This 3 pole bandpass filter using https://wetnet.net/rf_design/3pole.main.cgi design calculator online.
this bandpass filter is butterworth filter and intended to filter digital tv bandwith 470MHz to 800 MHz , and only this signal will detected after antenna.
Band Frequency 470 - 800 MHz
Center Frequency = 635 MHz
Bandwith = 400 MHz
Impedance = 50 ohm
Input this parameter to online filter calculator above :
Click Submit , and the result :
L1,L3 : 0.008763 uH ( 8,763 nH)
L2 : 0.039789 uH (39.789nH)
C1,C3 : 0,000008 uF ( 8 pF)
C2 : 0,000001753 uF (1,753 pF)
We must determine for Inductor L1,L2,L3 the coil diameter, coil length and number of turn of coil.
Using online calculator from http://www.qsl.net/in3otd/indcalc.html we can determine diameter, length and number of turn coil.
For L1,L3 :
L = 0,00876 uH , coil diameter will be d : 0,01 m , Coil length will be l : 0.01 m and number of turns , n : 1,1355
And for L2 :
L = 0,0396 uH , coil diameter will be d : 0,0259 m , coil length will be l : 0,01 m and number of turns n : 1,1355
The most important thing of build filter is the connection between component should be as close as possible, for avoiding wild capacitance that will generate noise, and the ground area of PCB should be good and wide.
On the next article, I will posting the implementation of this filter.
why Impedance = 50 ohm ?
ReplyDeletehello Rodrigo ,
ReplyDeleteSorry for late response, Thank you for your comment. Good Question ? After learn more about on literature , I found in Belden.com the answer, Why 50 Ohm ? It was related with 2 scientist research from Bell Labs in 1929 , Lloyd Espenscheid and Herman Affel.
They were going to send RF signals (4 MHz) for hundred of miles carrying a thousand telephone calls. They needed a cable that would carry high voltage and high power. In the graph below, you can see the ideal rating for each. For high voltage, the perfect impedance is 60 ohms. For high power, the perfect impedance is 30 ohms.
This means, clearly, that there is NO perfect impedance to do both. What they ended up with was a compromise number, and that number was 50 ohms. ( Source : http://www.belden.com ), I will make one article about 50 Ohm impedance for more clear