Fall 2008

Dayton WT3 Woofer Tester
A fantastic device that delivers everything it promises!
Article By Jeff Poth

Difficulty Level

  In my first formal Enjoy the Music.com review (I previously covered VSAC 2008), I'd like to begin with a disclaimer that I am technically a cable manufacturer (Poth Audio). This article in no way conflicts with my other company, as i do not manufacturer loudspeakers yet am a happy DIY enthusiast. Now let us move to a basic introduction paragraph. I love building hifi. I love hearing music.

Wow that was quick! More on the building hiifi part, and more specifically, the important part of my experience for this review: I've built dozens of speakers at this point, everything from tiny to huge, fullrangers to four-ways to subs. Generally I design by theory (and math) and knock up prototypes. Then I listen. Crossovers are generally tuned by ear, and then measured at a later time, once I get a ‘feel' for how the drivers work together. I often utilize a Marchand XM126 tube-based active crossover for testing various configurations. While it's a great piece of fun audio gear, it is even moreso a tool. Nothing is simple about high fidelity speakers, and to build ones own designs, things like active crossovers, huge boxes of expensive copper coils and capacitors, lots of sawdust and spent hours are necessary.

And in that vein, may I humbly submit the Woofer Tester 3. This slick little $100 toy works with your Windows machine for which the included software was written. It plugs into a USB port, and connects to the driver in question with alligator clip leads. Driver means: woofer, tweeter, midrange, ribbon, whatever sort of driver you like (some exotica may not be measurable, such as plasmas and piezo drivers).

Woofer test rig, and driver under test (AlNiCo magnet vintage JBL driver 122A)

After warm-up (you thought it was just tubes that warmed up didn't you!) the device settles in. You need to make sure that your windows settings are correct for it, which is comprised of maxing out your volume control, and a couple other small things. A quick calibration routine zeroes out the test lead resistances, and uses a resistor (included) to lock in the impedance scale. All of this takes but a few minutes, and is not a pain in the least, indeed, it's necessary to get the quality reads you get.

And you do indeed get quality reads, over 30,000 points on the impedance plot for starters. This is accomplished via a quick frequency sweep that's audible through the transducer (driver). You're also able to test the driver parameters. Reference efficiency and Vas (tied to efficiency) require a test box or ‘added mass' approach to measure, but you can still get valuable information on Fs, Qts, Qes, Qms without the added mass or test box, which allows you to check on plenty of issues. If the manufacturer's sensitivity spec is reliable, you have everything you need to do a precise box and crossover alignment. Drivers are best tested in a quiet environment on a 'test baffle', which is a rigid frame which allows the speaker to do its thing without excessive vibration.

Test results of a full-range driver. Note the ripple in the impedance at approximately 1.5-1.8kHz, showing the driver's first major breakup.

But wait there's more! It's not just drivers you can test, you can also test the impedance and phase profiles of complete speakers, crossovers, inductors and capacitors. Now we're getting somewhere, this functionality allows one to greatly improve tuning speed compared to some DIY development routines. Phrases like "Wow, I really did need a Zoebel network!" and "Geez I better downsize that cap!" will pour out of the experienced builder's mouth.

Yes I said experienced. This does not mean that this device is only useful for a seasoned builder, but the beginner will need some time working with this (and crossovers in general) and figuring out how impedance and phase information translate into what needs to be built in the crossover, and how it will affect the speaker's performance. Crossovers are the hardest part of a speaker. Unless you're building a full-range, which often means backhorn, which is a complex matter of an acoustic crossover. And even here, the information presented in an impedance plot is greatly useful; it shows horn resonances and operating bandwidth.

There are many uses for such a device. Quality control functionality is great. If you were building a high-end kit, you could make sure that all the parts were on-spec. You can use it to optimize port tuning in an enclosure, or to get the right amount of stuffing in a transmission line. It can show you problems with a crossover design, resonances within drivers themselves, and how the voicecoil inductance in a driver manifests itself. Let me repeat that- how the voicecoil inductance manifests itself. Indeed, voicecoil inductance is not a simple 1^st order linear transfer function; it's dictated by many factors and is highly non-ideal. By measuring it properly, rather than relying upon a manufacturer's spec which is typically Le at only one frequency, you're able to better control the driver within a crossover design. You can direct precision strikes at impedance peaks and breakup modes, and make sure that your drivers do what you're telling them to do.

Measurement of a completed speaker system with aperiodic venting (note the broken peak at 100Hz). Note also the impedance peak at 1.5kHz, showing the action of the crossover.

This device proved itself useful in sorting out which drivers have accurate specifications, and help in the design phase of the crossover.

For AlNiCo magnets, you can test whether the driver is on-spec -- match the parameters to the manufacturer's specification -- and you can see if the magnet has lost some strength. This is a real effect with AlNiCo, in which excessive power input can result in a 3dB loss in sensitivity, and dramatically changes the behavior of the driver. Qts rises due to higher Qes. In other words, since there's less magnetic flux in the gap due to the demagnetized material, there's less damping at resonance, so such changes totally invalidate box designs for these drivers. I managed to come up with a quad of the excellent JBL 123A that were still on-spec. This woofer was used in the L-100, one of the largest selling speakers of the time, and is essentially flat to several kilohertz, which is very rare for a driver of its size and power. Then I have some 122A, which had a much more powerful motor.... but now, it's significantly less so, due to overload. These woofers would be a poor replacement within the original system, but are still perfectly serviceable as high performance woofers; one simply needs to know how to use the altered woofer.

Conclusion
If I sound enthusiastic, that's because... I am! This is a fantastic device, delivering everything it promises, in a compact and easy to use package. The crowd who will get the most out of this are the DIY speaker builders who have some good electrical theory behind them, or those who are willing to invest some time in fine-tuning their projects and learn as they go. This is less useful for people who just want to assemble something, and aren't inclined to learn about the electric and acoustic theory. If you're like me, you're technically proficient, but not overly reliant upon measurement (since we hear very differently than microphones, I think a lot of measurements are overstated in terms of usefulness), and a device like this does an excellent job of bridging the gap- lots of information in a fast, easy to use fashion. A very practical solution for speaker tuning and tweaking indeed!

Details
Type: Computer aided loudspeaker measurement device.
www.daytonaudio.com