Help with TEM Cell- Calculating Field in uV/m - RF Cafe Forums

 Post subject: Help with TEM Cell- Calculating Field in uV/m

Posted: Mon Jan 10, 2005 2:47 pm 

Hello-

My expertise in RF is limited- I write code in C and assembly, design analog circuitry to a 1 megahertz or so, and push a lot of paper.

I am trying to calculate the field strength in a TEM cell given a signal generator output in dbM and the septum length of the cell. The signal generator output impedance is 50 ohm, the cell looks like a 50 ohm load, the testing frequency of 930 Mhz is well within the passband of the cell. Lets put aside cable loss, adaptor loss, swr, and the accuracy of my measuring instruments for the time being.

The TEM Cell manual gives a method for calculating field strength in cell at test specimen location:
1. Record the emf applied at cell input
2. Divide by the septum length
3. You get volts per meter.


I hook up a hp sig gen to the cell, and set the output to -86dBM

1. To calculate the emf applied to the cell, I use the formula P=E^2/R, rearranging to E= sqrt (RP). I use 50 ohms for R, and get P from P=
10*log(Px/1mW)
2. I divide by the septum length. Now I dont have the work in front of me but I get something close to 40-50 uV/M
3. I put a 50 ohm dipole antenna at the test point (gain is +2.5dbi), but measure a -107 dbM power at the test point.
4. Using the formulas from 1, I do not get 40-50uV/m...I get something substantially less

Whats going on? Should I be using 377 ohms instead of 50 ohms in step 4?

Please help this victim of a lumped systems education make progress!

Thanks.

 Post subject:

Posted: Wed Jan 26, 2005 12:28 am 

*bump*

Anyone?

Is "antenna" factor responsible for the 20 db difference? What is "antenna factor" for a dipole immersed parallel to a theoretical perfect TEM wave E-field and no higher modes?

Is "antenna factor" the same as "gain of an antenna"?

 Post subject:

Posted: Wed Feb 09, 2005 12:55 pm 

What is the septum length?

I do not follow how you are converting dBm to uV at the TEM input.
It looks like your starting with dBm (power referenced to 1mW) , converting to the same units and then sticking that number into a Ohm's Law equation.

-86dBm = 2.512pW
Now using your Ohms Law Calculation: P = E^2/R , R=50
E=11.2uV

What do you get when you take 11.2uV emf / septum length?
22.4uV is the input emf, no load.
11.2uV is the RF voltage in a 50Ohm system.

The DUT antenna is non-isotropic, how are you orientating it?

 Post subject:

Posted: Thu Feb 10, 2005 3:12 am 

The septum length is 0.22 meters.

The DUT antenna/PCB is clearly distorting the field. We can put an isolated receiver in and get a correct message 9/10 times @ -83 dbm. But when we put the bottom circuit board in (the receiver is mounted to it), we get the correct message 9/10 times @-86 dbM indicated input power. Its very repeatable +/- 1 dbM indicated input power.


I am told the reason why we measure -107 dbM with a 50 ohm dipole adjusted to the proper dimensions is something called "antenna factor" of a dipole. I am told my calculations do not have to use 377 ohm. I have always thought of this as "gain" of an antenna but I believe my understanding is incomplete.

 Post subject:

Posted: Thu Feb 10, 2005 3:17 am 

The DUT antenna is a loop antenna as in a 900 MHZ pager. It is aligned to be in parallel with the E field of pure TEM mode wave, as the dipole was.

 Post subject:

Posted: Thu Feb 10, 2005 3:23 am 

Sorry for the multiple replies but I am consulting my notes. Finally the DUT has been located in the cell such that the long axis of the antenna is half way between the septum edge and the opposite wall of the chamber.

 Post subject:

Posted: Thu Feb 10, 2005 12:47 pm 

A loop antenna?? Is is a OEM antenna or something you designed?
If your sensitive to the H field, then the 377Ohms would apply. However I doubt that is the case at 900MHz.

I am not sure on your PCB and antenna DUT size. I think the rule-of-thumb is to never exceed 2/3rds the cavity height and leave 1/3 between the DUT and septum open to prevent field distortion.

Here is a good read that I hope will answer some of your questions:
http://www.tdkrfsolutions.com/DataPDFs/antenna_paper_part3.pdf

 Post subject:

Posted: Mon Apr 25, 2005 11:55 pm 

In general, to calculate antenna factor, you begin with the input to the TEM cell in dBm. Subtract the reading on the spectrum analyzer (coming from the probe output) in dBuV (which is dBm plus 107). The result is your antenna factor.
To calulate the antenna factor for HP 11940/11941A near-field probes, use the formula: 104.3- Specan rdg(dBuV)-TEM input (dBm).