The Speaker Guru speaks: Darren Kuzma

Driver Designer Questions

1. Do we know any of your driver work?
There are a couple different levels of "design work"; general specing/picking and more involved true design work. The general stuff would include picking specific drivers to bring to the DIY market such as expanding existing lines to other sizes, impedances or making adjustments to the parameters of manufacturers' stock items. True design work would be "starting from scratch" and designing drivers from the ground up including selecting cones, coil designs, motors, etc. I've done general specing for many drivers in the Dayton Loudspeaker brand, including the Neo tweeters, several of the classic woofers, the aluminum woofers, Euro Series, IB's, Titanic MkIII's, and Quatros. I've also picked and/or customized many of the Tang Band drivers that Parts Express carries. I've done in-depth design work on the shielded DVC drivers and most recently, the Reference Series woofers. I think that the yet-to-be-released Reference Series subwoofers I've been working on will be some of the most impressive and heavily engineered/tweaked drivers in the Dayton Loudspeaker brand to date.

2. Do we have to spend a lot on woofers and tweeters to get good sound?
I feel that it is entirely possible to make a "good" or "very good" speaker from even the most inexpensive drivers. However, to make a "great" or "excellent" speaker, it requires a certain level of quality in the drivers themselves. Whether low cost equals low quality is another question altogether. There are many drivers that I would consider low cost which offer high performance - at least in certain respects. Likewise, there are many high cost drivers that I would consider of a low performance in an overall sense (their build quality may be very good, but their useful, real-world performance may not be that exciting). It is really a matter of deciding what requirements you want to place on a driver and then picking which one will meet those requirements regardless of price. If you can meet your goals for a lower cost, it's a bonus.

3. I mean really, there are like, ten parts in a woofer. How hard can this be?
You do have a point - and it's not like designing pacemakers. If something sounds a little bit less than perfect, it's not life or death. But there really are a lot of different levels of "designing" a driver. If you wanted to get into the nitty-gritty of it, you could spend a lot of time modeling and designing small portions of the magnetic motor system, cone or surround. If you were an engineer coming up with a totally new cone material, you could spend a year on the materials science, geometries, production and testing of the material. Likewise, if you were designing a new motor system, you could spend a lot of time modeling the magnetic circuit in the virtual domain, and then spend a lot of time and money developing the actual production methods to put the motor into use. The best design on paper may not be feasible when it comes to a production or quality control standpoint. There's been a lot of work in the last 100 years and there are many off-the-shelf cones, surrounds, and other components available. If you are not trying to break new ground, it is a matter of choosing the right combination of components to get the result that you want.

4. What are the three main differences between a $5.00 and a $100 woofer?
The three main factors that drive the cost up for boutique-type drivers:

However! - there may be a "new" driver that has a high cost because of the factors above which may not necessarily end up high performance. Likewise, you could put together a driver with components that aren't as new or unique which would have a high performance without all of these extra costs. Or you could have any combination of above factors in a driver, or just one or two of them, or none of them, and still have a successful (or UNsuccessful) product. Don't forget distribution markups and shipping/currency conversion factors. It is often difficult to know the amount of money you are spending that is actually going towards performance.

5. Are parts mass-produced and you pick them from a secret catalog?
This is true to a certain extent. There are companies all over the world that produce the components that go into a driver. In contrast, there are also many driver manufacturers that make components in-house. It is feasible that you could get a bunch of catalogs and flip through and pick out all of the parts for your driver, but just like any business, most manufacturers work with certain suppliers. There are also real-world limitations to your choices. People have to remember that a speaker driver is a product just like any other and limited by economies of scale and supply issues.

6. Are all good drivers made in Scandinavia?
There are certainly some good drivers made in Scandinavia, nobody will argue that. But I would not say ALL good drivers are made there. I can think of several examples of great drivers from Germany, France, Israel, Taiwan and the U.S. to name a few. It just happens that there are a lot of resources and knowledgeable people in the Scandinavian area who make things happen. In today's world economy and manufacturing, there's no reason that production has to be centered there. I'm afraid that in the coming months and years, even the classic Scandinavian companies will start moving some production to China in order to remain competitive.

A lot of speaker manufacturers say they use custom designs from Vifa or Seas. Are they really custom or just small changes made to stock drivers?
It depends on the size and scale of the speaker manufacturer and their economic situation. In general, a small company will start with stock drivers. As they get more capital and develop new products, they will come up with custom versions with minor changes that can be done with minimal tooling investment. As they grow into a large manufacturer with their own R&D and testing departments, they will usually move more into totally custom drivers. There are many exceptions to this rule, so take this advice with a grain of salt. I think a more common sense, street-smart answer would be: if it looks just like an off-the-shelf driver, it probably is (or at least is very close to it).

8. Besides stiffness, what do you have to say about cone material?
Coming from the supply side of things, I have no strong feelings about one cone material vs. the other. I think they all have different sounds, but one isn't necessarily better than the other. If you like detailed, accurate sound, then I think harder materials will give you more of what you are looking for. If you like warm, relaxed sound, then you'll more likely get that from paper or poly cone drivers. Some may say this is a very un-technical, subjective point of view, but that's really what it comes down to. I'll leave the technical aspects for another time.

9. Tweeters, metal or fabric?
From a technical standpoint, a metal dome tweeter will have the lowest distortion and most "piston-like" behavior throughout the majority of their working range. However, they can get into trouble when it comes to the 10k - 20k region where peakiness can lead to a "metallic" sound quality. Through the upper midrange and into the lower treble, you can't beat the accuracy of a metal dome tweeter. To eliminate the metal dome sound, you have to move the first breakup peak above 25 kHz or so. At 20 or 22kHz, I think it still audible. By the nature of a soft dome, there are all kinds of decoupling and standing waves going on within the diaphragm at the high frequencies. If you've ever looked at an impedance plot and see a strong rise, but wondered why the output doesn't drop, this is part of it. As the frequency increases, a less effective mass is being moved, making the dome more ring-radiator-like, and hence the sensitivity goes up to match the rise in impedance. This creates a certain amount of inaccuracy and distortion at the top end, no matter how good the tweeter. I have always liked metal dome tweeters. They always seem a little more realistic to me. I've had the best experience with aluminum, it seems to offer the clarity of metal, but with less breakup or metallic sound. Most soft dome tweeters that I have heard leave me feeling unexcited about their performance.

10. Why do expensive woofers seem harder to tame in the XO?
I don't think it has as much to do with the cost of the woofer as it does with the cone material. Many of the modern expensive drivers use metal or very stiff paper cones which can cause breakup problems up high. I like to think that if a woofer has an easy-to-see breakup peak up high, then you know that it will for the most part be clean through the lower frequencies. Conversely, if you can't see any breakup peaks, then there's probably something going on somewhere in the useful range of the driver that isn't showing up in a standard frequency response measurement. Just because the net response is more or less flat doesn't mean there aren't distortions and cancellations still going on. So, if it's got a big peak at 10kHz, you know where the problem is, and you can deal with it accordingly. If it's got a "smooth" response, then who knows what is going on?

11. Any tips to know if a driver is good or bad on paper?
To have any insight into the capabilities of a driver, you need to have some frequency response or impedance information available. I am a big believer in measurements and think that you can tell 95% of how a driver will sound if you have enough printed information available. Things that I look at first are the frequency response, is it dip/peak free in the region where I want to use? I look at the impedance plot and see if there are any "snags" in the curve in the range that I am using. Do these correspond to any anomalies in the frequency response? If there is a big snag in the impedance curve, but a smooth frequency response at that point, my first thought is that somebody fudged some data a little bit, which undermines my confidence in the measurement as a whole. But, if the data agrees and looks good, then most likely the performance of the driver will match how it looks on paper.
Don't forget to look at Xmax specs. If you're going to be expecting 100 dB playback levels and the driver just doesn't have the excursion to support it, then you need to rethink the design. No matter how good the woofer, if you're running it at it limits it isn't going to perform well.

12. So which is it, tubes or SS?
I guess I'm a SS guy. I use SS equipment exclusively but mostly because I don't have any tube stuff handy. I'd be willing to give tubes a try one day but figure I'll have plenty of time down the road to spend large amounts of money for a few watts.

13. What does the future hold for driver design?
Current transducer design dates back about 100 years, so it makes it hard to say. I think that the current type of dynamic driver is so ingrained and familiar to just about everyone in the world, it is difficult to step back and come up with a totally new approach. I can think of a few new emerging technologies that could dramatically change the way transducers are built, but it would take a lot of experimentation and development to really get them right. Most likely, drivers will continue to evolve more or less as they are today, but with newer and better materials and manufacturing processes.

speaker icon

Copyright Peter Jay Smith, 2005 Return to helarc.com