I just received an off-board message with some interesting questions prompted by this thread. The responses are posted here, as they may interest others.
Q1 (paraphrased): How would the performance of a system using a transmitter with 100 mW d-c input power to the final r-f amplifier, a 2.99-meter whip antenna, a 0.01-meter ground lead connected only to the negative d-c common of the transmitter, and no other conductors compare to that permitted by FCC §15.209?
Q2: Would it make any difference in performance if the system described in Q1 was elevated?
The table and chart below were generated some time ago, and give some insight into this. They are based on the use of the ground connection losses and other conditions shown.
FCC §15.209 permits a maximum field of 24,000/(Freq in kHz) at a distance of 30 meters from the transmit antenna -- which for 1650 kHz is about 14.55 µV/m. Note that a 14.55 µV/m, 1650 kHz field is well below the ambient r-f noise level in most locations outside of a screen room.
The performance of the configuration described in Q1 would be very poor compared to a compliant §15.219 system using a short conductor to a ground rod, as the antenna system would have such high reactance that nearly all the power available from the transmitter would be dissipated as heat, even if the antenna system could be tuned to resonance. Probably it would perform better than a system compliant with FCC §15.209, though.
The height of a Part 15 AM transmitter and its attached whip -- by itself -- has no significant effect on field intensity. What
does have a large effect on the field intensity of an elevated system is the radiation produced by the conductor(s) other than the 3-m whip, leading away from the transmitter. They, along with the 3-m whip form an off-center fed dipole, which has much better radiation efficiency than the 3-m whip with a very short ground lead, alone. This is easy to see in the NEC plot (2nd graphic below).