PLL System Design
One of the big mysteries in PLL/frequency synthesizer design seems to be, "Where do the specifications come from". I hope to be able to answer this in this web page. Radio level specifications don't generally specify what the phase noise on a local oscillator needs to be. Instead, the radio level standard will specify things like adjacent channel interferer levels, 2 tone interferer's, output spurious levels, etc. Exactly what is specified will depend on who wrote the specification and what they are trying to achieve. The specifications will be different, depending on whether it's a receiver or a transmitter that's being specified.
The actual phase noise specifications are derived from the radio level specifications, using information such as the reference frequency, the channel step size, the PLL architecture, the radio level architecture (Dual Conversion superhetrodyne, Low IF, Zero IF, I/Q Upconverting mixer etc.) cost, current, etc.
PllDesign.xls
I'm starting to write a PLL Linear Analysis tool in Excel, using a combination of Excel VBA code and the complex number support to crunch the numbers. The first pass is here Link (Excel Spreadsheet) I'd be interested in any feedback on the usefulness of this.
09/29/2006. The next version is here Link (Excel Spreadsheet). This is looking almost useful at this point, I've added calculations of the loop filter components for 2nd, 3rd, and 4th order loop filters on separate worksheets. I've also added calculations of the resistor noise contributions, but I'm still suspicious of the Reference Oscillator and VCO phase noise calculations.
11/16/2006. The next version is here Link (Excel Spreadsheet). A closer approximation to useful. I've finished closed loop phase noise for 2nd, 3rd, and 4th order filters, with VCO and reference oscillator phase noise as inputs from a worksheet.
08/16/2007. I've finished the last addition I'm planning on doing to this. I tweaked the standard component value calculations to keep the poles and zeros with standard components closer to the ideal version. At this point, I'm going to stop working on this, as I haven't received any feedback on this spreadsheet.Link (Excel Spreadsheet)