Basket weave-wound coils

Hey Ken,

You struck a spark of interest here with me. Reducing self-capacitance and proximity effect in a part 15 AM loading coil holds great promise because it increases the coil Q and decreases coil loss. The tireless efforts of crystal radio enthusiasts to increase coil Q do indeed transfer directly to the same requirement for part 15 AM transmitters.

Basket-weave coils reduce self-capacitance and proximity effect by staggering adjacent turns above and below each spoke in the basket. This provides some physical separation of adjacent wires, thus reducing turn to turn capacitance and proximity effect.

A similar reduction of capacitance and proximity effect can be achieved in a single-layer solenoid coil by "space winding". Typically, space winding means that turn to turn spacing is at least one full wire diameter. Ham radio coils are typically specified for this type of winding spacing. The added spacing results in a larger coil with more turns than would be required for a tightly spaced coil, but with lower capacitance and proximity effect.

Both basket-weave and space wound coils don't lend themselves to economical fabrication by today's coil manufacturers, so one option is for the user to roll his own. That's not easy either.

There is a great opportunity here to develop an optimum coil by taking from the crystal set enthusiasts and the ham radio enthusiasts. The goal would be a coil that can either be fabricated economically by a vendor, or is simple enough for a dyi user.

You will likely never see one of these super-optimum coils offered inside a commercial part 15 transmitter because it is just too big to fit inside our expectations of a small outdoor box mounted under the antenna. Most likely the super coil would need to be mounted externally between the transmitter and antenna. That's not a configuration that appeals to the typical non-technical part 15 crowd. But, boy would it sing!

For some perspective on the use of a high-Q loading coil...

A Part 15 AM system on 1650 kHz using a 3-m whip at the surface of the earth, and driven against a 25-ohm r-f ground* would gain about 1.2 dB in radiated power when using a coil with 5 ohms ESR compared to using one with 15 ohms ESR.

Even if the coil had zero ohms ESR, radiated power would improve by only 2 dB compared to using a 15 ohm coil in this system.

These are improvements, for sure, but maybe not as much as expected -- meaning that loading coils installed inside commercial Part 15 AM transmitters are not much of a compromise.

* a 25-ohm r-f ground probably is lower (better) than most Part 15 AM setups use.

I tried to make a basket weave coil.

It didn't work right but that doesn't

mean my esperimentation is over.

They are very cool.

Bruce

I have a small problem: I want to place or wind the coil on the PVC mast. Since a basket weave coil will be significantly wider diameter than a straight close-wound (with insulated wire which helps separate better than close-wound enamel coated magnet wire), how do I account for the difference in calculated number of turns?

Can I adjust by using the same wire length between start and end marks and just wind until it gets to the end mark?

If you seperate the windings from each other, this makes the coil more efficent.. (At least the width of the conductor being used)

Why not wind the coil with heavy monofilament fishing line about the same O.D. as the wire being used, at the same time, side by side. This would give you seperation between the windings and guessing the coil size (length) would be easy! (~twice as long)

Just a thought..

Would it be reasonable to speak with a basket-weaver artisan as to their possible employment as a coil maker?

Basket weaving is a proud and surviving craft and the skills possessed by weavers might be easily adaptive to coil construction.

Thanks for the reply. I planned to do just that with the 16 AWG mag wire I already have...although a basket weave is yet more efficient...more on that later.

However it doesn't solve my problem, which is more about tuning to resonance. The idea is to make the coil work with as little extra tuning as I can, so I want to calculate the correct number of turns in order to 1) not waste wire (I want to use some of it for a few of the bonded elements), and 2) eliminate residual eddies, self capacitance, wire resistance in the ground path.

Granted those are small considerations, but, again, at these minuscule power levels, every little thing counts.

Here's a link to a site page which I have used for L calculations. If you know the needed inductance you can vary the coil parameters to find the correct ones.

http://www.k7mem.com/Electronic_Notebook/inductors/coil_ind_calc.html

Making the coil longer will require more turns for the same inductance.

Neil

Question: Why does a space wound or basket weave coil require more turns?

My current coil is pretty much right on for 1650 kHz at 67 turns on a PVC sewer pipe (half the wall thickness of schedule 40 pipe) close wound except using 18 AWG mid/fine stranded wire core with standard vinyl insulation. The insulation provides spacing between the stranded wire cores of (obviously) twice the thickness of the insulation jacket. No idea what that is, but surely less than the diameter of the stranded wire core.

Adjusting radiator length to resonance came within reasonable tolerance to the OA 3 meter requirement, but Q bandwidth is pretty tight...it starts dropping off either side of a Sharpie pen line width. Advantage is - it needs no taps or air cap for tuning.

The reason the system has lost half the  signal strength is because I no longer have ocean ground...meaning I have to put out a lot of radials and deeper rods, more than I thought, because of the rocky soil. Also, there's a cement path going right across where many of the radials should go - have to run underneath 🙁

I don't expect to ever get artificial ground conductivity like the ocean, so the antenna needs to be more efficient with wider bandwidth. I need to shave off as much compromise as possible.

Thanks Neil, interesting page. However I still can't figure out why more turns would be needed for the same inductance because of spacing. I.e., more overall coil length of course, twice as much, but why would that cause  lower inductance?

I can't speak for basketweave coil equations, which don't seem to exist, but the theory is that the alternating wires of the basketweave pattern spaces adjacent wires apart by some amount depending on the diameter of the form rods even though turns are wound closely spaced along the coil length.

Spacing between adjacent turns is very important for reducing coil self-capacitance and thereby increasing coil Q and reducing coil losses.

You can stick with a single layer solenoid coil with spacing between the turns to get the same result. The total winding difficulty is probably less than for a basketweave, but it still requires some extra complications compared to a tightly spaced coil. The basketweave form, which requires a bunch of rods extending the length of the coil and spaced evenly around the perimeter of the coil is not necessary. Instead, you need a way to wind the coil with consistent space between adjacent turns. Using plastic PVC insulated wire is not a good idea because the plastic has a higher dielectric constant with higher losses, so you will get higher capacitance between turns due to the higher dielectric constant, and higher losses due to the dielectric losses. It's best to have air between the windings and use enamel magnet wire. Winding the coil with monofilament line in parallel with the wire turns is a good idea, as long as you can devise a way to finally remove the monofilament without changing the position of the wires. If they spring back to uneven spacing, you will not achieve the maximum spacing advantage. I have seen a scheme where a lathe is used to cut a shallow spiral groove on the coil form with the desired wire spacing. The wire is then wound to follow the spiral groove. Nice way if you have a lathe.

Here are two links that are relevant to coil Q and self-capacitance: http://techdoc.kvindesland.no/radio/ymse1/20061216154253848.pdf   http://www.ax84.com/static/rdh4/chapte11.pdf

Coil capacitance is very important for Part 15 transmitters because we are forced to work with very short antennas at fairly low frequencies. The short antenna capacitance is in the range of 25-35 pF. Even at the high end of the MW band, the frequency is still so low that a loading coil to resonate with such low antenna capacitance must have an unusually high inductance. Because of the large number of turns required, the inductor will have a high self-capacitance, approaching maybe half of the antenna capacitance, unless it is wound with space between the turns, and a large diameter to keep the length/diameter ratio within the recommended .4 to 1.5 range.

Finally to answer your latest question concerning why lengthening a coil with a constant number of turns reduces its inductance, you would have to go back to very basic physics from the 1800s (Michael Faraday and Henry) to learn about the magnetic field, but really all you need to do is look at the equation for inductance of a single layer solenoid coil which can be found all over the web. In computer form it looks like this:

L = d^2*N^2/(18*d+40*l)

where L is the inductance in uH, d=coil diameter in inches, N=number of turns, l=length of coil in inches. Since the coil length, l, is in the denominator, increasing the length will decrease inductance when diameter and number of turns are held constant.

What would be the reason to remove the monofilament line? It's plastic, (nonmetallic) like the form the coil is being wound on. Polyamide (nylon) Possibly less reactive than PVC..

The plan...as of now is to wind the coil on the bottom of a 4" dia. cage monopole of thin wall sewer pipe. Spacing will be accomplished by winding two wires at once by hand using dabs of hot glue as I go, then pulling one if the two out and sealing the remaining spaced winds with varnish.

The main advantages of the basket weave is crossed wires should work a bit better plus once removed carefully from the rods former and varnished, it will need no support and so becomes a true standalone air coil. In addition it can be easily built with fine grade multi wire Litz wire, which should produce yet more inductance In the upper 1/4 of the band.

Here is some more information that may help you wind the coil.

Here is a very good coil calculator suitable for the iterative process of finding the optimal coil configuration that has a wire pitch of 2 wire diameters (or more) for a given inductance and wire size and a length/diameter ratio in the sweet spot of .4 to 1.5: http://www.midnightscience.com/formulas-calculators.html#formulas5

Your cage monopole with 4 in diameter is good for optimizing Q because it has a capacitance of about 55 pF (compared to 25 pF for a CB whip). This reduces the coil inductance requirement to 237 uH at 1400 kHz. The lower inductance makes it easier to find an optimal coil configuration.

Here are some results from the calculator for a 4 in OD diameter form:

#16 wire, 77 turns, coil len. 8.11 in, L/d ratio 2.03, wire AC resistance 1.804

#18 wire, 66 turns, coil len. 5.62 in, L/d ratio 1.41, wire AC resistance 1.929

#20 wire, 59 turns, coil len. 3.98 in, L/d ratio 1.00, wire AC resistance 2.150

The coils using #18 and #20 wire both yield an L/d ratio within the sweet spot. There is a formula for AC resistance (skin effect) on the same page linked above. The AC resistance is slightly higher for #20 wire, but still low enough to be practically negligible. All three AC resistances yield a Q of around 1000, but that's not the real apparent Q which will be lower due to the coil self-capacitance. Actual loaded Q will be still lower due to the antenna ground loss resistance. Using Litz wire for this type of coil won't help much because the lower AC resistance won't make much difference in the final loaded Q.

Be sure to measure the actual OD of the PVC pipe coil form and measure the actual OD of the particular enamel magnet wire you are using. A cheap dial or digital caliper is good to use for this.

One thing that occurred to me about the basketweave configuration is that you effectively get a turn to turn pitch of 2 wire diameters (or more) when the coil is tightly wound. This will reduce the L/d ratio compared to a solenoid for the same wire size. That makes it easier to stay in the L/d sweet spot with a larger wire diameter when pitch is 1 wire diameter. Larger wire will reduce the AC resistance. Note that Litz wire with the same or lower AC resistance as #16 - #20 solid wire may throw you below the .4 lower edge of the L/d sweet spot.

BTW, for a basketweave, I think you can use the calculator to approximate the inductance and L/d by using a coil diameter equal to the average diameter halfway between the basketweave maximum and minimum diameter.

Total wire length required for a coil is equal to the circumference of the coil multiplied by the number of turns. Wire length = pi*diameter*number of turns.