This thread has been locked.

If you have a related question, please click the "Ask a related question" button in the top right corner. The newly created question will be automatically linked to this question.

TRF7970A Antenna Matching Quality Factor with sloa135a

Other Parts Discussed in Thread: TRF7970A

Hi, I am in the process of developing a system based around the TRF7970A and need to match my PCB coil to 50ohm. I believe I know how to do this using sloa135 for reference.

In sloa135a the desired quality factor is 6.78 which makes since the target bandwidth is 2MHz in order not to block the 848k modulation.

I was looking at application circuit in the TRF7970A datasheet when some confusion arose... The circuit depicted there shows:

But if I use the equation for the quality factor of the inductor with parallel resistor (same equation used in sloa135a) I get:

Q = Rp/XL = (6800)/(2*pi*13.56*1.5) = 53.3...

Isn't a quality factor of 53 too high for the application? I saw that 8-10 was recommended in a previous thread

Also, when I used smith chart to find the input impedance to the coil + matching (after R2) I get about 36 ohms instead of 50. Is this correct?

Thanks,

Eric Horton

  • Hello Eric,

    We will need to review the TRF7970A Datasheet circuit, it definitely appears to be off significantly. Please refer to the circuit on Page 8 of this document as an alternative: www.ti.com/.../slou372a.pdf

    And I agree with your Smith chart assessment, you want to be close to the 50 ohm point, 38 ohms is far off. Even the antenna impedance is larger than what I'd expect.

    Please continue to follow the app note which is accurate, and also check out the circuit in slou372a which from the Smith Chart plug in I did, matches up well, coming in at about 47.5 ohms, which is close enough given that the Smith Chart tool is for theoretical evaluation and placing the components on board results in variance.

    Edit: Clearly I should just stick with software... thanks for straightening me out there Josh! :)

  • Eric/Ralph

    the steps are the following, please don't overlook the details: 

    first you need the j + XL of the coil , then tune the resonant circuit, applying the Q killer (shunt resistor) to get correct Q (of the entire circuit!)

    the resulting smith chart is:

  • additional note, this is a higher Q,  longer range applications, with tags that use single sideband of 424kHz  this is OK in some cases

    for the tags which use 848kHz sideband response, the Q should be lower, this is still valid and in that case, using the same coil, it would look something like this

  • Okay! That all makes sense... I guess I should have realized there was a real part of the coil impedance I wasn't accounting for, surprised it's 1 ohm though (seems pretty high, I've been measuring in mili-ohms for my coil... but could be intentional as I don't know the details).

    Also, the Q value being high for longer range and more specific modulation makes sense so thanks for clearing that up! You guys are responsive!

    -Eric

  • Eric -
    i used 1Ohm here just for demo purposes, you are correct that normally a coil would be milliOhms (real), but this is another "fine adjust knob" to turn if you need it. (i.e. use thicker wire or traces for lower R to start, thinner for higher R to start) this technique is also more commonly used in inlay style tags, where you cannot afford to put a Q killer resistor, because they are commodities that have all the cost taken out of them that is possible, plus they are expected to be flat, flexible and possibly pass a mailing test.

    you should always measure your coil first with network analyzer or even a low cost device which is operating at your center frequency (for example an MFJ-259C, or similar), then use the accurate measurement to then either calculate or use Smith Chart tool to get right down to close to what you need for components.