Quick question on 2.4Ghz PCB design...

pcb trace impedance 2.4ghz

Hi, I've searched the forum and googled this one and I can't seem to find an answer, so I've come to you guys

Right, I have a pair of those 2.4Ghz A/V modules, but the ones I have don't have an SMA connector for the antenna. Instead they have a 3-pin header, with 2 GND pins and the centre one being the RF out. The output impedence of this is the standard 50 ohms and I'm planning to connect this via a PCB trace to an SMA connector, so I can use standard antenna's (those 'whip' type used in 802.11 PCI cards). Now, I have used 'TxLIne' software to work out the PCB trace width. For FR4, 1.6mm its 3.02mm. But, does the length matter? I'm not trying to make the trace radiate energy (ie, use is as a monopole) its just to make a connection between the pin from the module to an SMA connector.

Common sense tells me that it should be short so it doesn't act as an antenna, and since both ends of the connections (RFout and antenna) are both 50 ohms, no actual 'matching' is required.

Sorry if this is a common question, its just bugging me.

Thankyou for your time.

Buried(in)Code.

make 2.4ghz on pcb

Buriedcode said:

Hi, I've searched the forum and googled this one and I can't seem to find an answer, so I've come to you guys

Right, I have a pair of those 2.4Ghz A/V modules, but the ones I have don't have an SMA connector for the antenna. Instead they have a 3-pin header, with 2 GND pins and the centre one being the RF out. The output impedence of this is the standard 50 ohms and I'm planning to connect this via a PCB trace to an SMA connector, so I can use standard antenna's (those 'whip' type used in 802.11 PCI cards). Now, I have used 'TxLIne' software to work out the PCB trace width. For FR4, 1.6mm its 3.02mm. But, does the length matter? I'm not trying to make the trace radiate energy (ie, use is as a monopole) its just to make a connection between the pin from the module to an SMA connector.

Click to expand...

The conexion must be alwais as short as posible, but you must not have any problem if it's any lengt. If I realy understand the 3 pad conexion you tell us, it use to be for a coaxial cable direct conexion (using a metal part, similar to a trident). You can conect a peagy conector.

Buriedcode said:

Common sense tells me that it should be short so it doesn't act as an antenna, and since both ends of the connections (RFout and antenna) are both 50 ohms, no actual 'matching' is required.

Sorry if this is a common question, its just bugging me.

Thankyou for your time.

Buried(in)Code.

Click to expand...

TaPa

ghz pcb design

If your microstrip transmission line is 50 ohms, and if it sees 50 ohms impedance at the terminals, the radiation of the line will be minimal, and the length would matter only for insertion loss.
For insertion loss point of view, of course, shall be shorter as possible.

Thanks for your reply's guys
Good to know people have experience with this stuff, I don't really trust my own knowledge and its always good to confirm something with others before committing to a PCB I can't afford.

If your microstrip transmission line is 50 ohms, and if it sees 50 ohms impedance at the terminals, the radiation of the line will be minimal, and the length would matter only for insertion loss.

Click to expand...

Thats what I thought, I would probably have given up if any impedance matching was needed, since my knowledge of microwave is rusty to say the least. Hopefully I can get my hands on some measuring equipment to check any attenuation, output power etc..

Thankyou again.

BuriedCode