I'm relatively new to RF design and I have a 2.4GHz transceiver I need to match to a balun but I haven't been able to find good information on how to do that. The 'optimal load impedance' specified by the transceiver for the balanced output is something like 15 + j88 Ohm, and the baluns only shows balanced input impedances of 50 Ohm, or 100 Ohm, 150 Ohm, etc.


Any advice would be greatly appreciated.

Regards,
Angelo

Your question is not very clear to me. Do you want to build a circuit (RLC) that connected to the unbalanced input of a balun will be seen at its output as a balanced impedance of 15 + j88 ohm ? Is your system narrow or wide bandwidth ? Which balun are you using ? Best regards

Hi Alberto, thank you for your reply.

I believe what I need to do is create an RLC network that will match the balanced output of the transceiver to the balanced input of a balun. Most baluns I see state balanced impedance at 50, 100, 150, etc. Ohms, which I assume means the reactive component is zero. But all the transceivers I've seen specify the optimal load at the balanced differential outputs as R+jX so I assume connecting directly to a balun will result in a mismatch.

I would have a similar problem with various components I'm considering, but the particular parts I'm most likely to use are the Nordic nRF24LE1 RF micro and the TI CC2591 front end (which includes a balun).

nRF24LE1
product page http://www.nordicsemi.com/index.cfm?obj ... lay&pro=95
data sheet http://www.nordicsemi.no/files/Product/ ... c_v1_3.pdf

CC2591
product page http://focus.ti.com/docs/prod/folders/print/cc2591.html
data sheet http://focus.ti.com/lit/ds/symlink/cc2591.pdf

Thanks for your help!

Ok, I think the fastest way is to connect an external balun to the CC2591, as suggested in its data sheet, page 6. Then connect the SMA output (it is a 50 ohm unbalanced) to the matching network you can find in the nRF24LE1 datasheet, page 186, or at the link "http://www.nordicsemi.com/files/Product/white_paper/Tuning_the_nFR24xx_matching_network.pdf".
Another way (quite complex) can be the design of a differential matching network connecting directly the two ICs. Starting from the CC2591 you have to apply a differential matching from 50 to 15 ohm then you can add two inductor (one to each branch) having 44 ohm reactance each, @ 2.4 GHz. Since VDD_PA must supply both ANT1 and ANT2 you need also two high value inductors (let me say 39 nH) to connect those pins. To calculate the matching network you can use RFsim99 freeware SW: I suggest the lowpass implementation to reduce the harmonics content. To convert the matching circuit to differential you just have to duplicate the series inductor on each branch dividing by 2 the original value and leaving the capacitor to its original value. However if you are no so familiar with matching netwok I think this could be a not so easy task.

Greetings,
I just went through the entire design process to do just this. There are several ways to skin this cat but the simplest and best approach is to break down the functions to two parts. First use a 180 Degree hybrid to split the power with proper phase shift. Then use a lowpass matching network to step up from 50 Ohms to whatever the target. The LP is the lowest loss solution to matching and gives the benefit of additional frequency rejection. Make it an even number of sections for the assymetric impedance response. Starting with a series L for the 50 Ohm side you could possible match with 4 or 6 elements. This means that the coils will increase in value and the caps will decrease.

You can find a number of Tutorials on RF Circuit Matching on everything RF.

Here is the link to the Page which has tutorials on RF Circuit Matching: http://www.everythingrf.com/Articles/Matching