- Posts
Hello all,
What follows is a rather technical discussion of Part 15 AM antennnas and transmitters. It is my hope that this will stimulate some thought and discussion.
The conventional wisdom regarding part 15 AM antennas is to place an inductance (loading coil) in series with the antenna so one can tune the antenna for maximum effect. The only reason to place a series coil in a 3 meter antenna system is to make the feedpoint impedance of the coil/antenna look resistive to the transmittter. The assumption here is that the transmitter requires a resistive load. The coil at resonance with the antenna serves only to present a resistive load to the transmitter and will not in any way change the radiation characteristics of the antenna.
Hello all,
What follows is a rather technical discussion of Part 15 AM antennnas and transmitters. It is my hope that this will stimulate some thought and discussion.
The conventional wisdom regarding part 15 AM antennas is to place an inductance (loading coil) in series with the antenna so one can tune the antenna for maximum effect. The only reason to place a series coil in a 3 meter antenna system is to make the feedpoint impedance of the coil/antenna look resistive to the transmittter. The assumption here is that the transmitter requires a resistive load. The coil at resonance with the antenna serves only to present a resistive load to the transmitter and will not in any way change the radiation characteristics of the antenna.
In other words, the reason for loading coils and resonant circuits seems to be to improve the antenna radiation, but this can be a false assumption. At best, these efforts only serve to match the antenna feedpoint impedance to that which is “expected” by the transmitter and do nothing to change the actual radiating antenna electrical characteristics.
Now what is “expected from the transmitter”? I have done experiments where I determined that the Ramsey AM-25 gives maximum output power when the resistive load is 24 ohms. I have also done experiments with the SSTRAM AMT-3000 where the output power increases as the resistive load is increased beyond 1200 ohms resistive with no peak found (more work is needed with this).
Now, allow me to emphasize, the only purpose of a loading coil or matching network is to present the transmitter with an acceptable load impedance. What this means is that we want to present a load to the transmitter which will result in maximum radiated power. This does not mean that that this will occur when the complex conjugate of the antenna matches the complex transmiter output impedance when viewed as we consider efficiency. The maximum power transfer theorem if applied gives an efficiency of 50%. If the power companies used this, half the power they generate would be lost.
The guiding principle for maximum radiated power with a given transmitter input power is that the output impedance of the transmitter should be zero and for the power availble, the voltage should be maximized.
I performed some simulations of short AM antennas and it is striking that the impedance is very low resistive but very high reactive. Since the field strength and thus the radiated field is dependent upon the power dissipated in the model radiation resistance, I conclude that the best results will be obtained when the voltage at the antenna feedpoint is as high as possible within the power limits of the system.
The only power which does anything useful is power delivered to a resistance. I know that transmitters are not perfectly efficient, but assuming we have one, we need to deliver our 100 mW. into a resistance comprised of the coil, ground, and radiation resistance. If the total of all resistances in a part 15 AM antenna system is typically 30 ohms, but the reactance is -j2500 ohms what should we do?
100 mW real power into this load will mean a voltage applied = 144 volts. (Preal = Re(V^2)= Re(144^2)/(.30 – j2500)) = 99.52 mW.
So, the result is this. An antenna loading coil functions to present the transmitter with a resistive load upon the assumption that a given transmitter requires this. The loading coil does nothing to improve the antenna performance but serves only to cancel any antenna reactance in an attempt to match the transmitter requirements. The ideal part 15 AM transmitter should be able to produce 144 volts at the antenna terminal regardless of the load impedance and therefore given the reality of part 15 AM antennas should be designed to do so with a highly capacitive load. If this happens, the real power delivered to the antenna system will be P = Re (V^2/Z) = Re (144^2/(30-j2500)) = 100 mW real power.
It appears to me that if, by means of step up transformers and clever transmitter design, we can deliver 144 volts to an antenna, we do not need a loading coil. The big question is can the transmitter produce this voltage when presented with a highly reactive load.
Neil
Total posts : 45366Tesla lit light bulbs miles away using that idea until the generator gave out.
Total posts : 45366Yep. But he didn’t have to worry about Pt. 15 rules did he?
Neil
Total posts : 45366[quote=radio8z]… It appears to me that if, by means of step up transformers and clever transmitter design, we can deliver 144 volts to an antenna, we do not need a loading coil. The big question is can the transmitter produce this voltage when presented with a highly reactive load.[/quote]The real power in the antenna system is a function of the voltage and current magnitudes there and to their relative phase, and is maximum when current and voltage are in phase at the antenna feedpoint, and zero when they are 180 degrees out of phase. These two conditions correspond to a VSWR magnitude of 1.0 and infinity, respectively.
A highly reactive load means that current and voltage are not in phase, so the power delivered to the load is reduced. This is analogous to circuits on the a-c power grid, where reactances are used to maintain voltage and current in phase, so as to maximize deliverable power.
In the case of the 30 -j2500 ohm example above in this thread, a tx able to deliver 100% of its rated power into a 30 ohm non-reactive load would “see” a load VSWR of 6,946:1, which would accept/dissipate only 0.03% of the tx power. The remaining 99.97% is reflected back toward the tx, and most of that is dissipated by system losses.
It doesn’t matter from a power transfer and radiation viewpoint how much r-f voltage exists at an antenna feedpoint if no, or very little current is able to flow there. Without current, the antenna will not radiate the EM energy needed for its purpose.
//
Total posts : 45366Hello all,
I do not disagree with Rich’s analysis, but I would like to add some thoughts. What I did was a thought experiment and a little math to determine what would be required to deliver 100 mW. to the 30 ohm resistive antenna component assuming an ideal (100% eff.) transmitter and also assuming that the only power delivered to the final amplifier would be the real power delivered to the load (to stay legal).
Granted there are practical limitations which will severely reduce the efficiency of the system, mainly the inefficiency of common amplifier classes which may kill this approach. The volt amps in the load in this example is .099 + j8.29 or .099 watts + 8.29 VAR which would indicate that the I^2*R losses will be high, yet the current magnitude is only 58 mA. In my fantasy ideal situation, the transmitter will act as a constant voltage source (implies 0 output impedance) and will be able to provide the voltage and current without regard to the phase which is set by the load. This happens frequently in AC power circuits where there is a poor power factor (e.g. small fans, flourescent lights).
A key here seems to be a low or near zero output impedance which can be accomplished with feedback circuits (operational amplifiers for example have closed loop output Z’s in the tenths of ohms). Another view is that the reflected power is not dissipated in the transmitter since it sees 0 ohms.
I do realize that I am approaching this with assumptions about ideal circuitry but by doing so maybe I can get closer to ideal in practice.
Neil
***edit PS…a thought about what Rich wrote:
>It doesn’t matter from a power transfer and radiation viewpoint how much r-f voltage exists at an antenna feedpoint if no, or very little current is able to flow there. Without current, the antenna will not radiate the EM energy needed for its purpose.
You are correct but 144 volts applied to the model antenna impedance produces the maximum current constrained by the 100 mW limit for the final amp. The power dissipated in the radiatiion resistance will still be very small but that is the situation we have now and eliminating the coil losses will increase this power.
Neil
Total posts : 45366The calculations below put some perspective on this, assuming such was practical.
Parameter > Base-loaded > Proposed
Antenna Rr > 0.1 > 0.1 ohms
Antenna j > -2500 > -2500 ohms
Coil j > 2500 > 0 ohms
Coil R > 2 > 0 ohms
RF Ground Resistance > 29.9 > 29.9 ohms
VSWR > 1:1 > 6,946:1 (Ref = sum of resistances)
Reflection Coefficient > 0 > 99.97%ASSUMING THAT THE PART 15 TX COULD DELIVER ITS RATED POWER INTO THE VSWRs NOTED ABOVE
Antenna System Radiation Efficiency > 0.313 > 0.333 %
Antenna System Radiation Efficiency Improvement > 0 > 6.67 %
Field Strength Improvement > 0 > 3.28 %So even if this was possible/practical, wouldn’t the “coverage” benefit be rather minimal?
//
Total posts : 45366Rich,
I agree that a gain in radiated power of a few percent is not worth the effort and elimination of the loading coil will not produce much since its resistance is swamped by ground resistance in the model.
I am just not convinced (based on some preliminary antenna current measurements) that either of my two transmitters is providing anywhere near a reasonably expected current for the given final input power. I need to refine my measurement technique before I can cite data, right now it is just an impression. I will advise when I have some data which I can trust.
Neil
Total posts : 45366Rich,
I have been considering your posts on this thread and I have a question. You cite a reflection coefficient greater than 99%. Why would that be a problem if the delivered power (resistive) was at the legal limit?
If I could deliver 10 watts with a reflection coefficient of 99% would this not be equivalent to delivering .1 watt into a load with a reflection coefficient of 0?
If a transmitter is terminated in a resistive load due to a loading coil in series with a highly reactive antenna, does not the high reflection coefficient still exist at the junction of the loading coil and the antenna? Where does the reflected power go in this situation?
Thanks for considering these questions,
Neil
Total posts : 45366[quote=radio8z]You cite a reflection coefficient greater than 99%. Why would that be a problem if the delivered power (resistive) was at the legal limit? If I could deliver 10 watts with a reflection coefficient of 99% would this not be equivalent to delivering .1 watt into a load with a reflection coefficient of 0?[/quote]The first reason as far as Part 15 AM is concerned is that the FCC limits the input power of the final r-f stage of the tx to 100 mW. Another reason no less important is that few r-f amplifiers can withstand working into a load that reflects 99% of the power that the amplifier is trying to deliver. That reflected power adds dissipation (heat) and r-f current/voltage in the tx circuits that usually are in excess of what they can withstand.
[quote=radio8z]If a transmitter is terminated in a resistive load due to a loading coil in series with a highly reactive antenna, does not the high reflection coefficient still exist at the junction of the loading coil and the antenna?[/quote]A perfectly resonant antenna has only pure resistance, and no reactance. It makes no difference to the circuit whether or not that resonance is the result of the mechanical dimensions of the antenna alone, or is a combination of those mechanical dimensions and an input network (coil). The tx will see a load with zero reactance in both cases, and the reflection coefficient into a matching impedance at the antenna or antenna+coil feedpoint will be zero.
//
- You must be logged in to reply to this topic.
