Need Help in Loop Antenna Building

[Harros]

The next step to fix these problems is to change the connection point of R1 and of R2. For example, with R1, you should disconnect it from the top of L1 and create a "tap" on L1 about 1/10 from the bottom (that is, 10% of the total number of turns of wire) and then connect R1 to that tap. Similarly, disconnect R2 from C5/L3, create a tap on L3 at the same height as you did for L1, and reconnect R2 there.

By the way, I did not assign the Q of the coils during simulation, will it cause my simulation to be inaccurate? Should i do the simulation without coupling the source impedance and load impedance?

 

[RadioRon]

By the way, I did not assign the Q of the coils during simulation, will it cause my simulation to be inaccurate? Should i do the simulation without coupling the source impedance and load impedance?

Your sim is probably inaccurate. What Q does the sim assume? Infinite?

How would you do the sim without a source?

 

[Harros]

Your sim is probably inaccurate. What Q does the sim assume? Infinite?

How would you do the sim without a source?

I guess so. There is no such an option to allow me to set the Q factor of the inductor that i use in simulation. The freq response of the filter now is much better. But the bandwidth is uneven as i can see a wider band at upper freq. What should i do?

 

[RadioRon]

I guess so. There is no such an option to allow me to set the Q factor of the inductor that i use in simulation. The freq response of the filter now is much better. But the bandwidth is uneven as i can see a wider band at upper freq. What should i do?

You should add a resistor in series with each inductor. These resistors represent the resistive losses in the coil and each coil together with each resistor becomes a better model of a real inductor. The resistor value should be calculated by dividing the inductive reactance of the coil by its Q, and you can assume a Q of 80 for this.
Next, you can tune the resonant frequency of each of the three tanks circuits to seek the most desireable response. Tuning means to vary either the inductance or capacitance value of each tank circuit a small amount and also the value of the top-coupling capacitors. What is the most desireable response? If you are interested in passing only one sinusoid, at 1.00 MHz and you know this sinusoid is stable, then you can tune your circuit for the most extremely narrow response. If you considered it wise to have a bit more bandwidth in your filter to deal with variabilities then you might choose to tune one tank to exactly 1.000, another tank to 1.001 and another to .999 MHz to flatten the response. This final tuning is up to you.

Do not be concerned with a bit of assymettry to the response as you point out. This is harmless. Also, pay no attention to the response below -50 dB because when this filter is built, there will be unexpected parasitic coupling that will limit the response and not allow it to fall below -50 or -60 dB. So, your plot showing attenuation beyond -100 dB is not realistic.

This filter is working well.

I think that it would be a lot more informative if your x axis were expanded to show performance mainly from 100Khz to 10 MHz.

 

[Harros]

I can observe that the quality factor for most of the inductors sold in the market are around 35 to 45, should i put the corresponding Q of the particular inductor rather than the Q of 80? I am now working on designing the amplifier, is it alright for me to use 2n3904 in the amplifier?

I am now satisfied with the filter response (please refer to the pics, and circuit)... How do you think? I made the capacitors in the CL tank to be in parallel as i cant get a component with such a exact value , is it alright for us to do so?

Again, whats the type of connector that i should use to connect all the modules?

 

[RadioRon]

I can observe that the quality factor for most of the inductors sold in the market are around 35 to 45, should i put the corresponding Q of the particular inductor rather than the Q of 80? I am now working on designing the amplifier, is it alright for me to use 2n3904 in the amplifier?

I am now satisfied with the filter response (please refer to the pics, and circuit)... How do you think? I made the capacitors in the CL tank to be in parallel as i cant get a component with such a exact value , is it alright for us to do so?

Again, whats the type of connector that i should use to connect all the modules?

I agree that the filter response is now good. It is reasonable to use the Q of the actual inductors that you might use, so 45 would be OK to put in the sim. The 2N3904 is an adequate choice, although I think there might be better choices. For example, the 2N2222 might be a bit better. There are many popular types that would work well at this frequency. Many use a jfet such as U310, or 2N4416, or MPF102. I suggest other types because I don't often see the 2N3904 used at high frequencies, whereas the 2N2222 is quite popular.

You may put capacitors in parallel to achieve an unusual value, that will work fine. Many designers might choose a variable capacitor instead. One good reason for this is that the sim will not be perfectly accurate and you will find yourself wanting to vary the capacitor value a little bit anyway and a variable type makes this much easier.

The connector type is dependent on what exact types are available. If you are connecting RF signal, you should use coaxial cable and coaxial connectors. DC and low frequency control lines may use any type. Size is also an important parameter. Are your modules going to be large enough to allow convenient use of, for example, BNC type connectors. These are common and easily found. But if they are too big, many designers use SMA connectors which are overkill for your application. There are many very tiny types of connector as well, but these are harder to buy.

 

[Harros]

Any recommendation on the type of the amplifier that i should build? Recently, I found a documentation about the pathloss in near field. Can we take this model in our design?

 

[RadioRon]

In general, the best kind of amplifier would be a tuned input/tuned output common emitter/common source configuration. This sort typically provides some selectivity and resistance to overload from interferers, and is common, so finding design examples should be easy.


The paper is from someone who knows considerably more about this than I do, so I accept its accuracy but find that it can be confusing. For example, he makes an error on page 4 second paragraph. In this paragraph, it is critical to pay attention to the sign of the number he provides. He says...A path loss of -6 dB is typical, while very near might result in path loss of 60 dB (which is actually path gain), but then errs in saying pathloss at 1 wavelength is positive 18 dB when I believe he meant to say -18 dB.

Also, in my experience, the gain of an antenna, at least the gain that we typically measure, is only defined in the far field. So to use such gain figures in a near field equation seems somehow wrong to me. But this is likely my lack of understanding. He does not discuss this issue at all. Usually when we talk about near field between antennas, we use the term coupling and not gain. Not sure how this all fits together.

He also ignores the issue of what happens when you don't have co-polarization. This will become critical I think and will degrade the numbers he is predicting. However, it seems that you can control polarization so perhaps I'm worried about nothing here. Except for this critical point. Are you planning a tx antenna that emits both electric and magetic field? If your tx antenna is a simple loop, it will not have much electric field close to the antenna. If it is a short monopole or dipole, it won't have much magnetic field. What tx antenna do you plan?

 

[Harros]

The tx antenna that i am going to use is a ferrite core loop antenna, which can be found in the old radio. Is it alright to use this type of antenna? By the way, there is another documentation that i found yesterday, it is about the work done by the inventor of this distance measuring tech, but the prototype is working at 10.7MHz. What a coincidence, he use the same phase detector!
**broken link removed**

For the rx Rod Antenna, can I just implement this impedance buffer amplifier without redesigning it? **broken link removed**

 

[RadioRon]

The tx antenna that i am going to use is a ferrite core loop antenna, which can be found in the old radio. Is it alright to use this type of antenna? By the way, there is another documentation that i found yesterday, it is about the work done by the inventor of this distance measuring tech, but the prototype is working at 10.7MHz. What a coincidence, he use the same phase detector!
**broken link removed**

For the rx Rod Antenna, can I just implement this impedance buffer amplifier without redesigning it? **broken link removed**

Yes I think that impedance buffer will be fine without changes. You must select the appropriate coil inductance and they suggest 470 uH which seems OK to me.

I think that the TX antenna will be the subject of some experiments to find the best. If you use a ferrite core loop antenna, you will get mostly magnetic field and little electric field. I cannot immediately think of a form that will give equal electric and magnetic fields in an electrically small antenna so we might as well start with this loop and give it a try. According to that paper, the E field starts off with higher gain at short distance anyway, so it is ok to give an advantage to the H field by using the loop.

 

[Harros]

Hi, back to the 1MHz Oscillator, the output of the 1MHz Crystal Oscillator seems to be a square wave rather than a sine wave... Is there anything can be done to produce sine wave from this crystal oscillator. By the way, i cant find a 1MHz crystal in my place but the 1 MHz Crystal Oscillator. The lowest frequency of the crystal at my place is 1.8MHz (for UART communication use)... I am now out of idea on the sine wave oscillator that i should build...

I have an idea where the output of the crystal oscillator is connected to a 5th order Cheybyshev LPF to filter out all the harmonics and obtain the pure sine wave. I have done the simulation, it seems working fine (It functions just like a Pierre Oscillator here). However, i used a pulse voltage source to replace the crystal oscillator in simulation (this is because i cant find a crystal oscillator in the simulator). Is there any inaccuracy of this simulation? From the simulation, the waveform is located at the positive part of the v versus time plot. Should we do clamping on it to move down the waveform?

I cant find the data about output current of the crystal oscillator from the data sheet, how can i know the output current of the crystal oscillator?

Besides, i have got the rx loop antenna for H field detection. It is a AM loop antenna made by AIWA. I have enclosed the diagram here. Any comment on this antenna?

 

[RadioRon]

Hi, back to the 1MHz Oscillator, the output of the 1MHz Crystal Oscillator seems to be a square wave rather than a sine wave... Is there anything can be done to produce sine wave from this crystal oscillator. By the way, i cant find a 1MHz crystal in my place but the 1 MHz Crystal Oscillator. The lowest frequency of the crystal at my place is 1.8MHz (for UART communication use)... I am now out of idea on the sine wave oscillator that i should build...

I have an idea where the output of the crystal oscillator is connected to a 5th order Cheybyshev LPF to filter out all the harmonics and obtain the pure sine wave. I have done the simulation, it seems working fine (It functions just like a Pierre Oscillator here). However, i used a pulse voltage source to replace the crystal oscillator in simulation (this is because i cant find a crystal oscillator in the simulator). Is there any inaccuracy of this simulation? From the simulation, the waveform is located at the positive part of the v versus time plot. Should we do clamping on it to move down the waveform?

I cant find the data about output current of the crystal oscillator from the data sheet, how can i know the output current of the crystal oscillator?

Besides, i have got the rx loop antenna for H field detection. It is a AM loop antenna made by AIWA. I have enclosed the diagram here. Any comment on this antenna?

As you have discovered, it is commonly done to build a square wave oscillator and then simply filter out the harmonics. Your simulation is correct and your approach is good. Clamping is not necessary, you only need to use a DC blocking capacitor, ceramic type.

The crystal oscillator is rated to drive 50 pF capacitance or 10 TTL loads. The term "TTL load" is a standard way of saying how much current this device can deliver. See this thread:
https://forum.allaboutcircuits.com/showthread.php?t=8471
and read the post from Papabravo.

That AM loop will work OK. You can also use a ferrite rod AM antenna as is usually found inside small portable receivers. This has the advantage of being smaller than your antenna.

 

[Harros]

As you have discovered, it is commonly done to build a square wave oscillator and then simply filter out the harmonics. Your simulation is correct and your approach is good. Clamping is not necessary, you only need to use a DC blocking capacitor, ceramic type.

The crystal oscillator is rated to drive 50 pF capacitance of 10 TTL loads. The term "TTL load" is a standard way of saying how much current this device can deliver. See this thread:
https://forum.allaboutcircuits.com/showthread.php?t=8471
and read the post from Papabravo.

That AM loop will work OK. You can also use a ferrite rod AM antenna as is usually found inside small portable receivers. This has the advantage of being smaller than your antenna.

I have used the ferrite rod AM antenna for the transmitter antenna. Is it alright to use the same antenna at the receiver as well?

 

[RadioRon]

I have used the ferrite rod AM antenna for the transmitter antenna. Is it alright to use the same antenna at the receiver as well?

You remind me that it may be inappropriate to use a ferrite rod antenna for transmitting. The reason is that the ferrite rod may saturate with the amount of magnetic field that your transmitter will send out. For this reason, I think it may be best to avoid the ferrite rod for your transmitter and use your air-core loop antenna for that instead. The receiver, on the other hand, is receiving very tiny amounts of power, amounts too tiny to allow the rod to saturate, so the ferrite rod is very well suited to the receiver. After all, it is commonly used in this way.

 

[Harros]

You remind me that it may be inappropriate to use a ferrite rod antenna for transmitting. The reason is that the ferrite rod may saturate with the amount of magnetic field that your transmitter will send out. For this reason, I think it may be best to avoid the ferrite rod for your transmitter and use your air-core loop antenna for that instead. The receiver, on the other hand, is receiving very tiny amounts of power, amounts too tiny to allow the rod to saturate, so the ferrite rod is very well suited to the receiver. After all, it is commonly used in this way.

By the way, i am thinking of making the transmitter antenna to be in small sized (portable preferably). So, should i build the air core loop antenna on my own? Do you have any reference on the formula of the number of turns for the air-core loop stick antenna? What type of wire that i should use to wind the loop stick antenna?

How to infer the impedance of the crystal oscillator?

 

[RadioRon]

By the way, i am thinking of making the transmitter antenna to be in small sized (portable preferably). So, should i build the air core loop antenna on my own? Do you have any reference on the formula of the number of turns for the air-core loop stick antenna? What type of wire that i should use to wind the loop stick antenna?

How to infer the impedance of the crystal oscillator?

The air core antenna size and wire type is not very sensitive to variations and is very simple so yes you should make your own. Choose a size that you prefer, but remember that size relates directly to range. Find something non-conductive to use as a core and choose any convenient wire that has insulation on it and it a reasonable diameter to use for your size. It doesn't matter if it is stranded or solid, but copper is best. If your loop is quite small, like less than 2 inches across, you might consider using enamelled copper wire. This type has no plastic jacket on it, the only insulation is a thin layer of clear enamel which allows it to be wound quite closely on a form.

When deciding how many turns of wire, consider also how you plan to couple energy into this loop. Are you going to tap this coil or are you going to add a secondary loop?

Choose a number of turns of wire that will give you an inductance of about 10 microhenries. Use an estimator to help:
https://www.daycounter.com/Calculators/Air-Core-Inductor-Calculator.phtml
Will you resonate this antenna with a parallel capacitor?

The output impedance of the oscillator is inferred using ohm's law. The specification tells you what the output peak to peak voltage is, plus it tells you that the output drive capability is 10 ttl loads which lets you estimate the current output while delivering that peak to peak voltage. Now, just calculate using ohm's law and this voltage and current. This will give you a resistance.

 

[Harros]

I am going to add the secondary loop, as the primary loop will be terminated using a tuning capacitor... But i cant get a variable capacitor with high capacitance range. Can i add a fixed capacitor in parallel to the variable capacitor, and use that variable capacitor to fine tune the selectivity of the loop antenna? How can i determine the value of the fixed capacitor?

I am thinking of using the pvc pipe to build this antenna core. Is it alright to do so?

Regarding the output impedance of the crystal oscillator, the impedance is quite high: 4.7V / 16mA = 293k ohm.... Am i correct in calculating the output impedance?

 

[RadioRon]

I am going to add the secondary loop, as the primary loop will be terminated using a tuning capacitor... But i cant get a variable capacitor with high capacitance range. Can i add a fixed capacitor in parallel to the variable capacitor, and use that variable capacitor to fine tune the selectivity of the loop antenna? How can i determine the value of the fixed capacitor?

I am thinking of using the pvc pipe to build this antenna core. Is it alright to do so?

Regarding the output impedance of the crystal oscillator, the impedance is quite high: 4.7V / 16mA = 293k ohm.... Am i correct in calculating the output impedance?

It is ok to use a fixed capacitor in parallel with a variable one. Use a ceramic type for the fixed one. Make sure the variable one is about 10% of the total capacitance, or more, so that it gives you a reasonable tuning range. The total capacitance will be derived from the equation:

Fr=1/[2 x pi x sqrt(LC)] where Fr is the resonant frequency of the antenna, L is the inductance and C is the total parallel capacitance. This is an approximation since the secondary coil will shift the result a little bit, but this equation is the best starting point. PVC pipe is a good choice.

Your arithmetic about the output impedance is incorrect. Try again. You should end up with 293 ohms, not 293K ohms. This value is a simplification of the actual output impedance and is not exactly correct, but it is probably close enough to allow you to design your filter. The actual output impedance changes depending on which part of the waveform we are looking at. This is the result of using a limited signal (ie. square wave). But an approximation is fine, so this is not something to worry about. In fact, to be more accurate it would be better to estimate the RMS output voltage rather than the peak to peak voltage before calculating resistance. But since the output current is also an estimate it might be OK to not worry about it. If there is an error in the output resistance, I think that we are somewhat high and the actual result might be a bit lower. But go with 293 ohms and I think it will be close enough.

 

[Harros]

Yes, you are correct. I misread "mili" as "micro"...


I am going to couple energy to the loop using a secondary loop, I have used the inductance calculator to calculate the number of turns needed. There are 43 turns for the loop with 1 inch diameter and 4 inch length. Does the size of the enameled copper wire used affect the inductance of the loop?

How can i infer the number of turns needed for the secondary loop? How should i calculate the input impedance of the secondary loop?

By the way, for a 10uH inductor, a 2.5nF capacitor is needed for frequency tuning. However, the value of a variable capacitor ranges from few pF to about 100 pF... So, should i increase the inductance of the inductor in order to reduce the capacitance of the capacitor used?

 

[RadioRon]

Yes, you are correct. I misread "mili" as "micro"...


I am going to couple energy to the loop using a secondary loop, I have used the inductance calculator to calculate the number of turns needed. There are 43 turns for the loop with 1 inch diameter and 4 inch length. Does the size of the enameled copper wire used affect the inductance of the loop?

How can i infer the number of turns needed for the secondary loop? How should i calculate the input impedance of the secondary loop?

By the way, for a 10uH inductor, a 2.5nF capacitor is needed for frequency tuning. However, the value of a variable capacitor ranges from few pF to about 100 pF... So, should i increase the inductance of the inductor in order to reduce the capacitance of the capacitor used?

The diameter of the copper wire will have only a very small affect on the inductance until it so thick that you can't fit it onto the four inch form. The thickness will, however, have a significant affect on the inductor Q.

First you can calculate the resonant resistance of the primary tank circuit by determining the inductor Q, then calculating the equivalent series resistance of the coil, then converting this series R to a parallel R. This parallel R is the resonant resistance of the tank circuit since the reactance of the inductor cancels the reactance of the capacitor. If there was no resistance in the circuit at all, then the resistance at resonance is infinite. But practically, a rule of thumb value for the tank circuit might be estimated to be 3 Kohms. But go through the calculation to see what it should be.

Once you have an idea of what the tank resistance is at resonance, you can then study the concept of the loaded Q of a resonant circuit. This term "loaded" refers to the effect of putting additional resistance into the tank by tapping or coupling as we are doing. You will find that it is not too difficult to use the turns ratio of primary to secondary of your coupling method and the load impedance you will be attaching to the secondary loopo (50 ohms?) to find out how your secondary loop with its output load will reduce the net parallel resistance of the tank circuit.

Yes, it seems that my initial guess of the inductance value is bit low and you can increase it.