pfd linearity
时间:04-09
整理:3721RD
点击:
All
we all know that charge pumps used in fully integrated RF synth has a certain level of nonlinear characteristic: injected charge vs. input phase signal difference is not pefectly linear.
This non-linear behaviour impacts on mainly on PLL spurs, in particular in sigma-delta architecture.
I want to measure it in a lab (only PFD-CHP, not the whole PLL).
My basic idea is to (if inputs to PFD are available) inject two signals, with a very controlled phase shift (phase accuracy<<1deg) and measure the average current at charge pump output to calculate the injected charge.
Does anybody have hint, schematics, block diagram on lab characterization techniques of charge-pump non-linearity?
Does anybody have experience on that? Any other measurement technique?
How to generate and control this phase shifted signals with references in the 10-40MHz range?
Thank you in advance
Mazz
we all know that charge pumps used in fully integrated RF synth has a certain level of nonlinear characteristic: injected charge vs. input phase signal difference is not pefectly linear.
This non-linear behaviour impacts on mainly on PLL spurs, in particular in sigma-delta architecture.
I want to measure it in a lab (only PFD-CHP, not the whole PLL).
My basic idea is to (if inputs to PFD are available) inject two signals, with a very controlled phase shift (phase accuracy<<1deg) and measure the average current at charge pump output to calculate the injected charge.
Does anybody have hint, schematics, block diagram on lab characterization techniques of charge-pump non-linearity?
Does anybody have experience on that? Any other measurement technique?
How to generate and control this phase shifted signals with references in the 10-40MHz range?
Thank you in advance
Mazz
I think you can generate tunable phase shifts with an all-pass filters maded by analog multipliers or OTA amplifiers, then with an VNA as tone generatorn (later one already well known PLL+VCO for clean source signal of VNA), splits signal of phase shifter in two, you can measure accurate phase between "in phase" and shifted signals with VNA and finally low pass filter and accurate voltmeter...
BR
