methods for Reducing EMI Noise
The SiT9005 offers a variety of methods for improving radiated electromagnetic interference. One of these methods is through customizable drive strengths, which are used to optimize the clock rise/fall times. The rise/fall time of the clock varies based on the capacitive load that the clock must drive; higher capacitive load will result in slower rise and fall times. The table below gives an example of how capacitive load and the chosen drive strength affect the typical rise/fall times.
Expected rise/fall times according to drive strength and load capacitance. Table taken from the datasheet (page 6).
Within the Programmable Drive Strength section of the datasheet (page 5), SiTime has provided a hyperlink to application notes for “detailed information about rise/fall time control and drive strength selection.” However, after clicking on the link it appears that the website hasn’t been updated to include the SiT9005 part…. I couldn’t find any reference to it. In my opinion, this isn’t a big deal. After all, this is a new part, and I’m sure the focus has been on getting the part into the hands of engineers and providing a detailed datasheet (which has been accomplished). If nothing changes in say six months, then I’ll be a bit annoyed.
I also noticed a small documentation oversight. The Output Drive Strength descriptive block in the part numbering system diagram (see above) provides what appears to be a hyperlink: “See Tables 7 to 11 for rise/fall times.” Unfortunately, this link doesn’t work, and also the rise/fall times are in Tables 8 to 12. Again, this is not a big deal.
Another method for reducing EMI noise is spread spectrum modulation, also known as using a spread profile. The SiT9005 part offers two spread profiles: one is triangular, and the other has the shape of a Hershey’s kiss (see image below). I wonder, seriously, if the Hershey chocolate company approved the use of this terminology. And, by the way, this term is not limited to SiTime…other companies use it too.
The triangular and Hershey-kiss spread profiles. Image courtesy of signalprocessing.it.
Pin 1 Functionality Options
Pin 1 can have multiple functions (selectable via the part number configuration): output enable, standby (for low-power applications), or spread disable. The image below provides an explanation of these various options.
Pin 1 functionality options. Image taken from the datasheet (page 3).
The absolute maximum junction temperature rating of this device is listed as 150°C, and Note 5 warns us that “exceeding this temperature for [an] extended period of time may damage the device.” Hmmm…what exactly does “extended period of time” mean: seconds, minutes?
Also, I find table 6 (see image below) to be less than straightforward. Can it be assumed that the junction temperature is never more than 10°C higher than the ambient temperature? More importantly, the datasheet seems to be missing the thermal resistance data (junction-to-ambient and junction-to-case thermal resistance).
Maximum operating junction temperature. Table taken from the datasheet (page 3).
This new SiT9005 series of spread spectrum oscillators looks to be quite impressive, and the datasheet provides generous technical information (sans the thermal resistance data). Hopefully, after a few months, SiTime will revise the datasheet to fix the oversights.
Have you had a chance to use this new spread spectrum oscillator? If so, leave a comment and tell us about your experiences.