Post Go back to editing

Input voltage tuning speed for HMC739LP4 evaluation board?

Hi, I have a HMC739LP4 evaluation  board, which I want to use to setup a heterodyne phase locked loop. I wanted to know the fastest speed/frequency or response time, upto which I can tune the input voltage signal ? This is important for me, since I need to know if this VCO is suitable for my application. For e.g. Can I frequency modulate the (input voltage of the) VCO and achieve datasheet performance for a MHz range input voltage signals ? I have a tunable loop-filter that works upto 200 MHz. This information isn't present in the datasheet as far as I understand. Thanks...

Parents
  • Hi, 

    You might be surprised that you actually do have what you need to estimate the tuning speed. The biggest drivers are the source impedance of whatever you plan to drive the tune port with and the required change in tune voltage. The lower the source impedance (Rs) the faster it will tune. If you review the equivalent schematic shown for the VTUNE port you'll see a 4.0pF capacitor that is essentially in parallel with a varactor that at 0Vdc is 5.0pF (maximum capacitance = worst case). This combination comprises the bulk of the capacitance. The 250 ohm resistor along with Rs makes up most of the resistance. Simply use (Rs ohms + 250 ohms) * ( 4.0pF + 5.0pF) to derive the RC time constant. Let's assume the source impedance is 50 ohms so we have a total of 300 ohms * 9pF = 2.7nS. We know that it takes 5 time constants to charge a capacitor to 99.3% of our desired voltage so to remain perhaps a bit conservative we'll allow 6 time constants as there are parasitics and diode variation, etc., that we're not accounting for so 6 *2.7nS = 16.2nS (~62MHz) to tune from 13.0V to 0V or vice versa. Reducing Rs to 5 ohms we get 13.8nS or 72.6MHz.  If you're not tuning over the full bandwidth of the HMC739 you can achieve slightly faster tune times. The above equation is really aimed at deriving the modulation bandwidth. For tuning speed your times will likely be faster as your system may be able to "lock" or otherwise function when you get within 5% of the full charge in which case (3) or perhaps (4) RC time constants worst case may suffice. 

    Hope this helps, 

    Marty 

Reply
  • Hi, 

    You might be surprised that you actually do have what you need to estimate the tuning speed. The biggest drivers are the source impedance of whatever you plan to drive the tune port with and the required change in tune voltage. The lower the source impedance (Rs) the faster it will tune. If you review the equivalent schematic shown for the VTUNE port you'll see a 4.0pF capacitor that is essentially in parallel with a varactor that at 0Vdc is 5.0pF (maximum capacitance = worst case). This combination comprises the bulk of the capacitance. The 250 ohm resistor along with Rs makes up most of the resistance. Simply use (Rs ohms + 250 ohms) * ( 4.0pF + 5.0pF) to derive the RC time constant. Let's assume the source impedance is 50 ohms so we have a total of 300 ohms * 9pF = 2.7nS. We know that it takes 5 time constants to charge a capacitor to 99.3% of our desired voltage so to remain perhaps a bit conservative we'll allow 6 time constants as there are parasitics and diode variation, etc., that we're not accounting for so 6 *2.7nS = 16.2nS (~62MHz) to tune from 13.0V to 0V or vice versa. Reducing Rs to 5 ohms we get 13.8nS or 72.6MHz.  If you're not tuning over the full bandwidth of the HMC739 you can achieve slightly faster tune times. The above equation is really aimed at deriving the modulation bandwidth. For tuning speed your times will likely be faster as your system may be able to "lock" or otherwise function when you get within 5% of the full charge in which case (3) or perhaps (4) RC time constants worst case may suffice. 

    Hope this helps, 

    Marty 

Children
No Data