Just like yesterday
And I'll get on my knees and pray
We don't get fooled again
If things happen for a reason, it's a good thing I didn't do that lacquer touch up on the front just yet. After only a day indoors, her neck joint started failing and the action almost doubled. The intonation past the 4th fret was totally unacceptable, so I had no choice but to remove the neck again and give the reset another try. After using the old reliable clothes iron with a pair of metal putty knives to release the glue between the fingerboard and body, I made a simpler, kinder and gentler neck removal jig from a scrap piece of stair tread and a few pieces I salvaged from the first jig.
With a little steam from my Wagner steam machine and a few turns of the wing nuts, the neck slid out nice and easy. After a little cleanup of the wet hide glue with a rag, I left things to cool off overnight.
After gluing some thin shims to the neck heel side of the joint and some more sanding, I was ready for another attempt. My hunch was that the top plate was deflecting at the bridge from the string tension. This time, I decided to hold off on using the glue, so I could take some measurements at the bridge after a day or so, and put the strings back on. I measured the 12th fret action before and after playing her for an hour or so, and found the action was holding just under 1/8 of an inch. I also noticed a slight gap at her neck heel and body. I let her sit until the next day and, just like last time, found that her action had grown to over 1/8 of an inch and her intonation was unacceptable again. I also noticed there was a bit of relief in the neck. After flattening it out with a few tweaks of the truss rod, the action came back down to just under 1/8 of an inch and the intonation returned to an acceptable level.
Noting that the nut and saddle could stand to be trimmed some, and that I was still planning on restringing her with 10-gauge strings, I felt a little better about the way things were going. I took a close look at the heel and noticed the gap that shouldn't be there was also just a bit wider. It was still under 1/16 of an inch, but way too large. Before taking the neck off to work on the heel gap, I set up a makeshift device with a level for a flat edge and stacks of fender washers for measuring the top plate deflection at the bridge.
With the strings in tension, it took 10 fender washers at each end of the straight edge for the level to clear the high point of the bridge, leaving about a 1/16th of an inch gap. With the strings unwound, producing no tension, the gap between the bridge and the straight edge was virtually unchanged.
Since the top plate was not deflecting from the string tension, the only thing remaining that could have led to the change inaction was sloppiness in the dovetail joint. Knowing that, I glued some more thin shims at the tip of the neck heel end of the joint and left them to dry overnight.