Thanks for all the info.
I'm using the Shen with a Gauge Lifeline, Headway EDB-2, into an Acoustic Image Coda.
Using a bit of HPF on the amp, EQ is set mostly flat.
Open A string is quite boomy, C# on the same string is kinda dead.
I agree, not going to haul a 31 band rack graphic EQ around. Might try experimenting more with the notch filter on the Headway, see if that can sure the boom. What frequency should I be cutting? a narrow band ~55Hz right?
The way I find the right frequency with an EQ is to find the peak, then apply a dip. With parametric I start with a medium bandwidth. Turn up the overall gain until the bass is on the verge of feeding back. Then create about a 6dB medium bandwidth peak and sweep until I find the offending frequency. Find the exact frequency by narrowing the bandwidth and sweeping back and fourth. You want to find the peak with pretty much the narrowest bandwidth possible. After you identify the precise frequency, pull the gain down with the narrow filter.
The feedback frequency can sometimes vary with factors such as temperature. So in some instances you may want to slightly open up the bandwidth, but this is only with super narrow filters that most devices do not have. Otherwise you may need to recenter some of the notches as conditions change.
Since you generally can't peak with a notch filter, you simply turn up the gain until the bass is on the verge of feeding back and then sweep the notch to find the offending frequency. After you get close, try increasing the gain so more. If the same frequency rings again, try fine tuning the notch.
In the event you ever do work with a 30-or 31-band graphic, here is my procedure: Turn up the overall gain until the bass is on the verge of feeding back. Note any frequencies that pop out. Now drop the overall gain slightly, grab the two lowest bands on the EQ, and slowly bring them up. If nothing happens keep turning them up until they have maxed out. If feedback occurs, take note of the tone and drop the faders back down. Next move up one band and do the same thing. In other words you will peak two bands again, the lowest band was previously the top band, and the top band is new.
After you have tested three bands, assess and make a decision. Do you need to dip one of the first two bands or not. If you do decide to make a dip, you have to determine which of two bands to adjust, and then apply a relatively deep cut. The idea is you should dip the band that is closet to the problematic frequency. Sometimes the problem is almost dead center between the bands. If this occurs it's often better to make a shallower cut, because the effective bandwidth is wider. However, sometimes you get feedback that is close to each band. If one band feeds back way easier, try a deep cut on only the worst band. Sometimes it's better to us a relatively shallow cut across several adjacent bands.
The judgment you use to decide which band(s) to dip is something that develops with experience. So consider repeating if you don't get great results on the first attempt.
So far you have boosted three adjacent bands and made a decision whether you should apply a cut to the lowest two bands. Make your adjustment and then continue in the same fashion. Every time you move up one band, you make a decision on what to do with the lowest two bands out of three. Continue until you have tested and assessed all relevant bands on the EQ.
After you get comfortable, streamline the process however you see fit. Ideally you want to dip only the worst 4-6 feedback nodes. If you have "golden ears", you can simply progressively bring up the gain and when feedback starts you simply apply a dip at the corresponding frequency.
I like to keep on hand one on the overall gain while I am peaking the EQ controls, so I can quickly turn everything down if a really strong feedback spike occurs.
