Last Word with Tony Andrews, Part 1

with
Tony Andrews

Tony Andrews is a director of Funktion-One and co-founder of Turbosound with John Newsham. Pictured: a hirsuite Tony (third from left) in 1983 with John Newsham (left), Toby Hunt and Cliff Henricksen visiting from the US.

What got me started in audio? ’60s soul music.

By 1970 I was doing sound for bands and had built my first loudspeaker.

It was all bit of a chance occurrence really. We were listening to music and generally spacing out, when my brother picked up a smallish PA box and faced it into the corner of the room. Of course all the HF content was eliminated but the bass had gone up. Within a matter of days we’d built a 2 × 18-inch box facing into a corner with a lid on it.

I’m not saying that sub was revolutionary, but it set the direction we’re still on: working empirically. We check things as we go. Sure we use sophisticated measurement equipment – in fact, we built a huge device with Southampton University to do 3D balloon measurements – it can take 300 or 400kg and it runs automatically in 5° steps. But there’s no substitute for building and listening.

Most of our point-of-difference technology is in the driver and the way we load it.

The basic impedance mismatch of a moving diaphragm in the air is so great you’re really doing well at that level of componentry to get better than three or four percent efficiency, and typically the average hi-fi in your front room is only two percent efficient. But with proper horn loading you can get that up to something approaching 30 percent. That’s night and day, in terms of the amount of power you have at your disposal. 

But it’s not just the efficiency, it’s the nature of the sound you get out of horns. They’ve always been more immediate, involving and provide altogether better transients. We want that sort of performance down to 25Hz. That’s where we’re going at the moment.

We’re well into designing a 32-inch sub. To be honest, I’ve not properly tuned it yet because I’ve not experienced those frequencies in such quantity before. Then you’ve got the added problem: are there any records out there that have that sort of frequency content? Not many. We’ve got four tracks that give us something to work with it.

Our research and experimentation in the ’70s culminated in our first mass market product, the Turbosound TMS3.

We liked the sound of cone drivers — the sound of paper is perfect for rock ’n’ roll… well, just about everything really. At the time Electrovoice had a great eight-inch device that was high flux with a nice light cone. It was really good for midrange. We put a simple wooden horn on it and got another 8dB out of it over and above using it as a direct radiator. Problem is, you get that honkiness that puts everyone off horn loading anything other than a compression driver. 

Someone working with us at the time took a rolling pin out of the kitchen and put it in front of the cone while it was working and it immediately eliminated a whole shed load of cancellation and it began to sound half decent. From there we began experimenting with different geometry — something you still see in the patent [head to Google Patent Search for more].

We finished up with a horn that didn’t honk but had all this immediacy and efficiency. I think right off the bat we were at 107dB at 1W/1m. Also because the output was so well organised we were preserving a lot of HF, which allowed us to push the crossover point a lot higher.

That’s a big part of what made TMS3 special: pushing the compression driver crossover point into the 4kHz area. It almost made the midrange like a tweeter (with Flashlight we took that a stage further).

Compression drivers are inherently a pretty distorted device and they can’t do a vocal properly. To this day people rely on them down to 2kHz. It’s a mess, with a ton of intermodulation because the diaphragms are too small. On top of the diaphragm there’s some kind a metal involved and then a phase plug that’s gone largely unchanged since the 1930s. From there, people think they can fix the problems with electronics, which is complete and utter bullshit. You have to get the speaker right; you can’t increase speed or recover lost transients with electronics. And if you EQ the hell out of it — which I actually consider to be downright cheating — your phase is all over the place. You want things to be smooth, and if you have holes and spikes caused by some kind of argument in the diaphragm, you can’t fix that with EQ — you just make the argument worse. You can average it over time and kid yourself you have something approaching an even response, but the ear doesn’t work like that. Transients are everything for the ear. And it’s that immediate delivery of all frequencies, all at the same time … well, that’s proper audio. 

Some of the major systems out there have ridiculous amounts of EQ. Maybe six or seven points and ±15dB in some cases.

The TMS3 forced us to make a leap of faith. We had a rental company that was doing really well. We were out with Santana and even Frank Sinatra at one point. But it was extremely hard work, you didn’t always get paid, and the margins were bad — you’d be lucky if you’re breaking even. We looked at the TMS3 and thought, if we don’t start building and selling this now, someone else will. We literally gave the rental business to Britannia Row. That was a sad day.

We were renting a system to Frank Zappa and struck up a very positive audio relationship with his FOH engineer Ron Lorman who later worked with Miles Davis. He was also involved with  the Savoy Club in New York. We put some TMS3 in there as sidefill and everyone went ape shit over them and it went from there. Once the US got onto it, TMS3 had its own momentum.  

Next issue Tony Andrews tells you why he dislikes line array, but is launching something that’s as easy to rig as line array; and which digital consoles he likes (hint: not many!).