Eliminating RF Resonance and Calculating Choke Impedance

I don't understand the description of the transmitter & receiver by the doorway problem, but I can answer your choke impedance measurement question.

You need an rf signal source, and some way to measure the signal. A spectrum analyzer is best, but an rf detector into a voltmeter will do in a pinch.

The rf signal source has a 50 Ohm output impedance, and the spectrum analyzer or rf detector likewise has a 50 Ohm input impedance. Connect the the signal source to receiving equipment and make a baseline measurement.

Then insert the unknown impedance choke between signal source and receiver and make a second measurement. The difference in the two measured potentials is called the insertion loss and can be related directly to the choke impedance at the frequency of interest.

If the choke impedance is likely going to be low relative to 50 Ohms, you put it in shunt between the transmission line conductors. It will load the source that way and you will get a large insertion loss. If the impedance is likely to be high relative to 50 Ohms, then you place it in series with one of the transmission line conductors, and again the insertion loss will be high. If you do things backwards from the way I described, it will be difficult or impossible to get an accurate measurement.

If you have placed the choke in shunt, the equation relating insertion loss to choke impedance is:

IL = 1/[1 + (25/XL)],

where XL is the unknown choke impedance. From this equation for the insertion loss of shunt injection, you can see why it doesn't work when XL is large: the insertion loss approaches unity asymptotically and you can't get a good reading.

If you have placed the choke in series, the equation relating insertion loss to choke impedance is:

IL = 1/[1 + (XL/100)],

where XL is the unknown choke impedance. From this equation for the insertion loss of series injection, you can see why it doesn't work when XL is small: the insertion loss approaches unity asymptotically and you can't get a good reading.

If you have a spectrum analyzer available and are able to make logarithmic (dB) measurements, then you have the following relationships:

IL shunt (dB) = 20 * log {1/[1 + (25/XL)]} = -20*log [1 + (25/XL)], and

IL series (dB) = 20 * log {1/[1 + (XL/100)]} = -20*log [1 + (XL/100)]

Hope this is what you needed. If you can explain better about the doorway problem, I will take a crack at that as well.

Thank you for your help. To be honest, this may be a bit beyond my understanding. Though I think I get it for the most part, and have a few ideas.

Regarding the door; these units (receiver and transmitter) are operating at a set frequency and actively seeking a return signature from an rf circuit. The door tends to resonate at the frequency being transmitted and deliver a false positive. I am seeking to find ways to change the door resonance, eliminate it completely, or shield it. I believe, that since it is practically impossible to eliminate any and all gaps in the door construction, that these spaces act as capacitors and turn the entire door into a big tank circuit. Is there a thin coating material that can be applied to the door that will shield it from the rf signals generated by the transmitter? If the door is "grounded", will this prevent resonance, or simply dump electrical noise to the ground? Would it make more sense to try to build a shield that can be attached to one side of the antenna, or would this negatively impact the antennas ability to transmit and receive from other directions as well?

Also, getting back to the impedance question; I am able to hook up an oscilliscope to the transmitting antenna and get a visual of the signal. I can then use an on board potentiometer to adjust the (attenutation, I think) to get the highest possible signal. If I worked in reverse, by putting the choke in line with the antenna wires and comparing the signal to the baseline...using trial and error, number of wire turns around a toroidal ferrite, etc. If I kept trying each one until I got the greatest negative affect, I assume that would indicate the greatest impedance possible given the specific ferrite and wire being used. Then I can note this, and build multiple chokes with the exact same factors. The choke would be used to be placed in line with cables feeding power to the antennas with the intent to impede any RF signals in the specific ranges which may ride along these wires and enter the antenna. I realize this is a very layman's way of doing it, but would it work? Would there be a better way?

Thank you again for your assistance.

You might try bonding the door frame to a good low impedance ground and shunting some of that energy to ground. Given enough time, equipment, and money we (engineers) can usually find out the exact cause (physics) although with some guidelines and basic rules of thumb we can practice the "black art" and mitigate your problem without completely understanding it.

What this sounds like is an inventory control system at the entrance to a retail store. Or maybe it's at the back door to prevent "shrinkage" as it is sometimes called from employee theft. Am I on the right track?

If I assume you are dealing with a loop antenna (door frame), c=fλ.

3e8=1.8e6*λ

… λ≈ 167 meters at 1.8 MHz and

3e8=9.5e6*λ

…λ≈ 31.6 meters at 9.5 MHz

A typical retail store metal door frame may in fact present itself as a nice half-wave loop antenna at the higher frequencies you mentioned. See the wiki entry on loop antennas as a good starting point.

http://en.wikipedia.org/wiki/Loop_antenna

I hope this helps in your quest.

Sir Robin hit on my first idea which would be grounding the door. We have done this at some of my transmitter sites. Take decent gauge wire with some slack and wield it to the metal door and the frame. We then have another wire wielded to the bottom of the frame that connects to copper ground straps which all connects to copper stakes hammered a couple feet into the earth, which is our site ground. Depending on your site it might be easier to use a cold water pipe if you do not have a site ground.

Thank you for your pragmatic solution and description. I do believe this could work. As you see on my earlier post, I was leaning in this direction, but needed more descriptive advice.

As far as shielding.....have you tried anything of the sort in your facility? I initially thought of ferrite material, but then wondered if the absorption quality wouldn't simply diminish the transmitted signal. Might I be able to use another reflective sort of shield that could push the signals back to their intended direction and away from the door? Perhaps a "C" shaped attachment on the end of the antenna closest to the door. I cannot redesign the antenna being used, but, could perhaps add on to it to get a more directionalized output.

Any ideas?