Bill Firebaugh: The Well-Tempered Innovator Page 2

I don't know if I could assign a magnitude to the improvement, but you can sure see it when you're making flutter tests. And you can hear it. A friend of mine who has a home-built, very massive turntable like a Micro Seiki, uses a very stiff belt with it, made out of rubber-impregnated Kevlar. He tried flipping it over and he told me he very definitely notices a difference. He always uses it twisted now.

It was while I was experimenting with belts that I discovered something amazing. The belt goes slower than the pulley. And it's not because the belt is slipping, it's more like it's crawling. Here's what happens, I believe. On one side of the pulley the belt is tight, because it's having to overcome platter pivot friction, while on the other side it's loose, because it's just taking up the slack. And in between, around the pulley, those tension differences have to be adjusting to one another. So there's a sort of a wave of suddenly-changing tension crawling backward all the time through the pulley-contact area.

Holt: Wouldn't there have to be some slippage through this area?

Firebaugh: Sure, but see, the slippage is in one direction at one end of the belt wraparound, and it's in the opposite direction at the other. You'd expect them to cancel, except that the area where the belt is crawling is continually feeding part of the pulley-contact surface backward toward the tensioned part of the belt.

It's hard to visualize. And it took me a long time to figure it out. But what it does is, if you try and calculate the platter speed from the pulley speed and the two circumferences, it doesn't work out right. The platter doesn't run at the predicted speed; it will always be slow. I'll tell you, this drove me nuts until I found out what was going on. I wasn't the first person to discover this by any means. I was explaining my speed problems to my belt manufacturer, and he knew all about it. He confirmed that this is a known phenomenon in belt-drive systems, especially ones with relatively compliant belts. The pulley goes faster than the belt. But I had to learn it the hard way. I ended up having to correct the size of the pulley empirically to get the right speed.

Holt: Why does your turntable weigh 50 pounds?

Firebaugh: It weighs about 42 pounds. The base material is Medite, which is a very dense, high-quality fibrous board. It's resin based, like chipboard, but denser. And perhaps you know that I was using sand for a while to damp it further. In the first prototypes.

Holt: So I understood. But you aren't any more?

Firebaugh: Oh no, it was a menace! Sand is marvelously inert; it's really great stuff for damping turntable bases. And tonearm tubes—I pack the tube of the WTA with it—but it was no good in the turntable. It made the base very inert, but when I brought my prototype WTT to Las Vegas in January 1985, it was so dry there that my demo records built up a fierce electrostatic charge, and some of those grains of sand got stuck to them and ruined them. They all ended up with scratches. So that was the end of the sand.

I wound up with a Medite sandwich—three plates of Medite fastened together with sound-absorbent stuff. It looks like double-backed adhesive tape, but it's very sound-deadening. It's made by 3M. They've got a huge catalog of double-backed stuff with all sorts of different properties, made out of lots of different materials with different densities and thicknesses and compliances and so forth. You wouldn't believe the variety of stuff they make!

Holt: What about acoustical isolation?

Firebaugh: I don't attempt to isolate the platter from the base, but the whole turntable is suspended on four fairly stiff feet, whose compliance is such that the mass of the table provides isolation from the underlying surface from about 15Hz upwards.

Holt: Are the feet spring-loaded?

Firebaugh: No. I tried using springs, but I never could get the result that I wanted. It was always too jiggly and unstable, and intuitively, I know that if you touch something, and it jiggles up and down for several seconds, that's going to give you instability problems when you're playing a record.

Holt: Did you try damped springs?

Firebaugh: Yes, but the damping provisions were either ineffectual or prohibitively costly. My whole experience with springs soured me completely on them. By solving problem A they just introduced problems B, C, D, E, and F. So I gave up on springs.

Holt: Your 'table's suspension feels very stiff. Is that the way it's supposed to be?

Firebaugh: Yes. And that's the kind of compliance you need to isolate a mass of 42 lbs at frequencies above 15Hz. While 15Hz may seem kind of high, the type of stability it gives you makes the player very immune to such things as footballs and so forth. In my home, I have the record player sitting on a dining-room buffet in which there are many drawers and doors which I use to store audio junk in—old cartridges, arms, record-cleaning brushes, alignment protractors, all sorts of the things you accumulate through the years. I can open and close those drawers and doors without in any way disturbing the playing of a record.

Holt: You said you started out with "the usual" aluminum platter. Is that what you ended up with?

Firebaugh: Oh, no. Like everything else I started out with, it went through a lot of experimentation. The aluminum platter rang like a bell. It was very poorly damped. No, the final version uses an acrylic platter—Plexiglas. It's highly internally damped. It weighs about 3 lbs, and there's a thin rubber mat on it.

Holt: That's pretty light. Maybe that's why belt vibrations were affecting the sound.

Firebaugh: Maybe, but twisting the belt helped that very heavy platter my friend had on his home-brew 'table.

Holt: What about the motor suspension? That looked pretty unusual too.

Firebaugh: I went around the world on motors, and tried lots of different schemes for mounting the motor on the same assembly as the platter and then isolating it with springs. You know, that's the usual technique. And I just couldn't get it to work. I couldn't get the motor vibrations down to a level that I wanted, so I got kind of torqued off one night because I was at the end of my rope, and I took some lead—scuba divers' weights—and...

Holt: Which, I suppose, you just happened to have lying around!

Firebaugh: No, I do scuba diving. But I did happen to have a Coleman stove and a big pot, and I melted some lead and poured up a brick of it. I clamped the motor to that lead brick and said, "Okay, motor, if you want to vibrate, fine, you do that, but you're gonna have to vibrate this lead brick along with you." And it just completely eliminated all the motor vibrations.

Holt: Just by adding mass!

Firebaugh: Just by adding a lot of mass.

Holt: How big is this "brick"?

Firebaugh: It's four and a half inches square. I've forgotten how thick it is; an inch and a quarter, maybe.

Holt: How is it attached to the motor?

Firebaugh: It's clamped against the bottom of it. There's a metal plate at the top, with the spindle sticking through it, and then there are four long screws that pass through the plate and through the lead brick at the bottom. The motor's captured between those two.

Holt: How is the motor fastened to the turntable unit?

Firebaugh: It isn't. It's a completely separate assembly. But it doesn't stick out like an outrigger; it sets into a compartment at the left-hand side of the base. The motor stands up on four stiff rubber feet attached to the underside of the brick. You can't have the feet too compliant, because the whole motor will tend to rotate.

Holt: Was that your finished turntable?

Firebaugh: Not quite. The object of all this was to try to get the flutter and wow down to low levels, and I eventually wound up fluid-damping the motor. See, no motor manufacturer I contacted would guarantee me any rotational stability spec. You buy a 300-rpm motor, they guarantee it goes around 300 times per minute and that's all. But its speed during each revolution? That's completely uncontrolled. The manufacturers assume inertia will take care of that, but it doesn't completely.