Attached is an example of a configuration that would disturb the balance of the currents in the OWL:
The right side of the antenna and the left side of the antenna return currents to the unbalanced side of the tuner which are:
Not necessarily equal in amplitude (because C1 and C2 do not have the same values)
Are phase shifted with respect to the OWL input currents (because the round trip path length from the tuner to the end of the OWL, along each end of the antenna, and back through ground is a significant fraction of a wavelength... or more)
But he mentioned in the loop case that the lower frequencies showed out of balance lamps, not higher freqs.
I think I'd look at the lamp 6 inch spacing pickoff, maybe change that to a couple of feet for 40. Maybe change the pickoff point that's originally about one foot out from the MB. Might be that there's no "imbalance" at all on 40, but that since the antenna is only a 15 meter loop that one of the lamps is at a current null from a wave going all the way around, so to speak.
If some results are seen there then make a larger loop for 40. The small inductor has sufficient phase change that the lamps now 'seem' to indicate balance when in fact it was fine in the first place.
What is the ERP variance at the far field with and without the small inductor? I'll bet none.
Yeah, model it.