Graham Model 1.5 tonearm Page 3
Gilson (Wireless World, October 1981) points out that the side thrust force acting on the stylus increases with increasing overhang. Side thrust increases the stylus loading on the inner groove wall, decreases it for the outer groove wall, and tends to displace the cantilever from its center position in the groove. All of this can and does lead to increased distortion. He takes "hi-fi pundits" to task for giving manufacturers such a hard time for "less than optimum" arm geometry. Gilson is quite willing to settle on a value for overhang less than the Löfgren/Baerwald standard, and trade increased lateral tracking distortion for reduced side thrust.
He makes a valid point. If side thrust could easily be compensated for, he would not have a leg to stand on. But the side force varies so much across the surface of the record that it is impossible to effectively compensate for it. The best approach, according to Gilson, is to minimize it in the first place.
I imagine that Gilson would not be enamored of Löfgren's alternative alignment, which calls for increased overhang. I too am unimpressed with this alignment. The fact that the Model 1.5 arm allows a cartridge to be aligned in either the standard Baerwald/Löfgren geometry (which Graham simply calls Baerwald) or the alternative Löfgren geometry (Löfgren, according to Graham) gave me the opportunity to compare the two. I started out with the Benz MC-3 set up in the alternative Löfgren alignment, then switched to the standard alignment, taking the time to readjust the VTF and VTA because of the change in overhang. I did not, however, change the anti-skating during the testing, which was set at 1.5gm throughout. At least with this cartridge, I was much happier with the standard B/L alignment. In general, the soundstage was more transparent and more palpably focused. The upper octaves were smoother and inner detail was better resolved.
One of the issues that gets little attention in any discussion of lateral tracking alignment is the accuracy with which the audiophile can make the overhang and offset-angle adjustments in the home. The wide availability of commercial alignment gauges obscures the fact that they are difficult to apply accurately. The gauges depend on crude "eyeballing" to achieve their claimed alignments. Even if you got the overhang just right, with what sort of precision do you think you can judge the offset angle? Even a 1 degree error here will greatly change the maximum distortion figure.
Another problem is that gauges assume that the cantilever is perfectly parallel to the sides of cartridge body, and then use the body of the cartridge for alignment purposes. This is sometimes not the case. In my own sample of the Benz, the cantilever is somewhat skewed to the side as it emerges from the body. Unless the alignment procedure allows the use of the stylus tip and cantilever as reference, the final accuracy of the alignment may be a far cry from what the gauge intimates it should be.
The Graham Solution
Let's look at how the Model 1.5 addresses the various criteria for the "perfect arm," starting with the pivot design.
It may surprise some of you to find that Graham has chosen to go with a "unipivot" bearing. Nevertheless, a unipivot has a lot going for it: First, it is the simplest design. Second, it pre-loads the bearing surface to zero tolerance, and is capable of yielding the lowest possible friction in a mechanical design.
Bob sent me a videotape of a test he conducted where he pitted an SME IV against an undamped Model 1.5. Both arms carried the same cartridges and were statically balanced. Both arms were displaced vertically downward the same distance at the start of the test. The idea was to see which arm bobbed up and down the longest, as this would indirectly reveal the degree of vertical friction in the bearings. The SME pooped out after about 30 seconds. The 1.5 kept going for two and one half minutes before Bob stopped the arm for fear of having me fall asleep. He claims that the arm actually continues its pendulum-like motion for over five minutes.
Third, bearing performance does not drift out of tolerance, as there is nothing to adjust. Graham uses tungsten carbide for both the bearing cup and pivot. Both elements are said to be polished to stylus-tip tolerances. With a typical 7-9gm cartridge mounted on the arm, Graham calculates the loading on the bearing point to be in excess of 100 tons per square inch. With this sort of loading there is no play in the bearing to interfere with the transduction process. Finally, a unipivot design makes it easy to add damping fluid around the pivot point. Viscous silicone fluid (about 0.75ml) is used here to provide damping in both vertical and horizontal planes.