This is going to be a long reply so I’ll quote directly inline using CP for my previous comments and BR for Brent Regan’s comments then I’ll follow with CPN for my latest comments.
CP<<<Burying the radials in the carbon fiber or such that the radials were more to the inside would go a long way toward removing the beneficial effects of the radials.>>>

BR-<Really? That is a very interesting point but you completely lost me on your aluminum airplane analogy. Expanding your logic, in the case of an aluminum airplane, an antenna will have poor performance if mounted inside the airplane but we know (believe) it will behave much better if mounted outside. Shouldn't the aluminum damp the signal as you say the carbon will? We know this is not the case so it seems you were right about the "dumb thought". It would seem that, as you implied, that the AC impedance of the carbon structure is the critical, and unknown, element here. I wonder what is the impedance of the carbon skin at COM frequencies. DC, low current resistance is on the order of a few milliohms per square inch per inch. I doubt that we are so "lucky" as to have the skin impedance at the frequencies of interest to be anywhere near the critical damping impedance. >

CPN -- There’s a lot to cover here. First the aluminum can analogy. Suppose I put a radiating source inside a can with no holes (i.e., it’s battery operated.) The loss is very low, but real. The standing waves build up until the circulating currents times the real component of the loss (the IR loss) equal the output of the emitter and the container heats up. Now we puncture the wall. Depending on the size and relationship of the hole to the circulating current, the hole becomes a re-radiating antenna. For instance, a slot ½ wavelength long and perpendicular to the current is just as good as if it were a ½ wavelength wire antenna fed by an equal amount of current. This is one of the primary mechanisms in EMI problems with electronic equipment. The electronics inside generates currents on the box that typically has a cover. Said cover forms slots with its mating box (look at your typical PC for instance), and said slots re-radiate the problem EMI. At this point the primary loss mechanism is the loss through the slot and the wall losses fade into insignificance. So your comment about “critical damping resistance” is right in that it is unlikely to be right at 50 ohms, but that still doesn’t mean it can’t be adsorbing some significant portion of the energy.


BR <While this is "outside my area of expertise", I had understood that damping became less effective as you moved from the high (EMF) potential end of the antenna (tip) to the low potential end (ground plane). So, how do you bleed off energy that is at ground potential? The current is highest at the base but the resistance to the feedline is lowest>

CPN -- Here is the crux of my concern. You’re right about the EMF on the tips of the antenna. But an additional way to think about it is that you have a continuous transformer going from approx. 50 ohms at the feedpoint to hundreds of ohms at the tip. But to me, the important point to remember is that you can’t radiate from a monopole (a nice theoretical concept) you require a counterpoise in some form. (We’ll come back to this in your question below.) Generally this counterpoise is the groundplane formed by the skin in an airplane. But it’s important to realize an equal current also flows in this groundplane (i.e., the current you measure in the center of the coax – the feed – goes out the ¼ wave whip, through the air, back to the ground plane and returns on the shield If the plane is resistive, then electrical current is used to heat the resistance resulting in heat. One way to view this is that at these frequencies, the capacitive coupling between the tips of the antenna and the ground plane form a closed loop and the RF current circulates. The residual inductance of the wire is balanced by this capacitance, and the leftover is a real resistance which a different way of stating that the antenna is launching energy into space as electromagnetic radiation. But another thing about this field around the wires/ground plane is that it wants to minimize the impedance, and part of that minimization process is that the field wants to minimize the enclosed loop area if possible. So this means that it will try and occupy the surfaces closest to each other commensurate with conductivity which means some of the current will go onto the surface of the carbon fiber if it is in the way. Or another way to look at it is that imagine the ¼ whip standing on a groundplane. There are circulating current and voltage fields. If you place a wire in the field, it will intercept these fields and as a consequence will have a circulating current. This will reach a maximum if the wire is ½ wavelength long. This element will now re-radiate the energy. This is the principle behind the Yagi antenna. But what if this element is resistive? It will adsorb energy from the field and dissipate it as heat. This is the condition of a resistive ground plane in the energy field, i.e., the ground plane radials are inside the airplane. How bad this is depends on what the ground plane resistance is in this frequency range.


BR<And just when you thought it couldn't get better, the airplane skin is actually two skins separated by the core. What is the effect of the capacitive and inductive coupling between the skins?>

CPN -- Generally my rule of thumb is that if the spacing is significantly less than a wavelength, the capacitance will dominate and it can be treated as one plane (where I’m assuming the planar dimensions are significant portions of a wavelength.


BR<Even so, I always thought of a ground plane as an electrical "mirror" that makes a 1/4 wave antenna "look" like a half wave. It is not part of the system that receives the signal other than "holding" one end of the antenna so it can resonate. As such, it shouldn't matter if it is wrapped in air, carbon or wet noodles. For transmitting, the ground plane soaks up the antenna's "reflection" like a sponge. The earth makes a good ground plane, so why wouldn't copper/carbon? >

CPN -- So, first thinking of the ground as a reflector is good, generally thinking of it as a “sponge” is bad. Good ground planes are good conductors at the frequency of interest and have equal currents flowing in them. The earth is not as good, but you take advantage of the area principal. The area times the skin depth is much higher for the typical installation, such as a typical 550 KHz to 1.5 MHz AM transmitter. And in point of fact, many of these installations include buried radials to improve the efficiency of the installation. (Also to complicate matters, not all installations are ¼ wave resonant systems especially the AM just mentioned. A ¼ wave 550 KHz would be about 446 feet tall – that’s a tall antenna.) But as frequencies go up to FM and TV range, the antennas are either dipole or ¼ wave on ground planes, but in either case, the counterpoise conductivity is satisfied with actual conductors of metal.

BR<One could reason that either the carbon is a good conductor of RF or it isn't. In any case, adding copper radials wont hurt and may help.>

CPN -- I absolutely agree. The copper radials can only help. I’m just suggesting for maximum benefit they should be on the outside with the ¼ wave whip, especially on lossy medium such as carbon fiber construction.

BR<Sounds like it is time for more tests! >

Yep, you can’t beat an antenna range!

BR<In the meantime, theories aside, we do have empirical evidence from the lab and the field that ground radials on the inside of the carbon skin produce satisfactory results. Not being an expert, I tend to rely on experience.>

CPN -- I agree, I’m only quibbling about the word “satisfactory” and how that should be defined.

BR<PS: And what about polarization? Two thirds of the bent whip antenna on my airplane is horizontal and therefore 90 degrees out of polarization phase with COM ground transmissions. Wouldn't it work better if I took the kink out? ?

CPN -- Man, you don’t want much do you?<grin> This is a big area. I’ll just say I don’t know, the bend will introduce distortions to the antenna radiation pattern and is likely to reduce its effective height (i.e. it’s not quite as efficient as a ¼ wave whip. ¼ wave whips can be made much shorter with a technique called a “top hat” where a plate or disk is added to the top of the antenna to increase its capacity. The antenna length is shortened considerably, but the end result is that although it VSWR may be wonderful, it radiation efficiency is terrible because of the reduced physical dimensions have reduced its effective height. This is another aspect of the point I was originally trying to make, to wit, just because you have a good match – low VSWR – doesn’t mean you have an effective antenna.) To imagine the field from a ¼ wave vertical whip on a ground plane, drop a doughnut over the whip. The doughnut represents the field in space. The length of the vector from the junction of the whip and the grounplane to the outer surface of the doughnut represents the gain of the gain of the antenna in that direction. So one direction the antenna is “blind” is along the axis of the whip. When you bend the whip down, it would be a bit like squishing the doughnut under the narrowed area and expanding it directly opposite. In other words, you started making a directional antenna out of it and the lobe away from the bend starts lifting off the ground plane. It’s not necessarily less efficient (but keeping in mind the “top hat” example above it probably is a bit less efficient than a vertical ¼ wave whip), it’s just not even in all directions. Which, while I’m on the subject, suggests that ¼ wave whip on the belly is better than a ¼ wave whip on the top of the airplane. Look at the doughnut, the angle of radiation is about 45 degrees from the ground plane horizontal – now imagine which direction you want to transmit to/receive from when you’re at altitude. You’re above the radio stations (I hope, because any other concept scares me) so you need your antennas to angle the radiation down. The post from Bob Jude today is correct, the cone will start to lower the radiation angle. But the ultimate end point of that process is a vertical dipole whose pattern is perpendicular to the line of the dipole and it becomes a variation that is called a “bazooka.” So I hope that Bob’s picture is of the roof of his plane, not the belly, as I would think that would provide a better radiation angle in flight.

BR<Oh, the things we do for vanity and speed.>

CPN -- But occasionally I think we make allowances for safety and comfort else why carry around all that surplus weight in radios and navigation gear, and extra seats! <grin>

I hope this made sense – if not, ask questions and I’ll try again.
Regards,
Charles Patton
LNC2 360JM