Stephano directed me to this thread. I'm glad he did :)
I have a DL103D on a modified RB300 arm (expressimo back end, no VTF spring, Cardas wire, no plug in arm pillar base, VPI VTA adjuster, planed underside of headshell and end of arm tube).
For years, my general approach to upgrading and tweaking has been to identify each individual weakness and fix it. As a result, my tonearm has been setup to reduce the presence region resonance/shrillness in this arm/cart combination. I've applied several suggestions from Thorsten (decoupling cart from headshell, lead tape around armtube at 1/3 length) and set VTF and VTA to reduce the shrillness.
After reading a bit of this entire thread, I got a 3/4 ounce "lead substitute" weight, cut it in half, drilled the ends, loaded them with modeling clay (temporary glue) and stuck 'em on the axle nuts. There was an astonishing change as all previous customers have reported.
- large increase in bass resolution and extension
- soundstage is more solid, bigger, paradoxically with less well defined instruments
- shrillness mostly gone (yea!)
Then I noticed that the top octave or two were gone.
So I decided to set up the parameters that are tuned by ear -- VTA, VTF, and anti-skating -- again. Azimuth, overhang and offset are set correctly. Instead of setting these by ear to reduce the shrillness while maintaining a decent tonal balance, I followed a method that's new to me. Basically, you set VTA to get best focus on a mono vocal record, then set VTF to get best dynamics and realistic tone, then set anti-skating to get equal dynamics in both channels. See this post by Bernhard Kistner at
http://www.audioasylum.com/audio/vinyl/messages/45126.html.
First, I got some 1 oz. lead weights and redid the outriggers. Then set the arm up twice. After the first go round, there was an upper bass/lower midrange boost. This was due to the lead band around the arm. So I removed it and started over.
As things have turned out, there is now almost no distortion and the tonal balance is pretty much as it should be with only minor HF rolloff. Many many records that were really "zippy" and sibilant are very clean and natural now. Most bass is much tighter. Unfortunately, some old bass-shy Mercs are even more so, though the definition is stupendous.
I'd like to throw my hat into the amateur physicist ring and open up an issue.
Several people have noticed the change in soundstage similar to what I described. I believe that non-identical moments in the lateral and vertical planes are the culprit, causing crosstalk between the channels that has an unknown phase relationship to the signal in the originating channel.
If the cartridge body was perfectly stationary, azimuth perfectly set, and the groove was carrying a signal in once channel only, then the stylus would be moving in a plane at 45 degrees to the record, perfectly aligned with the generator for that channel. Now, if the cartridge is allowed to move in the vertical plane, then the stylus motion, relative to the cart, would be in a plane that is no longer aligned to the generator. This motion, as a vector, has a small error component in the plane of the other channel's generator. Depending on the actual motion of the cartridge in the vertical plane, this error component may be in phase with the dominant signal, out of phase, partly delayed, and this relationship may vary with frequency, arm resonant behavior, the behavior of the as an oscillating system over time, etc.
Fixing the arm in the lateral plane by raising the lateral moment, while the vertical moment stays relatively low to allow the arm to track warps, is equivalent to the situation I just described.
Thank you TWL for the mod.
- Eric, three years too late to this party