PN7160 Antenna matching qualification

cancel
Showing results for 
Show  only  | Search instead for 
Did you mean: 

PN7160 Antenna matching qualification

246 Views
michael_d_1983
Contributor I

Good day,

Could you please rate the quality of the antenna matching for the PN7160?

The antenna was measured with a VNA and calculated with the PN7160_matching_calculator.xls, then fine-tuned with the VNA.

00 Antenne eingebaut vermessen mit VNA.png

Our values:

La = 1.2uH
Ca = 8pF
Ra = 0.645R

L0 = 160nH
C0 = 330pF
C1 = 71pF
C2 = 68pF
Rq = 2R2

The Z almost matches the targeted 20R, unfortunately a small (-4.5j) imaginary component remains.

05 Matching 68p 133p 2R2 Basierend auf 04, finetuning.png

Is a small imaginary component okay, or should it be completely eliminated? Unfortunately, I can't manage to get rid of it.

Do you have any suggestions where I could start here?

By the way, reader function, I_TXLDO, AGC (both measured with the Factoryapp) look okay!

Thanks in advance,
Michael

0 Kudos
4 Replies

238 Views
Tomas_Parizek
NXP Employee
NXP Employee

Hello Michael ! 

I think generally it looks OK. But I have a few comments. 

1. Are the values you wrote the right ones? Especially the C2 capacitor? I put your values into the RF simulator and I can see quite a huge difference

Tomas_Parizek_0-1710921045200.png

I double-checked the matching in PN7160_matching_calculator.xls, and the C2 value should be around 135 pF. Then it looks more like your measurement.

Tomas_Parizek_1-1710921154959.png

2. Impedance (22.5 - 4.6j) Ohm is not bad. Of course, ideally, it is better to have exactly e.g. 20-22 Ohms real impedance. Then you have the most "efficient" system. It is also better from a manufacturing and components tolerances point of view. 

You can try to slightly increase C1 and then slightly decrease C2. You should be closer to the real axis. Generally, we recommend to keep the imaginary part up to +-3j 

 

3. Not sure how many "points" are you using for the VNA measurement. If possible, use 801 points. Then the Smith chart will be more "smooth" 

 

BR

Tomas 

0 Kudos

236 Views
michael_d_1983
Contributor I

Hello Tomas,

thank you very much for your suggestions - I'll try them out right away!

Regarding the values: Unfortunately a copy-paste error has crept in - sorry! The actual values for C1/C2 are:


C1 = 68p
C2 = 133p

Thank you,
Michael

0 Kudos

201 Views
Tomas_Parizek
NXP Employee
NXP Employee

Hello Michael, 

Of course! Let me know how has it turned out. 

About the values, don´t worry.  At least you can see that we're really double-checking each customer request  : )

BR

Tomas 

0 Kudos

180 Views
michael_d_1983
Contributor I

Hello Tomas!

I tried a bit more and came up with C1=71p / C2=131p (130p).
Here I get an almost real Z of 25R.

11 Matching 71p 131p 2R2 Basierend auf 04, finetuning xx.png
I accept the higher Z because the read range is actually very good and I then have reserves in terms of maximum I_TXLDO.
If I'm missing something serious, please let me know.

Thanks & BR,
Michael

0 Kudos