As in all things in my life, I was compelled into designing loudspeakers because I was dissatisfied with what were available. They were either too expensive (read that as crazy expensive), or cheap (terrible sounding). What I wanted were loudspeakers with outstanding performance but at an affordable price. That was twenty years ago. Today, it’s even worse. Some of the prices for “High-end” speakers are absolutely ridiculous.
After twenty years working on speakers, it gives me great pleasure to share my knowledge with fellow audio enthusiasts. With some basic woodworking skills to construct the speaker box, you’ll be able to build for yourself quality loudspeakers at minimal cost.
My approach is to use reasonably cheap drivers and try to get the best out of them. Definitely no driver that cost $300. Actually, I try to keep below $100. Less than $50 is even better. I do not believe that good audio needs to be expensive. I leave High-End to people with deep pockets.
What I strive for in my Speakers
In all my designs, it’s always the Mid-Range that’s the epicenter.
I demand Transparency. Voices must be clear and cut through the mix. Once that’s established, next in line is the treble. There must be no hardness. Highs must sound as sweet as possible. The last item is the bass. I want the mid-bass to have definition.
My speakers are designed for Free Space (ie 4 pi). Reason being that they sound clearest that way.
In my setup, the speakers are at least 3ft away from back and side walls. Listening distance between 6ft ~ 8ft, on axis.
I do not recommend placing them on the floor or against a back wall. This will have a detrimental effect on the sound. The bass will “bloom”, causing the vocals to lose it’s clarity.
LMS | OmniMic | SoundEasy | DATS
The above are the tools I use in designing speakers. However, designing speakers is not all science. In the end, I still use my ears for final judgement.
Simulations vs Actual Measurements
I do not design speakers based on computer simulations only. Reason being that when a driver is mounted onto a box, the frequency response is not the same as when it’s mounted on an IEC baffle. The only way is to build test boxes and work from there. It’s costly and laborious but accurate.
You will notice that most of my boxes are prototypes. This is due strictly to time constrain. All my research is self funded (read that as “labor of love”), I simply cannot afford the time to make nicely finished boxes. Feel free to use your imagination. As long as the internal volume is the same, they will sound fine. For material. I use either MDF or 3/4″ ply. I have no preference. Whatever is available in my workshop.