Polarization of 3DR Radio antenna

Hi there Carl,

Answering your question inline here as the silly comment wall does not allow a post of more than 5000 char....

I read your blog post about “Antenna Design Overview” and wanted to ask you the 1.5:1 VSWR. You write that it corresponds to a transmission efficiency of approx. 3%. When I did the calculations I got exactly 4%. Not that 3% is not approx 4% or that it makes a difference, I just wanted to check since I am measuring my own antenna and wanted to do the calculations correct. However I found another of your posts and there you write that it is 4% so I guess my calculations were correct.

Yes, to be more 'correct, 4% is closer to mathematically correct..I just have a short form calculator in my head that gets me 'close enough'....

Your questions:
Q1)

From one of your posts:

“The most important is to ensure the antenna is resonant at the desired frequency, that its polarisation and gain is appropriate for the application, that if it requires a BALUN one is present, and that the SWR is 'acceptable'.”

What I don’t understand here is the difference between a good SWR and a resonant antenna. In my bachelor thesis I built a 2.15 GHz MIMO antenna (4 dipoles and a ground plane). To tune the antennas I measured S11 while cutting them shorter and shorter until the “dip” in the reflection coefficient was at the correct freq. I thought that this tuning of S11 was equivalent to tuning VSWR and antenna resonance?

What you were doing was altering the antenna resonant frequency, which changed the feedpoint impedance as well - a by-product of the trimming procedure.

A 50ohm antenna feedpoint match gives you a good reflection coefficient, but that has nothing to do with the antenna resonant frequency. It simple means that the feedpoint is close to 50ohms resistive. Trimming the antenna length affects the feedpoint impedance, and trimming the length in the 'correct direction' may bring the feedpoint closer to 50ohms, but the antenna resonance can be altogether incorrect.

Imagine a 1/2 wave length of Aluminium tubing, the length of which is resonant at say 300MHz ( how we determined it is resonant we will discuss later on). Now lets feed it with some RF, and try to arrange the feed such that we have a good 500ohm feed impedance, and so a good SWR. A neat way to do this ( there are many ways..) is , for example, the Delta Match. This is simply a connection of the 2 'wires' of the antenna feedline to the antenna itself,in a sort of Y connection, each feedline wire connected to the antenna tubing spaced some distance from the tubing center point. If we connect both wires of the feedline exactly to the center of the tubing, that would be an RF short circuit, ie, 0 ohms. Now if we start to move each feedline wire away from the center, ie, one wire say a few cm left of center, and the other wire the same distance right of center, and we keep doing this while measuring S11, we will find a point where the impedance is 50ohms. The impedance will increase as we keep moving the feedline connection points further from the center. At the 50ohm point the SWR is optimum, and the antenna feed is matched - this WITHOUT affecting the antenna resonant length. The antenna is at its optimum - it is properly matched, AND it is resonant. Since the distance from center point for a 50ohm match is a proportion of the antenna tubing length, we could all achieve a match by trimming the tubing length - we would then at some point achieve a similar proportion of impedance without actually moving the feedline connecting positions. HOWEVER, we have at the same time altered the antenna resonant length, so the antenna no longer radiates efficiently at the desired frequency.

An antenna will only radiate at maximum efficiency when it is resonant - the feedpoint impedance does not need to be 50ohms for this to be. The only reason for the magic 50ohm feedpoint match is because a 'standard' impedance evolved and most Transmitters and Receivers todays exhibit a 50ohm input/output impedance. And for efficient power transfer, the impedance of the transmitter and its load ( the antenna) should be the same. ( there are other 'standard' impedances in this game as well - 300ohm, common in old TV systems, where the feedline was 300ohm twinlead and the antenna a folded dipole style - also 75ohm, as most TV and TV antenna are today..)

Regarding Resonance:

For the lower frequencies ( HF up to lower UHF) it is quite easy to determine Resonance to an acceptable accuracy. Replace the feedline at the feedpoint with a 1 or 2 turn coil ( this is oversimplified - the techniques are easily found in any good Ham radio handbook) and use a Grid Dip Meter to determine the dip frequency - this is the resonant frequency. The antenna can then be trimmed to get is spot on the required frequency. DO NOT trim any more when matching the antenna - now you have to match by other means - Stub, Gamma, Delta, Hairpin, or any other of a myriad of matching methods. A dipole is easily matched to 50ohm unbalanced feeder ( coax) by means of an inductive stub ( you will still need to apply a Balun or RF choke, to eliminate the RF currents on the Coax outer braid)

Determining resonance at the higher frequencies is more complex - a grid dip meter no longer provides useful readings.
Q2)

I have seen that a lot of people are using CP antennas for FPV but I have not found many that uses it for telemetry. Why is that? Is it because the video link is analog and more sensitive to multipath interference than the digital telemetry link?

I presume when you say 'FPV' you mean the live video transmission? If there is such a trend amongst the uninitiated, I would propose that it is due to antenna size - I believe that the video transmission are mostly in the 2.4GHz to 5.8GHz range, while the Telemetry is in the 433MHz range? if so, CP antenna for 433MHz are large and unwieldy - also the multipath issues at the VHF and UHF frequencies are less problematic.
Q3)

I posted a question about the polarization of the standard 3DR Radio 433 MHz in the antenna forum but I repeat it here:

I thought that the antenna was a metal straight wire but when I removed the plastic coating I found that the antenna is a small helical. In the far field a small circular loop is horizontally polarized. So this means that if I have the standard 3DR Antenna on a transmitter and a thin metal wire antenna on a receiver, the antennas should be orthogonal to each other for best performance. I this correct?

No, not so. The small helical you refer to works in the 'normal mode' - ie, it is vertically polarised if held such, as would be a vertical wire whip. Any helical antenna having a pitch spacing and helix diameter less than 0.1 wavelength is a normal mode helix. Such a helix radiates linear polarisation, in the plane of orientation. The 'other' helical, the axial mode helical, is a CP antenna and radiation is in the direction it 'points'. The loop antenna is also a linear polarised antenna, radiating as a dipole does, but with the direction reversed , ie, there is a sharp null to the signal when viewed through the center of the loop.

In your case, the two antenna must be matched in orientation for max signal coupling, ie, vertical to vertical, etc.

Also, your statement -In the far field a small circular loop is horizontally polarized. - obviously depends on the loop orientation.

Hope that helps a little..

Regards

Joe

Hello Carl,

The analysis of the Normal Mode Helix is actually a little more subtle than your approach.

For 'mostly linear and vertical radiation', the Normal mode helix must be held vertical, and the helix diameter and spacing between turns should be around 0.1 lambda max ( alpha then being the pitch angle of each turn of the helix)

Then there are basically three limiting cases of such an antenna -

When alpha = 0deg the antenna reduces to a single horizontal loop, with the obvious radiation pattern ( null through the centre and horizontal linear polarization.

When alpha > 0deg but less than 90deg - then each 'loop' can be treated as a loop with an attached vertical dipole element. The length of the dipole element depends on the value of alpha. This combination now radiates a more elliptical polarisation. The closer apha is to zero, the more horizontal and the closer to 90deg the more vertical the polarisation will be.

When alpha = 90deg, the antenna reduces to a plain vertical dipole, with obvious radiation characteristics.

The detailed analysis of this antenna is correctly and concisely presented in 'ANTENNAS' - John D Kraus, Chapter 7-19, pages 334 onwards. ( you do have Kraus.....)

The bottom line is that if the antenna is more than one turn, and the turns are not co-planar, the radiation will not be horizontally polarized, but somewhere between horizontal and vertical, depending on alpha...and the antenna still has to be resonant...All in all a poor antenna for a situation where you want some horizontal omnidirectional gain, and some efficiency.

good stuff, antennas!

Regards

Joe

Replies

Gisela & Joe Noci April 4, 2015 at 2:24pm

Hi there Carl,

Answering your question inline here as the silly comment wall does not allow a post of more than 5000 char....

I read your blog post about “Antenna Design Overview” and wanted to ask you the 1.5:1 VSWR. You write that it corresponds to a transmission efficiency of approx. 3%. When I did the calculations I got exactly 4%. Not that 3% is not approx 4% or that it makes a difference, I just wanted to check since I am measuring my own antenna and wanted to do the calculations correct. However I found another of your posts and there you write that it is 4% so I guess my calculations were correct.

Yes, to be more 'correct, 4% is closer to mathematically correct..I just have a short form calculator in my head that gets me 'close enough'....

Your questions:
Q1)

From one of your posts:

“The most important is to ensure the antenna is resonant at the desired frequency, that its polarisation and gain is appropriate for the application, that if it requires a BALUN one is present, and that the SWR is 'acceptable'.”

What I don’t understand here is the difference between a good SWR and a resonant antenna. In my bachelor thesis I built a 2.15 GHz MIMO antenna (4 dipoles and a ground plane). To tune the antennas I measured S11 while cutting them shorter and shorter until the “dip” in the reflection coefficient was at the correct freq. I thought that this tuning of S11 was equivalent to tuning VSWR and antenna resonance?

What you were doing was altering the antenna resonant frequency, which changed the feedpoint impedance as well - a by-product of the trimming procedure.

A 50ohm antenna feedpoint match gives you a good reflection coefficient, but that has nothing to do with the antenna resonant frequency. It simple means that the feedpoint is close to 50ohms resistive. Trimming the antenna length affects the feedpoint impedance, and trimming the length in the 'correct direction' may bring the feedpoint closer to 50ohms, but the antenna resonance can be altogether incorrect.

Imagine a 1/2 wave length of Aluminium tubing, the length of which is resonant at say 300MHz ( how we determined it is resonant we will discuss later on). Now lets feed it with some RF, and try to arrange the feed such that we have a good 500ohm feed impedance, and so a good SWR. A neat way to do this ( there are many ways..) is , for example, the Delta Match. This is simply a connection of the 2 'wires' of the antenna feedline to the antenna itself,in a sort of Y connection, each feedline wire connected to the antenna tubing spaced some distance from the tubing center point. If we connect both wires of the feedline exactly to the center of the tubing, that would be an RF short circuit, ie, 0 ohms. Now if we start to move each feedline wire away from the center, ie, one wire say a few cm left of center, and the other wire the same distance right of center, and we keep doing this while measuring S11, we will find a point where the impedance is 50ohms. The impedance will increase as we keep moving the feedline connection points further from the center. At the 50ohm point the SWR is optimum, and the antenna feed is matched - this WITHOUT affecting the antenna resonant length. The antenna is at its optimum - it is properly matched, AND it is resonant. Since the distance from center point for a 50ohm match is a proportion of the antenna tubing length, we could all achieve a match by trimming the tubing length - we would then at some point achieve a similar proportion of impedance without actually moving the feedline connecting positions. HOWEVER, we have at the same time altered the antenna resonant length, so the antenna no longer radiates efficiently at the desired frequency.

An antenna will only radiate at maximum efficiency when it is resonant - the feedpoint impedance does not need to be 50ohms for this to be. The only reason for the magic 50ohm feedpoint match is because a 'standard' impedance evolved and most Transmitters and Receivers todays exhibit a 50ohm input/output impedance. And for efficient power transfer, the impedance of the transmitter and its load ( the antenna) should be the same. ( there are other 'standard' impedances in this game as well - 300ohm, common in old TV systems, where the feedline was 300ohm twinlead and the antenna a folded dipole style - also 75ohm, as most TV and TV antenna are today..)

Regarding Resonance:

For the lower frequencies ( HF up to lower UHF) it is quite easy to determine Resonance to an acceptable accuracy. Replace the feedline at the feedpoint with a 1 or 2 turn coil ( this is oversimplified - the techniques are easily found in any good Ham radio handbook) and use a Grid Dip Meter to determine the dip frequency - this is the resonant frequency. The antenna can then be trimmed to get is spot on the required frequency. DO NOT trim any more when matching the antenna - now you have to match by other means - Stub, Gamma, Delta, Hairpin, or any other of a myriad of matching methods. A dipole is easily matched to 50ohm unbalanced feeder ( coax) by means of an inductive stub ( you will still need to apply a Balun or RF choke, to eliminate the RF currents on the Coax outer braid)

Determining resonance at the higher frequencies is more complex - a grid dip meter no longer provides useful readings.
Q2)

I have seen that a lot of people are using CP antennas for FPV but I have not found many that uses it for telemetry. Why is that? Is it because the video link is analog and more sensitive to multipath interference than the digital telemetry link?

I presume when you say 'FPV' you mean the live video transmission? If there is such a trend amongst the uninitiated, I would propose that it is due to antenna size - I believe that the video transmission are mostly in the 2.4GHz to 5.8GHz range, while the Telemetry is in the 433MHz range? if so, CP antenna for 433MHz are large and unwieldy - also the multipath issues at the VHF and UHF frequencies are less problematic.
Q3)

I posted a question about the polarization of the standard 3DR Radio 433 MHz in the antenna forum but I repeat it here:

I thought that the antenna was a metal straight wire but when I removed the plastic coating I found that the antenna is a small helical. In the far field a small circular loop is horizontally polarized. So this means that if I have the standard 3DR Antenna on a transmitter and a thin metal wire antenna on a receiver, the antennas should be orthogonal to each other for best performance. I this correct?

No, not so. The small helical you refer to works in the 'normal mode' - ie, it is vertically polarised if held such, as would be a vertical wire whip. Any helical antenna having a pitch spacing and helix diameter less than 0.1 wavelength is a normal mode helix. Such a helix radiates linear polarisation, in the plane of orientation. The 'other' helical, the axial mode helical, is a CP antenna and radiation is in the direction it 'points'. The loop antenna is also a linear polarised antenna, radiating as a dipole does, but with the direction reversed , ie, there is a sharp null to the signal when viewed through the center of the loop.

In your case, the two antenna must be matched in orientation for max signal coupling, ie, vertical to vertical, etc.

Also, your statement -In the far field a small circular loop is horizontally polarized. - obviously depends on the loop orientation.

Hope that helps a little..

Regards

Joe
- Ikaros > Gisela & Joe Noci April 5, 2015 at 1:58am
  
  Dear Joe,
  
  Thank you so much for the answers. I will study them in detail and come back if (when) I have further questions. However, for the last one I can comment right away. Either we mean the same thing or I am missing something.
  
  Cartesian coordinate system, theta is angle from z axis to xy-plane, phi is angle from x axis to y axis.
  
  I place the 3DR antenna in the origin along the z axis, the turns of the helical antenna is going around the z axis. To understand the helical antenna in normal mode (radius much less than lambda) it is sufficient to analyse one turn. The far field function of a small loop antenna parallel to the xy plane is similar to a electric dipole, donut shaped with a noll in the direction of the z axis. However, as far as I understand for the electric dipole the electric field is directed in theta direction, i.e. vertically polarized. But for the small loop antenna the electrical field is in phi direction, i.e. horizontal polarization. E.g. if we look in the direction of the x axis the E-field will be directed parallel to the z axis for the dipole and parallel to the y axis for the loop. Equation (5.72) in “Foundations of Antennas” by Per-Simon Kildal, equation (5-19b) in “Antenna Theory Third Edition” by Constantine A. Balanis.
  
  So this means that if I want to communicate between the loop antenna in the origin parallel to the xy plane and a dipole at let say (x=10, y=0, z=0) the dipole should be parallel to the y axis.
  
  Is this correct or all wrong? :)
  
  Thank you again for taking the time!
  
  Best Regards,
  
  Carl
  - Gisela & Joe Noci > Ikaros April 5, 2015 at 1:03pm
    
    Hello Carl,
    
    The analysis of the Normal Mode Helix is actually a little more subtle than your approach.
    
    For 'mostly linear and vertical radiation', the Normal mode helix must be held vertical, and the helix diameter and spacing between turns should be around 0.1 lambda max ( alpha then being the pitch angle of each turn of the helix)
    
    Then there are basically three limiting cases of such an antenna -
    
    When alpha = 0deg the antenna reduces to a single horizontal loop, with the obvious radiation pattern ( null through the centre and horizontal linear polarization.
    
    When alpha > 0deg but less than 90deg - then each 'loop' can be treated as a loop with an attached vertical dipole element. The length of the dipole element depends on the value of alpha. This combination now radiates a more elliptical polarisation. The closer apha is to zero, the more horizontal and the closer to 90deg the more vertical the polarisation will be.
    
    When alpha = 90deg, the antenna reduces to a plain vertical dipole, with obvious radiation characteristics.
    
    The detailed analysis of this antenna is correctly and concisely presented in 'ANTENNAS' - John D Kraus, Chapter 7-19, pages 334 onwards. ( you do have Kraus.....)
    
    The bottom line is that if the antenna is more than one turn, and the turns are not co-planar, the radiation will not be horizontally polarized, but somewhere between horizontal and vertical, depending on alpha...and the antenna still has to be resonant...All in all a poor antenna for a situation where you want some horizontal omnidirectional gain, and some efficiency.
    
    good stuff, antennas!
    
    Regards
    
    Joe
    - Ikaros > Gisela & Joe Noci April 6, 2015 at 12:55am
      
      That is a very clear and informative answer!
      
      The detailed analysis of this antenna is correctly and concisely presented in 'ANTENNAS' - John D Kraus, Chapter 7-19, pages 334 onwards. ( you do have Kraus.....)
      
      Found it in the library :-)
      
      Thank you!
      
      Regards
      
      Carl

This reply was deleted.

Discussions

Polarization of 3DR Radio antenna

Replies

Activity