I made a thread about nut compensation some time ago. Since then I've been making nut compensations on many old guitars. But sometimes I've had problems with the saddle and the thick E, A and D strings. I could not reach the intonation point even with a thicker saddle bone than I should need. Thinking about it I realized that using a tailpiece when measuring may not be correct with a stringpin bridge.
I used a whole day trying three different methods to measure the intonation points.
1 The new gizmo for the saddle (and still the loose frets at the nut) with only two strings at tension.
2 The first setup with the new gizmo but with all the strings at tension when measuring.
3 The way I've done it before with a tailpiece and movable tangless zero "frets" for the nut and drills on top of the bridge.
The difference between method 1/2 and 3 is a bit chocking but interesting, it seems that the angle for the saddle on a glued down bridge with string pins needs to be bigger than on a floating bridge with a tailpiece. The difference on the saddle for the thick E string was a whopping 2.5 mm, and no difference for the thin e string. The nut compensation using the old method was successful because I always use the stroboscope tuner when filing to the intonation point at the saddle. But sometimes I needed to redo the saddle with a wider slot when I didn't reach the intonation point.
To my relief there is no difference (apart from measurement errors) between method 1 and 2. I had some problems reading the stroboscope with the A string, may explain the difference at the nut for method 1 and 2. With only one string at the time to measure it's easy to work with the drill bits (the second string at tension on the other side of the fretboard is needed to keep the zero fret nut/spacer in place). The method 1 is probably the better measured one, having three strings on top of the drill bit was cumbersome. The only real difference is that the drill bit had to be 0.3 mm thinner on this guitar with all strings at tension to get the correct string height at the 12th fret.
Here are my measurement results for the three cases using the same guitar and strings.
1 Measuring single strings with "gizmo", distance from the backside of the 1th fret
to nut intonation point (mm) and saddle intonation point (cm)
E 35.00 58.85
A 35.18 58.70
D 36.77 58.65
G 35.48 58.60
b 36.16 58.50
e 35.79 58.40
2 Measuring with all strings at tension using "gizmo"
E 34.96 58.85
A 34.70 58.65
D 36.56 58.65
G 35.41 58.60
b 36.20 58.55
e 35.86 58.40
3 The old tailpiece method
E 34.29 58.60
A 34.83 58.50
D 36.50 58.50
G 35.31 58.50
b 35.89 58.50
e 35.77 58.40
As before I use my set of drill bits with 0.1 mm steps between 1 and 6 mm to set the 12th fret string height before measuring the open string, 3d and 12th fretted notes with the stoboscope tuner making sure all the three notes are correct at the same time. To make this happen, the movable bits of zero frets at the nut and the "paper clip"/drill bit at the saddle are adjusted to and forth until a solid match is found.
I used 2.6 mm 12th fret string height for the E and 1.6 mm for the e strings (Newtone Heritage 0,12, 618 mm mensur) and the string height of the strings in between following a straight line.
2.6 / 2.4 / 2.2 / 2 / 1.8 / 1.6
That gave the following measured string heights above the 1th fret using a 0.05 mm feeler gauge under the tangless fret bits used to intonate:
0.45 / 0.4 / 0.4 / 0.35 / 0.3 / 0.25
The "gizmo" works very well. With the middle screw through the threaded brass pin it is easy to move the intonation point, the "paper clip" wire. The loose drill bit is securely wedged between the "paper clip" and string when adjusting the screw. The "paper clip" can be moved some distance if needed with the set screws. The two nuts are tightened against each other and will turn with the middle screw.
The aluminium for the bar will be replaced with a square piece of 6x6 mm brass when I get the material. Aluminium is too soft for the set screws in the long run.