microstrip l c

Hi all,
I know a few reason about that but I have more information to understand this stuff ?
Can someone explain why we prefer microstrip filters to the others ?

It all depends upon the manufacturing methods of you cirucut. L and C filter elements cost extra. If you are already having transmission lines in your circuit, the extra transmission lines in the filters are low cost.

in high frequency, for example,above the 3GHz, this is the only choose

Yes, it's true. At high frequency, especially millimeter-wave, it's quite difficult to implement L and C elements on circuit board which has an equivalent size of wavelengths.

Yep, this is right - if you're building a LC filter for very high frequencies you're also building an antenna at the same time. Also inductors and capacitors have bigger losses and at high frequencies, an L can become a C and vice versa (if you e.g. look at datasheets for some capacitors, you will find diagrams showing that or tables telling you the maximum frequency the capacitor still has its usual capacitance at)

Also tolerance of L C circuits may cause problems on high frequency. But microstrip line generates better results.

Inductor and capacitor at microwave frequency is hard to find. some value of inductor and capacitor cannot be implemented e.g.0.05 nH it is very small value. It is almost impossible to use wire inductor.

You'd also find that microstrip filters have an insanely high roll off, that would be quite difficult to obtain from an LC filter. I've done experiments with microstrip filters with rolloffs of about 700 dB/decade.

Low cost? It depends: using space on an RF-grade substrate also cost money, but for frequencies > 3 GHz this should be lower. But a better reason is that at these frequencies tolerances are higher and Q-factor are lower for these filter.

Therefore for good filters, often microstrip is the best choice in this frequency range.

There are a lot of reasons. Yes, it is right that lamped L and C elements will work only far below self-resonant frequency (SRF). For best 0402 size components SRF can be as high as 6 GHz. Even so high number is not high enough to design good filter with lumped elements on microwave frequencies. The price is important, microstrip filter can be done very small and inexpensive, especially when very narrow bandwidth is not required. But there is one very important reason that probably overweight other. This is very high value of unloaded Q-factor for microstrip based resonators. On microwave frequencies it is easy to get Q about several hundreds. With special Q optimization technology I got unloaded Q about 5000 on 5 GHz. This provides very good steepness for filter slopes, excellent attenuation in stop-bands and low insertion loss. This is not widely known point. Actually band-pass filter insertion loss depends primarily on filter order and the ratio of loaded Q to unloaded. We cannot change filter order and loaded Q because these numbers are defined by the spec. But having high value for unloaded Q of filter's resonators allows to get very low insertion loss. For example, for broadband (this ensure low loaded Q) band-pass filter on 4 GHz central frequency it is possible to get insertion loss below 1 dB for 9th order Chebyshev filter with return loses about -15 dB. Definitely it is impossible for lumped LC filters.

Best regards,
RF-OM