Crystal oscillator interference with GPS signals and EMC issues

小浅白白

Crystal oscillator interference with GPS signals and EMC issues [Copy link]

Hello everyone: Recently, I was working on a project. The machine added GPS function. Because the mold of the machine was determined a long time ago, the GPS function was also added after the option. When testing the GPS signal, the result was very poor. There was no problem testing the GPS module and antenna separately. The main reason was that the interference of the machine itself was too large. The main source of interference was that the main board, core board and the module used all had a 25M crystal oscillator. The frequency of GPS was 1575M, which was exactly 63 times the frequency of 25M. The result of scanning with the instrument showed that the main source of interference was also the crystal oscillator. Because I am a novice in this area, I have a question. How many times can the frequency of the crystal oscillator be multiplied at most? In my understanding, the crystal oscillator can multiply the frequency to interfere, but the energy radiated at the same time of multiplication must be attenuated, and it is impossible to multiply it all the time. Another problem is that when the machine scans EMC, how to calculate the frequency of the interference source as shown in the figure below? Is there any calculation method? I don’t know much and the question I asked is too simple. Please enlighten me, or if there is any information you can recommend, I would like to thank you very much.

Arvinˇ

Putting interference sources like crystal oscillators together with GPS modules is not a good design. Moreover, the EMC test shown in the figure only targets the interference radiated from the machine itself to the outside, and does not reflect the interference of the internal crystal oscillator on the GPS signal. The only way to reduce interference is to add some physical shielding layers between the crystal oscillator and the GPS module.

chunyang

Unless your crystal oscillator has a long connection line, the interference is mostly introduced from the power supply/ground. The solution is to strengthen decoupling, including the use of magnetic beads and capacitors. For finished products that cannot be changed by circuit, you can only think of shielding, and you can try to insert grounded aluminum foil.