poorren
Junior Member level 3
Hello folks,
I'm trying to build a homebrew RF receiver board. The receiver is of 2 stage down-converter schema. The first IF SAW filter of that receiver is selected to 190Mhz center frequency. In actual measurement the 2IF output, I found a Spur in middle of receiver passband. With some experiments, I found the spur is due to the harmonic product of reference clock of RF board at 190Mhz!!!
The receiver board used an external Ref clock 10Mhz. The clock is buffered and distributed with a simple 74AC04 chip(around 2.5Vpp). When doing the layout of this area, all the Ref clock signal wires are buried in mid layer of PCB (between two Ground layers) with ground vias around to construct a wall. And the placement of Ref clock is far from receiver IF area. Even using this rigorous layout methodology, I still couldn't prevent the 19th harmonic of ref clock diving into the IF path. Based on the measured Spur power at 3IF output and receiver IF gain, I could inversely compute the power this 19th harmonic interference is about -115dBm. So, the board layout isolation is larger than 120dB. I once attempt to solve by using a small capacitor 27pF shunt at 74AC04 signal output. Little improvement is achieved. ~~~ However, if I put my finger to touch the 74AC04 pins, the power of 190Mhz spur could be obviously reduced by about 5dB. I couldn't explain what the effect of my finger is.
One thing to note is I don't have a shield box to cover on the RF receiver now. (will be available next week...) But, my doubt now is if this harmonic interference couples by space, or board, or GND. My big concern is even the shield box won't help me much.
I would like to write this post as open questions, and like to hear your ideas and thinking!
Please reply if you think any points. Thanks advance!
Jeff
I'm trying to build a homebrew RF receiver board. The receiver is of 2 stage down-converter schema. The first IF SAW filter of that receiver is selected to 190Mhz center frequency. In actual measurement the 2IF output, I found a Spur in middle of receiver passband. With some experiments, I found the spur is due to the harmonic product of reference clock of RF board at 190Mhz!!!
The receiver board used an external Ref clock 10Mhz. The clock is buffered and distributed with a simple 74AC04 chip(around 2.5Vpp). When doing the layout of this area, all the Ref clock signal wires are buried in mid layer of PCB (between two Ground layers) with ground vias around to construct a wall. And the placement of Ref clock is far from receiver IF area. Even using this rigorous layout methodology, I still couldn't prevent the 19th harmonic of ref clock diving into the IF path. Based on the measured Spur power at 3IF output and receiver IF gain, I could inversely compute the power this 19th harmonic interference is about -115dBm. So, the board layout isolation is larger than 120dB. I once attempt to solve by using a small capacitor 27pF shunt at 74AC04 signal output. Little improvement is achieved. ~~~ However, if I put my finger to touch the 74AC04 pins, the power of 190Mhz spur could be obviously reduced by about 5dB. I couldn't explain what the effect of my finger is.
One thing to note is I don't have a shield box to cover on the RF receiver now. (will be available next week...) But, my doubt now is if this harmonic interference couples by space, or board, or GND. My big concern is even the shield box won't help me much.
I would like to write this post as open questions, and like to hear your ideas and thinking!
Please reply if you think any points. Thanks advance!
Jeff