BH7HQT These are not ideal conditions, as they are environment-dependent. For a puller type GP, only three things can be adjusted (and these are the only three), namely: the vibrator/ground stake and the ground net.
If the ground itself is already very poor, such as having stakes driven into concrete, it will not necessarily exhibit a 50 ohm resistance. If the stakes are driven into moist soil, they may sometimes exhibit a 50 ohm resistance. While longer and more extensive grounding networks are undoubtedly better, how much of an improvement do they offer?
I searched online for some tests related to this.https://w6iwi.org/az/Vert30/radials/ Regarding the section on "Relative Field Strength vs. Number of Radials," RBN's report shows that the number of ground nets varies from 0 to 16 across the 20m/15m band, with SNR fluctuating only around 1 dB. However, neither 80m nor 40m are lengths suitable for operation with a pull-out GP.
In summary, I'm saying that if you can directly adjust the oscillator by inserting the cable GP into the ground and achieving proper standing waves, then you won't need a ground net. If you can't achieve proper standing waves, you will need to add another ground net. As for whether you still need to add additional ground nets even after having achieved proper standing waves... it depends on whether you want to pursue that 1-2dB gain.
Of course, this problem also occurs in UV (Ultra-Violet) applications. When the hand-held antenna is connected to a coaxial cable with a center-fed design, the standing wave ratio (SWR) is already within acceptable limits. At that point, do we really need to add "a real ground" or "good grounding"? I generally don't (because buying those three small ground anchors takes up space and also costs an additional ¥