Interesting Coil Stuff

Thanks for the link. The table at the bottom of the article shows the typical Q for a close wound enameled wire coil to be 200. This is close to the Q measured for my coil which was 180.

The table shows the Q for Large #2 mix iron on 1.8MHz (toroid?) to be 500 to 600. At 1670 kHz this would give a coil resistance at RF of 6.3 Ohms compared to 18 Ohms for my coil which has a Q of 180.

Repeating a calculation from <a href="" http://www.part15.us/comment/28293#comment-28293" ;" another thread, the radiated power for a coil RF resistance of 6.3 Ohms would be Pr = (.085/(6.3 + 24.5))*(.12) = 0.33 mW compared to 0.24 mW for the 18 Ohm coil. This would be an increase in field strength of 1.17 which would predict a 17% increase in range.

Using my observed listenable range of 1.3 miles as a basis, the new range would be 1.5 miles. Not a great improvement but probably worth doing assuming that the Q of 500 is accurate.

Neil

These posts have convinced me of one thing, reducing the capacitance between windings by spacing is worth the trouble.

The size wire I used to wind my first coil was based on wire I already owned, which I think is probably a common approach. It is Belden 22 AWG Uni-Strand lead wire with either gray or green insulation. I can tell you the color of the insulation has no affect.

My coil-form is triangular rather than round, which I am told makes no difference. They say the coil could even be square.

My coil loads my AMT3000 exactly according to the manual, so if I improved the spacing between windings it would also load up.... what could be more efficient than loading properly, one way or the other? I just got confused.

Carl queried "...what could be more efficient than loading properly, one way or the other?"

There are two factors in play with the loading coil. One is that the coil presents the proper inductance to the antenna system to negate the capacitive reactance of the antenna radiator. Any reasonable Q inductor of the correct inductance value will do this regardless of the Q involved. This would be seen as "loading properly".

In addition to the inductance, the coil adds a resistance to RF which can be modeled as being in series with the inductance. This resistance (R) introduces power loss in the antenna system. Though the transmitter most likely will load there will be power loss in this resistance which is power taken away from the radiated signal.

This resistance is expressed in the Q with a high Q representing a low resistance and low power loss (R = (2*pi*frequency*inductance)/Q). Both R and Q depend on frequency so it is most meaningful when speaking of either to state the frequency of operation.

The R and thus the Q of a coil is affected by the current distribution in the cross section of the conductor. The skin effect describes one attribute where the current is forced to flow primarily at the surface of the wire so in cross section it would appear as a ring with the outside of the ring being the surface of the conductor. As the frequency is raised this ring thickness narrows which means less cross sectional area is conducting current resulting in higher resistance. This is the same as using smaller wire for DC in that smaller wire has less cross sectional area raising the resistance.

Other physical factors affect the shape and size of the conducting cross section area such as wire spacing. The high Q designs for coils are attempts to maximize the cross sectional conducting area as much as practical to give the highest Q and lowest R and lowest loss.

As shown in my calculations, which are for a real system based on measurements, there is a point of diminishing return for the effort to create a high Q coil which for my system seems to be at a Q of about 180. There is more to be gained with high Q if the ground loss resistance is smaller but for my system the Q of 180 is good. An air wound coil will give good results but a hunk of wire wound on a pencil (very low Q) will not.

In summary, higher Q is better but there is a point of diminishing return.

Neil