Call Search
     

New to Ham Radio?
My Profile

Community
Articles
Forums
News
Reviews
Friends Remembered
Strays
Survey Question

Operating
Contesting
DX Cluster Spots
Propagation

Resources
Calendar
Classifieds
Ham Exams
Ham Links
List Archives
News Articles
Product Reviews
QSL Managers

Site Info
eHam Help (FAQ)
Support the site
The eHam Team
Advertising Info
Vision Statement
About eHam.net

   Home   Help Search  
Pages: [1] 2 Next   Go Down
  Print  
Author Topic: Testing a loop antenna  (Read 3404 times)
KI6NUJ
Member

Posts: 118


WWW

Ignore
« on: March 17, 2008, 01:47:27 AM »

Hi,

Is there a way to test the main loop and the feeding Faraday loop of a loop antenna, separately, to ensure that they are functioning correctly? Any tests I can run on each independently to isolate any faulty parts or configuration (like loop length) etc?

Thanks,

- Siddhartha
Logged
W5FYI
Member

Posts: 1044




Ignore
« Reply #1 on: March 17, 2008, 04:47:04 AM »

I assume you are talking about testing a "magnetic" loop. If so, the main loop should easily dip a grid dip meter without the feed loop, but I don't think the feed Faraday-shielded loop can be tested independently. It needs to be coupled to the large loop in order for it to properly respond to testing. Is there a reason for you wanting to test them separately?
Logged
KL7AJ
Member

Posts: 329


WWW

Ignore
« Reply #2 on: March 17, 2008, 09:29:58 AM »

I concur with the grid-dip meter....the handiest gadget on the planet.

The Faraday shielding effectiveness is difficult to test...however if you just want to test the capacitance to the coil, you should be able to do that with a capacitance bridge, or even a modern DVM, if it's above 100pf or so.

Of course, the proof is in the pudding.  If you have a means of generating a weak signal (again, a grid dip oscillator is great for this) at a thousand feet away or so, you can test the directivity (or null) quite nicely.

eric
Logged
VK1OD
Member

Posts: 1697




Ignore
« Reply #3 on: March 18, 2008, 12:54:00 AM »

Siddhartha,

There is no magic to the so called Faraday Loop or Shielded Loop. The explanations that assert that it responds only to magnetic field and not electric field due to the 'shielding effect' are bunk.

The small loop behaves just like a small loop with the coax feed attached to the gap in the ring formed by the outer conductor. Depending on whether the centre conductor of the coax connects to the shield at the gap, or goes right around to the T you may have an inductive stub in series at the feed point (in the case of the latter), but the antenna is the outside surface of the outside conductor of the coax ring. The advantage of this feed configuration is better symmetry / balance of the loop so that common mode feed line current is minimised.

On its own, the small loop series impedance will have a modest inductive reactance, and a small resistance (comprising a very small radiation resistance and a small loss resistance). This is not very interesting in terms of the total system.

The main loop should show a strong resonance by itself, as might be sensed with a loosley coupled GDO.

With the total system, the main loop should dominate behavior, and you should observe a distinct resonance (zero reactance) looking into the coax at the T of the feed loop. If the resistance at resonance is too small (wrt 50), make the coupling loop smaller and vice versa. The large loop and small loop act like an impedance step up transformer (with ratio approximately equal to the inverse of the square of the ratio of loop areas).

Don't forget that if you measure the impedance looking into the coax at any distance from the T and if VSWR>1, true feed point impedance (ie at the T) will be transformed and observed resonance will not coincide with resonance of the loop. If you make measurements remote from the T, do you know how to find Z at the T?

Does this help?

Owen
Logged
VK1OD
Member

Posts: 1697




Ignore
« Reply #4 on: March 18, 2008, 11:54:28 AM »


Siddhartha,

To work a little example.

Lets say you were making a transmitting loop for 80m, and you chose a 1m dia circular main loop.

Lets assume that it requires a circular feed loop of 0.2m dia, and you will connect the centre conductor of the coax forming the shielded loop to the shield at the T.

Considering the feed loop in isolate (ie, not coupled to the main loop):

The inductance of the loop of RG58C/U at the gap is about 1.1µH, its reactance at 3.6MHz is about 25. The radiation resistance is dwarfed by the loss resistance, and total resistance will be milliohms.

Now, a 0.314m s/c inductive stub appears in series with the gap, and its impedance is about j1.8 with some milliohms of series resistance.

The combined impedance of j26.8 with some milliohms of series resistance is transformed by the half loop of coax to perhaps something like 0.1+j29 at the T.

So, the small loop looks like a small inductive reactance with a series resistance that is so low you probably won't be able to measure it.

Remember, if you measure it at the end of a further length of transmission line, it will be different because the load is transformed by the transmission line.

Now when you couple the feed loop to the main loop, the impedance will be dominated by the main loop as I noted in my earlier post.

Owen

PS: I hope I got the numbers right, this forum format being non revisable is not really suited to this kind of discussion.
Logged
VK1OD
Member

Posts: 1697




Ignore
« Reply #5 on: March 18, 2008, 12:39:39 PM »


An inevitible mistake:

Considering the feed loop in isolation (ie, not coupled to the main loop):

The inductance of the loop of RG58C/U at the gap is about 0.55µH, its reactance at 3.6MHz is about 13. The radiation resistance is dwarfed by the loss resistance, and total resistance will be milliohms.

Now, a 0.314m s/c inductive stub appears in series with the gap, and its impedance is about j1.8 with some milliohms of series resistance.

The combined impedance of j14.8 with some milliohms of series resistance is transformed by the half loop of coax to perhaps something like 0.1+j17 at the T.

So, the small loop in isolation looks like a small inductive reactance with a series resistance that is so low you probably won't be able to measure it.
Logged
KI6NUJ
Member

Posts: 118


WWW

Ignore
« Reply #6 on: March 18, 2008, 04:34:57 PM »

Thanks for the replies, Owen.

The *loop* is a square of 4 ft x 4ft made from RG58 cable. It is a prototype that I will tweak and then migrate to a permanent copper pipe hexagon. The Faraday loop is a 1/5th piece of RG8 loop sitting opposite to a air variable capacitor (50-500pf) on the loop.

I connected the entire contraption to my TS-570D using a 12ft RG8 cable. I tried to tune it to 40m and 20m using the rough noise method. Got nothing on 20m and on 40m, I could hear some CWers chatting away when the xcvr was NOT in the CW mode!! Rotating the antenna made no difference and neither does rotating the capacitor dial (no change in noise).

I know the xcvr is fine because I took it to the local HRO store where they connected it to a good known antenna and all worked well. The main loop is really simple so I can't figure out what can go wrong there. Again, its just 4x4ft of RG58 wrapped around two wooden thin planks on opposite sides. The end at which the capacitor is soldered, there is a gap in the loop of about 3 inches. And on the opposite side of the capacitor is the faraday loop.

I will provide some more meaningful numbers once I receive the capacitance meter, MFJ207 antenna analyzer and a Heathkit impedance meter that I ordered.

Thanks,

- Siddhartha

 
Logged
VK1OD
Member

Posts: 1697




Ignore
« Reply #7 on: March 18, 2008, 04:59:59 PM »

"The *loop* is a square of 4 ft x 4ft made from RG58 cable. It is a prototype that I will tweak and then migrate to a permanent copper pipe hexagon. The Faraday loop is a 1/5th piece of RG8 loop sitting opposite to a air variable capacitor (50-500pf) on the loop.

I connected the entire contraption to my TS-570D using a 12ft RG8 cable. I tried to tune it to 40m and 20m using the rough noise method. Got nothing on 20m and on 40m, I could hear some CWers chatting away when the xcvr was NOT in the CW mode!! Rotating the antenna made no difference and neither does rotating the capacitor dial (no change in noise)."

Where did you get that design?

The main loop has an inductance of around 5µH. That will resonate at 14MHz with about 25pF, so your cap is too large.

It should resonate at 7MHz with about 100pF.

If I understand what you have done, your feed loop is 1/5 the size of the main loop, that would be 1.2/5m or  0.24m a side. You should probably hear strong signals on the isolated feed loop, and it should exhibit very pronounced nulls when the plane of the loop is at right angles to the signal source. (Remember that incoming signals at 7MHz will be from the sky.)

If that works, couple it to the main loop and you should hear a distinct peak in noise when tuned, but it is very narrow in bandwidth (ie critical in tuning, especially with such a large capacitor).

It is a waste of time trying to prototype your loop as you have in respect of impedance matching. You may learn about the concepts, but the prototype is of little use as a prototype. The R component of the main loop is mostly loss, and your loss is much higher than say 20mm dia copper pipe. Loss in tuning caps is also an issue.

Physical symmetry is critical to the balance required for the deep nulls that are usually sought. Any effort in using the 'shielded loop' feed is wasted if you don't achieve near perfect physical symmetry of the antenna system (loops, feedline, environment).

Owen



 
Logged
KI6NUJ
Member

Posts: 118


WWW

Ignore
« Reply #8 on: March 19, 2008, 01:01:44 AM »

Owen,

This design is based on g4fon's antenna here:
http://www.g4fon.net/MagLoop.htm
http://www.g4fon.net/MagLoopTwo.htm

Can you please explain the calculations that you made? I used this calculator:
http://brneurosci.org/loopfrequency.html

Here is what I get for my loop dimensions (the air variable capacitor is 15-500pf, sorry)
----------xxxxxxxxxxx-----------
Input parameters
Length of one side: 121.92
Number of turns: 1
Length of coil: 487.68
Minimum value of cap: 15
Maximum value of cap: 500

Results
Distributed capacitance: 42.0625 picofarads
Inductance : 0.342472 microhenrys
Capacitance range : 57.0625 to 542.062 picofarads
Nominal Frequency range: 12162.5 to 70220.2 kHz
Actual Frequency range : 11681.1 to 36002.5 kHz
----------xxxxxxxxxxx-----------


Since it is a three section capacitor, I can tie the three sections together for 45-1500pf, right? That gives me:

----------xxxxxxxxxxx-----------
Input parameters
Length of one side: 121.92
Number of turns: 1
Length of coil: 487.68
Minimum value of cap: 45
Maximum value of cap: 1500

Results
Distributed capacitance: 42.0625 picofarads
Inductance : 0.342472 microhenrys
Capacitance range : 87.0625 to 1542.06 picofarads
Nominal Frequency range: 7022.02 to 40541.6 kHz
Actual Frequency range : 6925.59 to 29146.9 kHz
----------xxxxxxxxxxx-----------

Again, your responses are much appreciated.

Thanks,

- Siddhartha
Logged
VK1OD
Member

Posts: 1697




Ignore
« Reply #9 on: March 19, 2008, 02:54:45 AM »


Siddhartha,

1. The design at G4FON is nothing like you have been talking about, so I assume all the time in typing up advice was wasted.

2. The G4FON design you identify is a radiating dummy load (see below), apparently justified on the basis it is for QRP.

3. The calculator you cite gives an unrealistic value for the inductance of a 1.2m square loop, the value is more like 5µH than 0.30µH. Try a few other calculators and see if you can find some consistent answers.

4. An NEC model of a loop of 1.2m sq of 3mm dia conductor in free space at 7.1MHz gives a loop reactance of around 265 ohms and radiation resistance of 0.024ohms. Conductor resistance will be around 0.4ohms, so just on that basis, the antenna is less than 6% efficient, and there is more loss to come.

5. G4FON's 5t/2t matching transformer implies the feedpoint impedance is 8ohms. The radiation resistance of the loop mounted 2m above average ground is 0.07ohms. If the feedpoint Z is indeed 8 ohms, the antenna is 0.07/8pu or 0.8% efficient (excluding loss in the matching transformer), that means a gain of -20dBi. With an antenna like this, you are radiating QRP even if you feed it 500W.

6. The above is why serious transmitting loops are made with low resistance conductors, eg 20mm dia copper tube, and great attention is paid to a tuning capacitor with extremely low equivalent series resistance (ESR).

7. You might do better to find a reputable design... though it is challenging because there is a fair bit of bunk around about loops, feed / matching arrangements, and performance.

8. Think about conservation of energy and small antennas. The power that isn't lost as heat is radiated. The challenge is to keep the heat losses to an acceptable figure... and the above antenna looks like it converts close to 99% of your transmitter power into heat.

Owen
Logged
W8JI
Member

Posts: 9304


WWW

Ignore
« Reply #10 on: March 19, 2008, 03:23:34 AM »

Having built and measured dozens of loops, some with FS measurements for FDA approvial, I agree with VK1OD.
 

You are wasting time with a "Faraday Shield" on the coupling link. That idea is a myth that never does anything. It's very easy to understand why when you look at what it does and how big it is.

Secondly, braiding has horrible RF resistance. It's worse than most people expect. As a matter of fact at 30MHz the braid from RG-8 cable when aged just a bit has about the same resistance as a number 12 or 14 solid wire. This is because current must flow over many thousands of pressure connections. The resistance when clean and well compacted is about four times the resistance of a solid wire with the same circular mills.

Third, the capacitor is a poor design. The radiation resistance of the loop is so low the capacitor needs to be a special type. It also has to handle very high voltages even at low power (unless you are heating up things and wasting power).

Still, even though efficiency is low, you should see distinct resonance. I'd look at a few frequencies with a receiver and see if I could get a noise peak when tuning the cap. You should hear a very profound noise peak when the loop crosses resonance.

If you don't, then I would check the capacitor out of circuit for shorts, the loop for series resistance, and the coupling loop for construction. As a matter of fact I'd get rid of that silly Faraday shielded link (it won't do a thing) and just use a piece of stiff solid wire as a link.

 
73 Tom

Logged
N3OX
Member

Posts: 8852


WWW

Ignore
« Reply #11 on: March 19, 2008, 04:25:10 AM »

"As a matter of fact I'd get rid of that silly Faraday shielded link (it won't do a thing) and just use a piece of stiff solid wire as a link.
"

That's what I did.  Also a vacuum variable capacitor and 7/8" copper tubing:

http://www.n3ox.net/projects/magloop/magloop1_lg.jpg

It handles 100W applied and works quite well for a limited space antenna but still maybe 10dB down from my other antennas... if all I had was this small transmitting loop, I'd still be having a lot of fun working DX, but it's already much weaker than modest verticals, small beams and so forth.  

If I added a lot more loss with a poor design and went to QRP power levels to keep from burning things up and I'm not sure I'd be happy ;-)

Just food for thought.  IMO, it's likely worth spending the money for a vacuum variable capacitor and a gearmotor to build a remotely tunable very high Q loop with all soldered copper tubing joints.  If you're going to stick with QRP power levels, a good air variable capacitor will work fine... but I'll add a third vote to changing the design you hope to use.

Dan
Logged

73,
Dan
http://www.n3ox.net

Monkey/silicon cyborg, beeping at rocks since 1995.
WA3SKN
Member

Posts: 5419




Ignore
« Reply #12 on: March 19, 2008, 01:27:39 PM »

Siddhartha...
Heed the other posts!
From the original post, I thought you were building a small Faraday loop for receive RDF work.  It appears you are trying to make a small transmitting loop, typical g4fon type design.
Small transmitting loops are a poor design!  They have high losses and I cannot recommend them... unless a very last resort!
Consider other compromise antennas first... hidden antennas, vertical antennas, attic antennas, etc will all work better than the small loop.  You are better off with a mobile whip mounted on a refrigerator!

-Mike.
Logged
N3OX
Member

Posts: 8852


WWW

Ignore
« Reply #13 on: March 19, 2008, 02:03:12 PM »

"Consider other compromise antennas first... hidden antennas, vertical antennas, attic antennas, etc will all work better than the small loop. You are better off with a mobile whip mounted on a refrigerator!
"

Now hang on a second.  A copper tubing, all soldered loop with an all welded/soldered butterfly capacitor or a vacuum variable with several square inches of silver plated contact area on the clamps will blow a mobile whip out of the water.

A well constructed small transmitting loop with attention taken to minimize loss resistance is a good antenna for the size, in my opinion (as someone who likes a good antenna and has done direct comparisons between a decent magloop and large verticals and modestly high beams)

I agree that the G4FON is a poor design with respect to losses.  A loop where some care is taken to stamp out losses is actually pretty good and I'd consider it to be a good choice in the space.

All tiny antennas have low radiation resistance and require you to take a bit of effort to stamp out resistive losses.  If you're trying to build an antenna that works well, for short verticals, excellent ground systems and good loading are mandatory.  For small transmitting loops, fat conductors and very, very, very low resistance connections are mandatory.  

All decently radiating tiny antennas will have small  bandwidth and will require frequent retuning.  I'd hate using a small transmitting loop without the motor drive, but it's not so hard to add one.

You can't point to any broad class of antennas and say "that's a bad type of antenna"

There's a big difference between a good, well installed screwdriver mobile antenna and a hamstick.  There's a big difference between a 7/8" copper soldered loop with a quality vacuum or air variable capacitor and a #12 wire loop with capacitors selected by a slide switch.

I'm not convinced that a hamstick on the fridge would beat a G4FON loop ... but I know my loop would beat either of them.

Dan








Logged

73,
Dan
http://www.n3ox.net

Monkey/silicon cyborg, beeping at rocks since 1995.
KI6NUJ
Member

Posts: 118


WWW

Ignore
« Reply #14 on: March 19, 2008, 02:22:41 PM »

I think I confused everyone by referring to g4fon's design. Sorry about that. I think things would have been easier to understand if I simply took a photograph of the antenna and posted it. Well, because I am not home, the following drawing is the best I can do from office. Hope that helps. I took the 16ft length of RG58 from g4fon and the capacitor values (covering 40pf).

http://picasaweb.google.com/reach.siddhartha/LoopAntenna/photo#5179562373427236162

Owen - Your advice isn't wasted Smiley I really value your and everyone's replies.

Mike - As I interpret Owen, Tom and Dan's posts, they are saying use better quality components and pay good attention to details like symmetry. Anyways, I have no attic or roof access to play with.

I am off to Vegas for a few days. I will be back next week and ready to do some changes to the loop, make measurements and share the information so I can improve it.

Q - I saw a design with two loops perpendicular to each other but I did not see any theory I could find to support that. I believe its called the box loop. Any comments on how the *box* increases the aperture of the loop? Here is the link:
http://www.standpipe.com/w2bri/article1.htm

Thanks again and please do keep reviewing and posting.

- Siddhartha
Logged
Pages: [1] 2 Next   Go Up
  Print  
 
Jump to:  

Powered by MySQL Powered by PHP Powered by SMF 1.1.11 | SMF © 2006-2009, Simple Machines LLC Valid XHTML 1.0! Valid CSS!