Antenna matching network

Status
Not open for further replies.
About existing network did I wrote:
In worst case is coverage a bit short but it will work good enough to verify that it is alive.
Click to expand...
What that means is that component places exist for later measurement what values that should be plugged in.
Of course if transmitter is 50+j0 Ohm and transmission lines are ideal and no need to improve neither ESD or harmonics or decouple DC bias, then are no components needed.
Existing component values are not based on any measurements, so why simulate 50 Ohm to 50 Ohm?
Have you ever measured impedance of a speaker? Datasheet says 4 Ohm? In best case is it correct with 50% tolerance within a limited frequency range.
Now you have a transmitter for which datasheet says 50 Ohm? If you check datasheet for say nrf24L01 or CC2510, two different well known brands of very common 2.4GHz chips, do they at least specify the reactive part in their data sheet.
Stating that impedance is 50 Ohm in datasheet is as best not to far away from actual truth, but without specified reactive part is it worthless to even think about any simulation.
This was not designed by me I don't know the reason why he put this matching structure in external antenna.
Click to expand...
As I also wrote previously, if possible, is it clever to leave space for a network even if you when drawing schematic not is sure if it is needed and not even know why it may be needed.
 


So I can start with an input and output impedance that is close to 50 ohms and contain some small reactive parts also and use this to design something first. That could be be a good solution. Or as FvM wrote to use a low pass filter with 50-50 ohm input-output impedance.
 

I am totally blind in this case, do not know what TX circuit you design for, do not know if there are transmission-lines or how well they are calculated for actual PCB material or if there are any limits given how close to 50+j0 Ohm actual impedance must be designed for. A cheap toy can have wider limits but if you design a precision signal-generator is decimals of interest.
With background of what I know about your design, do I recommend better safe then sorry.
 


The pcb traces are well designed and they are very close to 50 ohms. The final board is going to be used for a commercial product as a WLAN gateway and the boards are going to be manufactured professionally, so I suppose I can make this assumption. After that if I put a 0 hm resistor instead of an unwanted series component this wouldn't be a problem right?
 

E Kafeman said:
0 Ohm serial is no problem.
Click to expand...

Is there a manufacturer of external antennas at 2.44 Ghz who can provide a datasheet with antennas S parameters so I can make simulations with this parameters?
 

Some provide S11 parameters. Especially for monopoles are these parameters only correct for the by provider given size of ground-plane. At 2.4 GHz, is it common with other types of antennas which visually looks like monopoles (one black plastic pin with a RP SMA in bottom) but are internally designed like end-feed dipoles or as collinear elements. These are usually less depending on ground-plane for its impedance.
It is however relative common that these antennas are mass-produced by a mechanical producer which can mold plastic pins but have no clue how impedance is measured. They do then in data-sheet state that antenna impedance is 50 Ohm and black. Probably equal truth as that your transmitter have 50 Ohm impedance .
Most antenna impedances are relative okey, but have also seen some real bad examples.
 
Reactions: ctzof

    ctzof

    Points: 2
    Helpful Answer Positive Rating
When designing a transmitter with a connector for an external antenna, you'll assume 50 ohm impedance. Full stop.
 
Reactions: ctzof

    ctzof

    Points: 2
    Helpful Answer Positive Rating

In theory you can perform any narrowband impedance transformation with a two-component matching circuit like L6+C165 in your original schematic (though depending on the antenna impedance they may not end up being an inductor and capacitor as drawn). The problem is that the match might demand odd component values. So adding a third component for a tee or pi network adds some additional flexibility. Again, you already have the makings of a tee network in your schematic. But for the external antenna, you shouldn't need any matching network, just a 50ohm microstrip, assuming the antenna port wants to see 50 ohms.
 


Ok but at least want this C157 capacitor as an ESD protection so I suppose this going to affect the final impedance so I need to have some additional components to retune the antenna
 


I need also this ESD protection capacitor C157 because the board is going to be tested with some input voltages for ESD compliance. Probably is going to affect matching network performance right. Is a good idea to put it there? Is there another way to achieve ESD protection? And finally can this capacitor be integrated to the existing matching network?
 

I don't see C157 on any of your schematics.

You can get some protection by making your network a with a highpass frequency response, but your components should be rated for the voltages that will occur from the ESD event. Probably not going to happen if they are little 0402 packages.
 

Its not exactly a capacitor is an ESD protection component with name C157. Is in the first post at the output of external antenna. The thing is that all the connectors of the board have to be tested for ESD that's why I want to put this component while at the same time want to keep the antenna matched. Is there a way for this.
 

Status
Not open for further replies.

Similar threads

Menu
Log in
Register
Cookies are required to use this site. You must accept them to continue using the site. Learn more…