I built my first full range driver project using Fostex FE-164 drivers a couple of years ago. This project was the ML TQWT speaker that has been really popular on the DIY forums and has been successfully copied many times. The ML TQWT represented a very inexpensive experiment to see what was achievable with a full range driver. I was very pleased with the results and the feedback from a number of DIYer's, having copied the design, has been extremely positive. After listening to this speaker system for a number of months, I concluded that real potential existed for full range drivers.
This encouraged me to start looking at the higher end drivers from AER, Lowther, and Supravox. The Supravox drivers had a little too high of a Qts for the quarter wavelength style of enclosures I had in mind. AER was a strong candidate but at the time only one very expensive driver was available. I was attracted to Lowther because I liked the idea of a series of drivers in any particular model line. The different Lowther models (PMA, PMC, DX, and EX) are available with multiple magnets of increasing strength. The magnet variable represents a degree of freedom, and potential upgrade path, for any proposed enclosure design.
The Fostex FE-164 had a Qts that was near the maximum value for any of the Lowther drivers. The question remaining was if a Lowther driver with a stronger magnet was selected, what challenge in the enclosure design would accompany a much lower Qts value. To try and answer this question, I hedged a little and purchased a pair of Fostex FE-208 Sigma drivers and built a second full range driver project. Again, I was pleased with the results I achieved with the less expensive Fostex driver. Even more important, this speaker was a final proof of concept that made me very comfortable using a driver with a Qts value anywhere in the range between 0.2 and 0.4. These Qts values almost bracket all of the Lowther drivers in any given model line.
For over a year, I read everything I could find on the Internet related to Lowther drivers. From the information and opinions I collected on the various forums and dedicated websites, there did not appear to be a consistent picture. Lowther speakers appeared to produce a love/hate response from those that had heard the different commonly available enclosures. Because of my location, I was not going to be able to sample several different Lowther speaker designs before deciding if this was the right direction for my next DIY speaker project.
Having already built two full range driver systems, I began to very carefully reread the descriptions on the forums of the sound characteristics produced by various Lowther speaker systems. Reading between the lines, I started to see a couple of consistent trends in some of these reports. The positives expressed included superior mid-range response and imaging. The negatives expressed typically concerned a very weak bass and a shouty overly dominant mid-range response. While the descriptions of the mid-range response appears to be inconsistent, I interpreted them to be primarily caused by the particular driver and enclosure combination coupled with a baffle step response problem. A baffle step problem will kill any perception of decent bass output and overstate the mid-range response. No matter what direct radiating speaker enclosure, potentially a 6 dB drop in output may be observed in the bass frequencies unless a correction circuit is included in the design. The observations being described were some of the same things that I had heard, measured, and corrected in my two Fostex speaker systems.
A typical full range driver set-up will most likely employ a low wattage tube amp with a direct connection between the terminals on the amp and those on the drivers. In these systems the efficiency of the driver is critical and the assumption is made that any passive circuit device between the amp and the speaker will degrade the total system performance. Usually, a high efficiency driver comes with a large strong magnet resulting in a low Qts. In my opinion, the choice of a low wattage amp and the requirement of a direct connection to a high efficiency driver significantly limits the flexibility in the speaker system design. I am not saying this approach is incorrect or inferior, I am just pointing out potential limitations that come with this approach.
My approach to full range driver design is totally different from the majority of people using this style of speaker. For my system, I use a 200 watt/channel solid state amp. This high power rating doesn't automatically cause a threat to the survival of a full range driver at normal or even elevated listening levels. Using this amp enables a number of design options to be possible.
Based on SPL response measurements of a finished speaker enclosure, I design a minimal amount of passive correction circuitry to re-balance the SPL spectrum and correct any baffle step problems. I feel that the improvements in the perceived bass response, created by the resulting balanced SPL frequency response, are far more significant then any potential degradation caused by these passive circuit components. In fact, I have not been able to measure any unanticipated changes in the SPL frequency response after adding an engineered baffle step correction circuit.
Adding correction circuitry will make the speaker sound different. The stong dominant mid-range response will be lowered and more in line with the bass output. That strong mid-range response can be very enticing and can lure the listener into thinking that they are hearing more detail and greater dynamics. But after extended listening I came to the conclusion that the balance was much better in a speaker with correction circuitry. There was an accurate balance between the bass and the mid-range and the mid-range had not really lost any of the detail or dynamics.
Engineering the correction circuitry is the key, applying a trial and error approach to designing and building correction circuits will probably do more harm then good. A trial and error style of circuit design may be the source for the rule of thumb that passive circuit elements degrade full range speaker performance. However, coupled with introducing correction circuits is generally a loss in speaker efficiency, but with the power provided by a solid state amp this is not a significant concern.
After much consideration and discussion with Jon Ver Halen of Lowther-America, I decided to try the DX series of Lowther full range drivers. I aquired DX2, DX3, and DX4 drivers with the new rolled-edge whizzer cone from Lowther-America. I really enjoyed discussing Lowther drivers and speaker design options with Jon and will continue to discuss this project with him as I progress.
Every project I design and build follows the same basic path. Measure the Thiele-Small parameters of the raw drivers after a suitable break-in period, based on the measurements design a quarter wavelength style of enclosure, build and finish the enclosure, assemble the drivers in the enclosure and measure the impedance and SPL frequency responses, design and install any required correction circuits, measure the finished speaker system, and document the project. For this Lowther project, I have decided to document each of these steps as the project progresses. The following is the first installment in this project and I will be adding more as things progress.
I welcome any comments on this Lowther project. Please feel free to e-mail me directly, I will do my best to respond to each e-mail promptly.