Analog Corner

(Originally published in Stereophile, May 12th, 1998)

I've got this friend Shirley. Married with two kids, she appears to be your typical suburban middle-aged housewife—but somehow her music genes got short-circuited. While most of her neighbors have become Yanni-fied (if they pay attention to music at all), Shirley is a Rolling Stones fanatic.

You are too? Really? Have you been on the Stones' "Bridges to Babylon" tour, attending concerts on the schedule throughout North America, and some in Europe too? Shirley has. (Her husband, Jerry, stays home—he's more of a Kinks aficionado.) Do you spend every spare minute on some kind of internet Stones "Glimmer Line''? Shirley does. Do you own every CD the Stones have ever released, plus stacks of live bootlegs? Do you play them constantly, memorizing who plays what on every track? Do you have a set list from every show on the current tour? Do you have the Stones' lips'n'tongue logo lovingly painted on your fingernails—a different visual and color combo on each one? Shirley does.

Recently she forwarded me an e-mail she'd received from someone who wanted to know if the new British 180gm pressing of Sticky Fingers was sourced from a digital master—as if that was a good thing. (It is from a digital source, by the way.) I e-mailed her that, rather than explaining why the opposite was true, she should come over and I would play an analog version for her vs Bob Ludwig's recent Virgin Records CD remaster.

Last week she showed up with her friend Pat, another Stones fanatic. I sat them down for a serious listening session. Pat was skeptical as I took out an original British Decca pressing of Beggar's Banquet. "How many plays before the needle ruins the grooves?" he asked. "Well, Pat, I'm going to play you a record I've owned and played incessantly for 31 years."

When "Jigsaw Puzzle" ended, Pat and Shirley were wobbly with pleasure. "It's a different song," Pat exclaimed. "I've never heard half of that stuff, and that sounds like a real electric guitar." And, and, and!—when the evening was over, and after some direct comparisons between ancient, oft-played LPs and the highly praised CD remasters, both Shirley and Pat agreed: there was no comparison between LPs and CDs. I don't have to tell you which won out, or why, or by how much.

What really amazed me was how these nonaudiophiles expressed what they heard and how it differed from what they were used to hearing. Without using any of the worn-out audiophile vernacular, Pat and Shirley were each able to describe the sensation of hearing instrumental lines laid out with ease—allowing them, for example, to follow Bill Wyman's bass throughout a tune. Both were struck by how much more of the intent behind what the musicians were playing came through. And, of course, both heard instruments (especially the difficult-to-delineate piano) buried for decades of listening on nonaudiophile gear. Neither could believe, for example, that Brian Jones actually plays harmonica throughout "High and Dry," not just at the beginning and during a few scattered breaks.

I played them German, Japanese, and American pressings of Exile On Main Street, and they were able to hear and describe the differences though neither had ever spent a minute of their lives listening this way. Shirley said, "I can't believe I can hear differences between pressings from different countries." Both described the sound from the CDs as being "all mooshed together," and both noted that just when the music reached what should have been a dynamic peak, it "pooped out." Both were really surprised to hear that analog bass was more harmonically complete and much better focused than digital—and believe me, it wasn't the CD player (a Bow-tech ZZ-8) that was at fault.

What's sad is that both Pat and Shirley have original vinyl pressings. They just haven't bothered playing them for the past decade because they thought—well, you know what they thought. They don't think it no mo'! Will they invest in audiophile-grade analog playback gear? Stay tuned.

Wally to the rescue
Coincidentally, the morning of the night Shirley and Pat visited, a box showed up full of Wally Malewicz's latest setup accessories.$s1 Using them, and some tips he'd shown me at CES, I made subtle but significant changes to my analog front-end that really improved the sound. So I thought this would be a good time to do a cartridge setup, Wally style.

You start with Wally's Wally-VTA-tractor ($80), which looks like a black skate blade about 2½" long attached to a tapped cartridge-mounting plate. But this device has nothing to do with "skating." Instead, it allows you to adjust VTA starting with the arm dead parallel to the record surface—no guessing required, even with a tapered armtube. Having that verifiable reference point to return to makes playing with VTA almost fun—and certainly anxiety-free.

To use it, you install your cartridge and set VTF (vertical tracking force). Don't worry about alignment for now. Wally supplies a set of height-measuring gauges. You find the one that reaches from the record surface to the top of the cartridge body—say, 19.5mm. Now you remove the cartridge and install the 16.5mm-tall VTA-tractor in its place, first adding three 1mm shims to the top of the mounting plate. (The device comes with a set of 1mm, 0.5mm and 0.25mm shims.) If your cartridge measures less than 16.5mm deep, Wally can supply you with a shorter VTA-tractor.

With the bladelike Wally-VTA-tractor in place of your cartridge, get out the thinnest and thickest records in your collection (eg, an RCA Dynaflex and a Mobile Fidelity Anadisc 200 200gm job, respectively). Start with the thickest on the platter and adjust VTA until the "blade" is perfectly flat on the surface of the record. If you're fortunate enough to have an arm with a height gauge—like a Graham, ET, VPI, or Triplanar—note the setting. Now repeat with the thinnest record.

Reinstall your cartridge and, using the higher of the two positions, set your overhang with, I suggest, the Wally-tractor ($80) custom-made for your brand of arm. The precise arc your stylus should follow across the record is laser-etched on a piece of mirrored acrylic drilled to fit over your spindle. Begin by approximating the overhang and then setting VTF in the middle of the cartridge's suggested range. Malewicz's Wally-scale tracking-force gauge was not available when this column was being written, but for about $80 he's claiming 0.05gm accuracy using a unipivot fulcrum and laboratory-grade counterweights—more on that next time.

When you've adjusted the overhang so the stylus stays centered in the etch all across the arc (don't settle for anything less!), double-check your VTF. Remember that increasing VTF pushes the stylus slightly forward; decreasing VTF pulls it slightly back. (It also changes SRA—stylus rake angle—which some consider more critical than VTA. More about that in another column.) Hopefully you won't have to readjust your overhang when you've got VTF correct, but remember: unfortunately, all of these adjustments interact with one another to varying degrees.

Ideally, the stylus should traverse a straight line across the record surface, but unless you have a true linear-tracking arm, that is impossible. As the stylus arcs across the record surface, it will cross the ideal straight tracking line at two null points. Wally puts three etched parallel lines at each null point. With your stylus tip on the arc trace at the center line, the cantilever should be precisely parallel to the two other lines (make sure antiskating is disconnected). If it's not, adjust the cartridge body. If your cartridge has been properly manufactured and your arm's offset properly designed, this shouldn't be necessary.

Now you have your cartridge body precisely parallel to the record and your stylus tip traversing the ideal arc across the record—all verifiable, thanks to Wally's devices. You're off to a great start.

Before you set azimuth, you have to set the antiskating. Skating is a real phenomenon that must be dealt with. Without getting into the whys, with an offset (headshell angled toward spindle) pivoting arm, forces are created that make the arm want to "skate" toward the center of the record. Antiskating is a force applied in the opposite direction to counterbalance that force. Unless you apply such a force, precisely calculated to offset the skating force, the arm will mistrack at high modulation levels and, equally bad if not worse, the stylus will wear unevenly—on the inside. If you apply too much antiskating, stylus wear will also be uneven, but on the outside.

According to Wally, research done years ago by Thorens demonstrated that the antiskating force should be approximately 10% of the tracking force across most of the record, increasing to 13% toward the center. He's designed a Gyro Gearloose–like device that lets you set antiskating and measure your arm's bearing friction. The Wally Skater ($85) consists of an acrylic plate that fits over the spindle and sits on the platter, attached to a vertically mounted, foot-long acrylic rod. Protruding from the rod at right angles are two more acrylic rods—one close to the platter, one at the top. Attached to the one at the top is a sliding device to which are attached two threads. At the end of one is a plumb bob, at the end of the other is a loop you place around your tonearm's finger lift.

You have to be very careful when first using this contraption or you're likely to swing your stylus into oblivion. I recommend securing the stylus guard before you proceed. First you adjust the plumb-bob thread so the bob hangs just above the lower rod, which contains a sideforce scale marked off in 1% increments. Attach the other thread to the finger lift and adjust the thread so the stylus floats between 1/8" and 1/4" above the platter. Play around with it a bit before removing the stylus guard. But never use this device with a felt or sticky mat; inevitably, the thread will slip and the stylus will contact the surface below. If it's felt or rubber, bye-bye stylus.

Once you've got the hang of this device (pun intended), and with the protruding rods at a right angle to the arm and the antiskating disconnected, the plumb-bob and finger-lift strings should be opposite each other at the same marking on the scale—wherever you have the plumb bob. If you move the top slider from which both the plumb-bob and finger-lift strings are suspended, the arm should instantaneously follow the plumb bob. This will be the case with top-quality arms. If there's a lag, you can measure the distance the plumb bob moves before the arm does. If your tracking force is 2gm (2000 milligrams) and it takes three marks on the scale before the arm moves, that's 3% (each mark is 1%), or 60mg of resistance that you must take into account when setting antiskating. In other words, subtract the 3% from the 10% of antiskating you wish to apply, since the arm's horizontal friction supplies 3% of it.

Okay, your eyes are beginning to glaze over; but once you play with the gizmo for a few minutes, it's not difficult. To set antiskating on arms with essentially zero friction, simply apply a sufficient amount for the arm to move back toward the outside of the platter 10 markings from where the plumb bob points. Using the Wally-skater, you can measure the antiskating device's effectiveness over the entire surface of the record and potentially compensate for any variations caused by the particular antiskating device used by your arm's designer. It's really neat!

Next we have to adjust azimuth, which, when set correctly, puts each channel's coil parallel to the groove. The groove walls are at 45º angles to each other; for accurate reproduction, the coils must be too.

Using the electrical "null" method recommended in the past by your correspondent and many others is simply WRONG, WRONG, WRONG—I apologize for misleading you. That method will ensure that the electrical output of the two generator systems is equal, but that's not really the goal, which is to minimize crosstalk and maximize separation between the channels.

To do that, you need at least one, preferably two digital voltmeters and a well-made test LP containing tones first in one channel only, then in the other. Such a disc is not currently being made, so this requires a trip to the used-record bins. If you find a test record with a "Channel Identification" band giving you a minute of 1kHz modulation in the left channel followed by a minute of 1kHz in the right channel, you're in business. I use the Command Stereo Check Out (CSC 100), but Shure's Trackability test records nos. 4 and 5 are also good; they're out there if you look. Not all test records are accurately cut, so be careful.

If you can afford a digital voltmeter that converts voltage to decibels (dB), your job will be easier, but you can get the job done even with an inexpensive RadioShack model. Begin by putting the meter's probes into the left channel's amplifier output jacks (positive to positive, negative to negative, of course). Play your test record's left-channel modulation track at a comfortable volume level and an output of between 5 and 10 volts. Don't worry if the meter fluctuates a bit due to pressing imperfections—obtain an average. When the right-channel modulation plays, note the voltage. It will be much lower—hopefully about 0.2V or 200mV. That's the crosstalk into the left channel: the amount of leakage from what should be playing only in the right channel.

If you have two voltmeters hooked up, one to each amplifier channel, you can get both measurements at the same time. Otherwise you'll have to move the meter probes to the right channel and play the record again. This time you'll get the smaller voltage when the left-channel tone is playing. When you're finished, you'll have "big" and "little" voltage measurements for both channels.

The goal is to compare not voltages, but decibels. So convert the difference between the left channel's "big voltage" from the "little voltage" (ie, what comes through the left channel when the right-channel modulation is playing) into dB, and do the same with the right channel. Remember: each channel's generator system is slightly different, as are the channels of your preamp; so don't compare voltages, compare the difference in dB. Wally Malewicz will be happy to send "Analog Corner" readers a free voltage-to-decibel conversion chart.

You can also buy a logarithmic calculator (not expensive these days) and use this formula: crosstalk in dB = 20 x log (V1/V2) (Footnote 1). Since I'm mathlexic, I'll give you the example Wally gave me: If you measure 7.65V in the left channel when the left channel is being modulated (V1), and 0.262V in the left channel when the right channel is being modulated (V2), and then divide V1 by V2, you'll get 29.198. The log of 29.198 is 1.465; multiply this by 20 to get 29.3dB crosstalk in the left channel. Repeat with the right and see how close you are in decibels.

Remember that you're measuring the difference between the voltages, so it doesn't matter if you start with 3V or 5V or 10V when you set your volume on the modulated channel. Minimizing crosstalk maximizes separation, which opens up the soundstage and helps to lock in the images and solidify the sonic picture.

If you were very lucky when you set up your arm, the difference in dB will be 10% or less (30dB vs 33dB, for example). You're done! But don't count on it. Most likely one channel will be 35dB and the other 20dB, which is too great a difference. Which way do you shift azimuth? You'll have to experiment to find out with your setup. The best thing to do is write down the results you first obtained, shift your azimuth (however it's accomplished with your arm) slightly one way (note which way!), and measure again. If the difference in decibels increases, go the other way until you're within 10%. Yes, it's time-consuming. Yes, it's a pain in the butt. Yes, you will be rewarded for your troubles.

So now you've got your tracking force to within 0.05gm of the middle of the cartridge manufacturer's recommended VTF, your VTA set precisely for parallel to the record, your overhang verifiably spot-on across the entire record surface, your antiskating at or near the ideal 1/10th VTF, and your azimuth set electronically for minimal equal crosstalk between channels. You're dialed in, baby!

Well, not quite: Most of us find that ideal VTA is slightly below parallel to the record, and frequently the maximum recommended tracking force yields the best tracking and smoothest sound. But at least now you can play with VTA, and if you get lost you can easily get back to parallel with both thick and thin discs. Now all you have to do is play with cartridge loading....

Four tips:

1) I find the Little Lite gooseneck miniature lamp indispensable for cartridge setup, especially getting the overhang correct with the mirrored acrylic gauge—the light angle spells the difference between clarity and confusion.

2) See "Analog Corner" in the July 1997 issue of Stereophile (Vol.20 No.7) for more setup tips—especially practical ones. But keep in mind that where there are technical differences, the present installment supersedes those earlier recommendations—especially when it comes to azimuth adjustment.

3) If you're afraid to do your own setup on an expensive cartridge, get yourself a cheap Grado and learn the ropes. When you get that Grado locked in, you'll be amazed at how good it sounds.

4) I heartily recommend Wally's setup tools—especially the Wally-tractor. True, you may use them only occasionally, but they're well worth the money. If you own a retail store and set up cartridges for customers, or if you belong to an audiophile club, I definitely recommend all of Wally's tools.

CD/LP notes
Long-time audiophile favorite Friday Night in San Francisco, with Al DiMeola, John McLaughlin, and Paco DeLucia, was first issued on LP in 1982 (Columbia FC 37152) and has recently been reissued on both CD and 180gm LP. The LP is from Vivante Productions in London on the Philips label, and while it's probably not kosher to be sold Stateside, it will likely be available through the usual mail-order suspects.

The CD is from Sony/Legacy, and, even though it doesn't say so on the jewelbox, it's an HDCD. I compared the original Bob Ludwig–mastered LP with both new issues. First off, the HDCD is really spectacular—yes, you read it in "Analog Corner." It sounds timbrally close to the original LP and offers incredible detail and the kind of decay you associate more with not brushing than with CD sound (that's a compliment). If you found the 1981 LP exciting but too bright and strident, try the new LP—it's richer and warmer, giving you more instrumental "body" but clearly softer (some would say "sludgier") transients. Which is "correct''? Talk among yourselves.

I've been raving for years about a record called Mel Tormé and Friends Recorded Live at Marty's, New York City (Finesse W2X 37484). I asked DCC Compact Classics' Steve Hoffman about a reissue, but he told me the tape was "unavailable." Last week I got a CD version in the mail from DCC, not on gold CD but as part of their aluminum-disc DCC Jazz series. I put it on and it sounded pretty good...until I heard pops and ticks. This CD has been transferred from LPs! I called Hoffman to find out what kind of turntable he'd used. Turns out the only tape he could find was a dub from LP made by producer Norman Schwartz. Check it out anyway.

Next time: Mikey visits the Library of Congress' recorded archives.

---

Footnote 1: Many years ago, I wrote a small computer program to do this calculation. E-mail me at john.atkinson@sorc.com and I'll send you the program free of charge.—JA