Very short loaded verticals over a good ground plane usually have a very low feed point impedance and a particular value of inductance across the feedpoint can effectually raise the impedance to match the coax better. The problem is worse on the lower bands where the antenna is really short and usually 20m and above will not have as much loading effect from coils in the traps and will not require external matching.
The shunt coil is tuned for the best match on the lowest band or split between the two lowest bands and its effectively out of the circuit on the higher bands where its not needed.
In my experience with a Butternut HF9V, the stock coil is about right for the antenna being on a pipe in the ground with no radials or on a roof with just a few radials. If the ground plane is better (like a big copper roof) the impedance on the lower bands goes lower and spreading the coil apart will change the match in the right direction.
The ground plane on the huge copper roof I dealt with caused me to remove a few turns and I still had to spread the coil apart some to get the match down on the lowest band. BTW, you want as much ground plane under a vertical as possible and your 36 radials is a good start. People who simply stick that antenna on a pipe in the ground with no radials might have a good match but the performance will be way down compared to your 36 or more radials.
prcguy
Thanks for the response. I've been over my assembly a dozen times, and I can't see anything I missed. The shunt I is something I've wondered about. I can't find too much info on adjusting it, I hate to shorten it and then find out I didn't need to.