Simulating an RFID tag

EDAboard.com Forum Index -> Electromagnetic Design and Simulation -> Simulating an RFID tag

Author

Message

Catalyst

Joined: 22 Jan 2005
Posts: 167

14 Nov 2005 17:42 maxwell software for rfid

Hi,

I was wondering if anyone can suggest a good EM simulator that is suitable for RFID tags (planar spiral antenna). I'm looking at 13.56MHz for the chosen frequency.

Can anyone point me in the right direction because there are so many packages out there, I don't know where to begin Sad

Thanks for your time

Cat

jian

Joined: 08 Sep 2003
Posts: 873
Helped: 151

15 Nov 2005 3:51 ie3d max threads

Hi, Catalyst: You can try the free eval of IE3D from www.zeland.com. I am from Zeland Software, Inc. There are quite some IE3D users in RFID. Thanks!

Catalyst

Joined: 22 Jan 2005
Posts: 167

15 Nov 2005 10:42 rfid maxwell 3d

are there any guides or tutorials that are RFID tag related?

Infact, will IE3D accurately simulate a 13.56Mhz structure as all the options i've seen upto now are only defined in GHz.

jian

Joined: 08 Sep 2003
Posts: 873
Helped: 151

15 Nov 2005 17:05 16-j380

Hi, Catalyst: It is accurate for that frequency range. Please give it a try.

Catalyst

Joined: 22 Jan 2005
Posts: 167

16 Nov 2005 12:49 rfid_tutorial.pdf

OK, I'm looking into the program more, theres a lot to take in.

Could you please explain to me how to simulate and measure the inductance of a structure at a specific frequency?

Kind Regards Smile

Cat

Google
AdSense
Google Adsense

16 Nov 2005 12:49 Ads

rautio

Joined: 21 Feb 2005
Posts: 297
Helped: 23
Location: North Syracuse, NY

16 Nov 2005 13:06 rfid tag antenna presentation

Attached are three files. 13.56MHzRFID.pdf illustrates how to use Sonnet to analyze RFID. RFID_tutorial.pdf is a tutorial I got from the Microchip web site. It explains the entire 13.56 MHz RFID technology in detail. The third file is a zipped directory of all the Sonnet circuits in 13.56MHzRFID.pdf. About half of them can be analyzed using the free SonnetLite (www.sonnetsoftware.com).

I just returned from a two week trip to Asia. I spent one Saturday in Tokyo as a tourist (visinting postage stamp museums, they have some really good ones!). I took quite a few trains. The JR trains all now have an RFID option. You carry a credit card sized RFID card (13.56 MHz, 8 turn coil, a friend took one apart to see what was inside). You place the card on the reader spot and then just walk through. It is OK to leave the card inside your wallet and place your wallet on the reader spot. Works really well and is very fast. You can even make purchases at stores with it. Cool stuff!

Sorry, but you need login in to view this attachment

Catalyst

Joined: 22 Jan 2005
Posts: 167

16 Nov 2005 16:04 ext: pdf rfid technology in detail

Thanks rautio for those files.

Could HFSS or MWS also be used to simulate structure at 13.56MHz? I've always come across alot of problems in the past.

jian

Joined: 08 Sep 2003
Posts: 873
Helped: 151

16 Nov 2005 17:18 sonnet rfid chip

Hi, Catalyst: After simulation, MODUA is invoked to display the s-parameters. You can select PROCESS->FIND 1-PORT EQ CKT and MODUA will find the parameters for the equivalent circuit for you at all frequency points.

Catalyst

Joined: 22 Jan 2005
Posts: 167

16 Nov 2005 18:06 Re: Simulating an RFID tag

Thanks for your reply jian, that worked fine. Cheers, I am going to have a more in-depth look into IE3D

However, I would also like to know whether HFSS and CSW could be used to simulate a planar structure at 13.56MHz as well?

Has anyone in this forum had any experience in this area?

Regards

Cat Smile

rautio

Joined: 21 Feb 2005
Posts: 297
Helped: 23
Location: North Syracuse, NY

17 Nov 2005 1:36 Re: Simulating an RFID tag

Yes, certainly HFSS and CST can be used. They differ from planar tools by being volume meshing tools. I would recommend the following procedure:

1) Do as much as you can using planar tools (properly used, they are more accurate and faster for planar circuits).

2) If possible, do analysis with both a shielded and an unshielded planar EM tool. This is especially easy using ADS and Momentum and Sonnet because both Momentum and Sonnet have very good interfaces into ADS. Just a couple clicks does the trick. The degree of difference between the results is an indication of how accurately the results have been calculated. If there is a large difference, you should check to make sure you have set the problem correctly for both tools.

2) Do at least one case in both planar and volume mesh. Make sure you get the same result (to within your requirements).

3) Now, using the volume meshing tools, you can test things like deforming the RFID tag, the effect of objects being near the RFID tag, the effect of the finite area of the RFID tag and reader, and the effect of odd orientations between the tag and the reader.

This is all in line with a presentation I just gave at the Microwave Exhibition in Japan (Yokohama) last week, "Which EM Analysis Should I Use". I also gave this presentation in Taiwan on Monday. In this presentation I recommend a "dream team" of four tools: HFSS, CST, Agilent Momentum, and Sonnet (even though Sonnet is the only tool we sell in Asia). I detail both advantages and disadvantages of each tool (it is espeically important to know disadvantages in order to do good engineering). Then I show how having all four tools allows extremely powerful EM analysis capability, much better than having just one tool. With the usual number of people, 60 to 70, the Sonnet seminar is usually the best attended of all the commercial seminars at this conference. This time we had 130 attendees. It was very gratifying to see.

Catalyst

Joined: 22 Jan 2005
Posts: 167

18 Nov 2005 10:04 Re: Simulating an RFID tag

Nice response rautio Smile

One of the problems I have is to decide which type of port I need to use to setup an accurate analysis.

In HFSS, there are Wave ports, lumped ports, voltage/currnet excitations. There are also numerous options in Sonnet.

For an RFID tag setup, which ports should be utilised in each of these programs?

Regards

Cat

rautio

Joined: 21 Feb 2005
Posts: 297
Helped: 23
Location: North Syracuse, NY

18 Nov 2005 13:10 Re: Simulating an RFID tag

Hi Cat -- In Sonnet, there is only one kind of port. It is an infinitesimal gap with a voltage impressed across it. This port can be put in different locations. For example, on the edge (so that the sidewall provides a perfect ground reference), it is called a sidewall port. On the interior, it is called an interior port. However, it now no longer has the same ground reference as the sidewall ports. So, you can click an option (by double clicking on the port) to make it an auto-grounded port. In this case vias are added to ground to make the ground refernce. If the port is a via, then we call it a via port. For RFID, just build on the circuits I attached above, and you should do just fine.

As for HFSS, I just know the high level trade-offs. I think it is likely that port choice will be important, but I can not advise you on which port to use. If you are on support, by all means talk to their support on this matter.

As for CST, if you are on support or considering purchase and in North America, we (at Sonnet) will be glad to help. I do not have the knowledge to help you on CST, however, we do have several support employees who are extremely knowledgeable on the matter. Otherwise contact CST directly.

And of course, I am sure there are others who can provide the needed information right here on this forum.

eng_tafa

Joined: 15 Feb 2006
Posts: 3

15 Feb 2006 12:41 Re: Simulating an RFID tag

thnx ur discussion was very useful to me
thnk u again

eng_tafa

Joined: 15 Feb 2006
Posts: 3

16 Feb 2006 11:15 Re: Simulating an RFID tag

hi every body
peace abon you all
i would like to ask about the quality factor of RFID antenna
all my results give alarge Q >100
and i need Q<40 what can i do
thanx

jimliu12

Joined: 01 Mar 2006
Posts: 11
Helped: 1

25 Mar 2006 16:43 Simulating an RFID tag

you can increase R, and just resistance

itaifrenkel

Joined: 09 Oct 2004
Posts: 147
Helped: 20
Location: Israel

26 Mar 2006 21:33 Re: Simulating an RFID tag

Hello Catalyst,

For low frequencies you should not use Ansoft HFSS but rather Ansoft Maxwell 3D.

Abstract: RFID area of applications is currently undergoing constant change in regard of standardization. As any other frequency allocation issue, RFID applications are generally governed by government legislation and as such it may differ in some countries. Standardization activities are seeking to create a more uniform environment in this respect. The most significant carrier frequencies for RFID applications that have been identified to cover low, medium and high ranges are 125 KHz, 13.56 MHz and 2.45 GHz. This paper deals with low and medium RFID applications which appear to enjoy a considerable degree of uniformity of standards and acceptance worldwide.

<a href = "http://www.ansoft.com/news/articles/RFID_Systems_with_Maxwell_Software.pdf">A Study of RFID Devices and Efficient Ways of Simulating RFID Systems with Maxwell Software.</a>

This was performed with Maxwell v10. Maxwell version 11 has the modern "HFSS-like" user interface.

Regards,
Itai

Logan_

Joined: 12 Dec 2006
Posts: 38

12 Dec 2006 6:48 Re: Simulating an RFID tag

I design an RFID coil tag antenna and I am having problems with impedance matching . How to matching coil antenna with an integrate circuit?

RFSimulator

Joined: 24 Jan 2004
Posts: 176
Helped: 15
Location: Europe

12 Dec 2006 7:38 Re: Simulating an RFID tag

Hi,

The RFID guys I know design the antenna directly in a ways that it fits to the output impedance of the RFID chip. They do not use any extra matching network because this would make the RFID more expensive.

If you know the complex output impedance of the RFID chip you will use, you can optimise the antenna geometry to match this impedance. Many simulation tools (I know for CST but I guess other tools can do the same) allow you to optimise the geometry directly to match a specific impedance. Does this help?

F.

badlok

Joined: 03 Jan 2007
Posts: 6

03 Jan 2007 16:39 Re: Simulating an RFID tag

Hi all,

I have some question about the read range, the read range is proportation to the tag antenna's gain, I want to know that----------- The antenna gain is simulated in ie3d is correct?

Because my tag's antenna is directly match the RFID chips (16-j380) but in the ie3d the default input impedance is 50 ohm, hence when I get the antenna's gain simulation result, I don't know it is correct or not

thank all.

jian

Joined: 08 Sep 2003
Posts: 873
Helped: 151

03 Jan 2007 17:02 Re: Simulating an RFID tag

Hi, Badlok:

The gain calculated in IE3D does consider the mis-match to the normalization impedance. However, you should try to use the conjugate matched gain which will consider the antenna to be conjugately matched. It should be the value you want.

Best regards,

badlok

Joined: 03 Jan 2007
Posts: 6

03 Jan 2007 17:40 Re: Simulating an RFID tag

Hi Jian,

What is the conjugate matched gain ? For my antenna design, the chip's impedance is 16-j380, hence I make the antenna's input impedance is 16+j380 to match (max. power transfer)

I don't know how to calculate the gain base on ie3d provided gain. I have calculated but the result seems wrong.......

Best Regards

jian

Joined: 08 Sep 2003
Posts: 873
Helped: 151

03 Jan 2007 18:07 Re: Simulating an RFID tag

Hi, Badlok:

Conjugate Match Gain is a gain implemented in IE3D/PATTERNVIEW Version 11 and later for RFID designers. We noticed that RFIDc always have complex output impedances. Some people tried to use s-parameters normalized to complex Zc for such cases. We consider using complex Zc normalized s-parameters is an incorrect concept. As it is mentioned in some other thread, complex Zc normalized s-parameters may create |S| > 1 even for a passive structure. This is the problem caused by using complex Zc. Instead, users can do either of the following way:

1. Use some C or L to correct the output impedance to make it as Zouput = R + j X and X = 0. Then, you can use the R value for s-parameter normalization.

2. On IE3D 11, we implemented the conjugate match gain. We assume the antenna impedance Zin = R - j X while the RFID impedance is Zoutput = R + j X. Then, we also remove the mis-match portion of the loss in calculating the gain. Basically, it is the best gain you can achieve for the RFID with given Zoutput. It is available on IE3D/PATTERNVIEW 11 or later.

Best regards,

badlok

Joined: 03 Jan 2007
Posts: 6

04 Jan 2007 20:30 Re: Simulating an RFID tag

Dear Jian,

Thx for your reply, I also have some question about ie3d.

My IE3D is ver. 11, I don't know how to set up the conjugate match gain. In simulation set up window, I click box of the Radiation Pattern File then see my attachement. Is it a method of conjugate match gain?

If yes, how to set the value? My RFID chip is 16-j380 and my anteena's input impedance is designed to 16+j380

Thank once again for your kindly reply

Best Regards

Sorry, but you need login in to view this attachment

jian

Joined: 08 Sep 2003
Posts: 873
Helped: 151

05 Jan 2007 2:35 Re: Simulating an RFID tag

Hi, Badlok:

There is no setup for it. On the IE3D simulation, you just check to do the pattern calculation. After the simulation, IE3D will do the pattern calculation save save the .pat file. Then, it will invoke PATTERNVIEW to display the pattern for you. You just select the .pat file in the Pattern List. Select EDIT->Pattern Properties and it will tell you the Conjugate Match Gain (CM Gain) at each frequency. We have not implemented the display of CM Gain in graph yet. It is a good idea to display it in graphs. At this time, you can only find the values in the Pattern Properties dialog.

best regards,

virtaul

Joined: 04 Sep 2006
Posts: 4

12 Jan 2007 8:25 Re: Simulating an RFID tag

HI!!

If I simualte the RFID tag in HFSS, set up the port whose impedance is R and reduce jX with RLC lumped boundary, I observer the gain of pattern.

Are the method I use and result correct??
Do I consider the S parameter of the simulation in HFSS?

ruey

navuho

Joined: 05 Jun 2001
Posts: 110
Helped: 14

12 Jan 2007 21:54 Re: Simulating an RFID tag

virtaul wrote:

I observer the gain of pattern.

Look to this topic :
http://www.edaboard.com/ftopic128935.html

badlok

Joined: 03 Jan 2007
Posts: 6

15 Jan 2007 20:28 Re: Simulating an RFID tag

Hi Jian,
My RFID chip is 16-j380 and my anteena's input impedance is designed to 16+j380. How to know IE3D the Conjugate Match Gain is reference to my chip ??

For examplem, if I don't set up anything, just let IE3D do the pattern calculation, what does the Conjugate Match Gain mean??

Thank you for your kindly reply

Best Regard

---------------------------------------------------------------------------------------------

Hi, Badlok:

There is no setup for it. On the IE3D simulation, you just check to do the pattern calculation. After the simulation, IE3D will do the pattern calculation save save the .pat file. Then, it will invoke PATTERNVIEW to display the pattern for you. You just select the .pat file in the Pattern List. Select EDIT->Pattern Properties and it will tell you the Conjugate Match Gain (CM Gain) at each frequency. We have not implemented the display of CM Gain in graph yet. It is a good idea to display it in graphs. At this time, you can only find the values in the Pattern Properties dialog.

best regards,

jian

Joined: 08 Sep 2003
Posts: 873
Helped: 151

15 Jan 2007 20:45 Re: Simulating an RFID tag

Hi, Badlok:

I checked it and it is Conjugate Match Efficiency (not gain). Howver, it means the same thing. It is not referencing to your chip's internal impedance. It is referencing to the antenna's input impedance.

From what I see, if your chip's internal impedance is 16-j380. You should try to optimize your antenna's input impedance to 16+j380. Then, you simulate your structure again. You can just use 50-ohm for the Zc of your port.

You will get 3 efficiency values while you display the pattern properties:

1. Radiation Efficiency
2. Antenna Efficiency
3. Conjugate Match Efficiency.

Basically,

Radiation_Efficiency = Radiated_Power / Net_Input_Power
Antenna_Efficiency = Radiated_Power / Incident_Power_50_Ohms
Conjugate_Match_Efficiency = Radiated_Power / ( Net_Input_Power + Power_consumed_in_source)

where
Net_Input_Power = Incident_Power_50_ohms - Reflected_power_50_ohms
At conjugate match, Power_consumed_in_source = Net_Input_Power (assuming Z_source = Conj( Z_antenna)). Therefore, Conjugate_Match_Efficiency = 50% of Radiation Efficiency.

On PATTERNVIEW of IE3D, we provide GAIN value. If you want to know the best gain you can achieve on your antenna, you should calculate it as:

GAIN_CONJUGATE_MATCH = GAIN * CONJUGATE_MATCH_EFFICIENCY / ANTENNA_EFFICIENCY (everything is in linear scale, not dB).

Best regards,

Regards.

3dandyw

Joined: 18 Jan 2007
Posts: 36
Helped: 1

18 Jan 2007 17:28 Re: Simulating an RFID tag

HI,

i was looking for similar ap.
i got answer reading his app notes

www.flomerics.com/news/news_details.jsp?newsId=335 -
Modeling An RFID Tag Antenna With MicroStripes

I was impressed about simulation speed !
AW

andrewmorcos

Joined: 10 Apr 2008
Posts: 10

10 Apr 2008 16:49 Re: Simulating an RFID tag

hello,
i am new to RFIDs
and my thesis is to do an positionning system using RFIDs.. i need to simulate this system on CST MWS...
Anyone can give me some hints? thnak u