S parameters & simulator

[GVVIN]

s parameters gain

Hi all,

I am designing a loop antenna (for magnetic field) ,I am measuring its performance by measuring coupling S21 with a second antenna of same type kept at a distance. I want to know whether my approch is right.

My doubt is S parameter calculations are based on incident and reflected power and calculation based on voltages and impedance (50 ohms Typ).
we say S21 is the forward voltage gain .

Will the measurement based on S parameter give a correct measure of magnetic field coupled.???? I think yes but my simulations are making me think otherwise.....

I am using CST µwave studio, is there any other way i could measure power received.

Thanks in Advance
GV

 

[RF-OM]

nfc antenna simulation

I am not sure I understood everything right.

1. S-parameters are voltage ratios measured when ports are loaded with required characteristic impedance. They are not power and this is why they are used so much in RF engineering.

2. If your antennas are identical and directivities in the direction between them are also the same you may use such antennas to measure the gain. Just divide the S21 by 2 (reduce by 3 dB).

3. magnetic field coupling depends on antenna properties. If it is pure electric antenna you will not get magnetic field information.

4. You may calculate the electric field strange at the antenna aperture. There is easy formula, but you need to know (or calculate) antenna factor first.

5. Think about antenna as the matching circuit between RF power source and air (120PI impedance). It may help.

Best regards,
RF-OM

 

[GVVIN]

s-parameter rfid

Thanks RF-OM,

I will put my doubt in another way, While testing antennas in Anechoic chamber, we have 2 antennas 1 AUT and the other an isotropic antenna. and we rotate and measure received power.

I want to repeat the same using simulation, Here I am using same type antenna having ports and measuring S21. Will it give a figure of received power. Rx Power= Tx power -S21 in dB.

Thanks and Regards
GV

 

[RF-OM]

s21 measurement

I think you may do it, but you need to count Tx antenna gain and RX antenna gain. Also, it is almost midnight here and I am not antenna engineer and did not performed antenna tests for long time. I will ask Azulykit (we are working together) to ask your question tomorrow. He is experienced antenna engineer and knows this stuff better.

Best regards,

Rf-OM

 

[Azulykit]

friis equation s-parameter

You can compare antennas using an s21 measurement in the manner you propose. There are additional factors if separation distance, direction (pattern), polarization and multipath that may be considerations as well. The other assumption that is difficult to verify is if the antennas are really identical. Do you really need to take this path?

Don't forget to consider the Friis equation in your calculations (e. g. space loss). It may be that you would be interested in a link budget and the associated terms as you seem to be after received power. B. Sklar has a useful discussion in his book if you want more details.

I prefer to measure gain and pattern to characterize antenna performance.

I am not familiar with CST so I will not comment there. I suspect that modeling two antennas at a large spacing will tax the computer significantly.

 

[GVVIN]

square loop antenna simulation s parameter

Thanks Azulykit,

I am doing design for Tag antenna for near field. when I thought of S21 I felt it is just a measure of electric field or power calculated from , Voltage square by (impedance)50 and magnetic field is not taken into account. Your comments Pls.

It may be that you would be interested in a link budget and the associated terms as you seem to be after received power. B. Sklar has a useful discussion in his book if you want more details

Yaa I am trying to do link budget with Tag. It will be greate if you could share book from B.Sklar, if you have the same.

Thanks in advance
Regards
GV

 

[Azulykit]

My copies of the book are on paper.......

Now that it is more clear that you have your antennas in close proximity the link calculation is not indicated. The apparent space loss will probably not follow 1/(range)² as it would with distant antennas.

I suspect that you are in what is sometimes known as the reactive near field where coupling measurements and testing is a challenge. There are four distinct methods of transferring power between circuits: conduction, capacitance, inductance, or EM radiation. The first three are very strong in the immediate vicinity of the antennas while EM radiation dominates at large distances. You do not have a direct connection so my suggestion is that you focus on induction and capacitance as coupling vehicles.

This is not really an antenna issue in the usual sense. Gain and pattern measurements assume significantly greater separations. My earlier comments do not really apply here.

The simulation using CST is more reasonable as the problem volume will not be huge. What is your frequency? I would anticipate that you are somewhere around 1 GHz but that is just a guess (WAG). What is the separation? How big are your loops. To get meaningful responses those are fundamental parameters that need to be provided.

Thinking of this as an s21 question still seems reasonable. You can get that from the simulator and from a network analyzer. You may not need to be in an anechoic chamber for the testing.

 

[GVVIN]

yes RFID around 900 MHz, max distance is 250 mm
Loops are around 9 mm radius , may be overall max 70 mm x 70mm, size of PET.

Thanks and Regards
GV

 

[RF-OM]

To GVVIN,

I think it is right and you do not need anechoic chamber. But if your application is for warehouse and usually there are the metal shelves you probably need to do the test on the real shelf and keep NA at least 1 or 1.5 meter away. There probably will be some other parts on the shelf and you may try to put them on real distances from DUT. Here the actual antenna receiving angle become important and may require adjustment. You may find the best angle by moving other devices closer and away from DUT. Now I am talking from the practical RFID experience, not from the naked theory.

 

[zirtapoz]

To GVVIN;

as my experiences about RFID(i m working for an integrator firm on RFID) near field tags have some problems. we are using Alien squiggle tags for near field in some applications. we r cutting them and just loops and asic are being tag. (near field tag) it s working with our special UHF loop antenna aproximately 120mm.(2W. reader)

But we thought that near field has no use commercially. cos no customer needed it. All customers wish to read tags with high ranges. so i think near field ll die. in case of small tags needed (if the size of tag is important such as in jewellery, some liquid applications for magnetic cuopling) than, near field is effective. But as i understood your size are bigger. So what is the point for working near field? i wondered. in my opinion 250mm. for near field is impossible with electromagnetic regulations of countries.. as to me UHF Rfid was born for reading higher distance. if not so and size does not matter, we would be still using 13.56 Mhz. with near field coupling such as Mi fare.

 

[RF-OM]

I have a little bit another information about RF near field usage. There is special new term NFC (Near Field Communications) in the latest telecommunications engineering. This technology will have deal with short distance low power devices and be used for RFID and many other applications. One of the most promising areas is cell phones. Very soon everybody will have NFC deice inside the phone and use cell phone as ID, as credit card for payments and in many other situations that require identification, payments, access to databases or services and so on. Airline, railroad, theatrical and other tickets will be replaced with NFC devices. Plastic credit cards will be obsolete. There are indefinite number of applications and soon NFC will be very important part of our life.

 

[zirtapoz]

yes sure there ll be. And we still have some NFC applications in some Point of sale which provided especially by banks. But they all working very near field.(13.56 Mhz.) Not UHF.

 

[RF-OM]

I worked on NFC problems about two years ago. It was just preliminary investigation and 13.56 MHz was considered. I am not sure what is going on today in this area. 13.56 MHz have pluses and minuses and 915 and 2400 has own advantages. It is hard to say which one is better, but it looks like 2400 is more secure than 13.56 at least.