Welcome to my website which primarily focuses on audio-related topics, specifically do-it-yourself speaker design. Here you can find several different DIY loudspeaker projects that I've designed and built. I regret to say that many of my speaker projects have become obsolete as certain drivers are no longer manufactured. These projects remain on my site primarily since they contain information about specific design strategies and in some cases these drivers can be on the used market. Some of my non-audio related hobbies and interests can be found on the Interests page. As I gather information regarding speaker design and audio in general I will add it to the Audiofiles page, my audio library so to speak. My email address and educational background can be found on the About Me page. My Links page contains several handy links to many different web sites categorized by type. The links page barely brushes the surface of audio/speaker related web sites to be found on the internet, but it provides a good start for anyone interested in DIY speaker design. Thank you for visiting and enjoy the site!
My BackgroundMy name is Roman Bednarek and I've been building speakers for about a decade now. I have a background in Electrical Engineering with a Master's Degree in Wireless Communications. I began my first speaker project back in 1998. I started with a low budget TMM design using Bose clearance drivers and several different tweeters. I built my first set of cabinets out of plywood and Poplar hardwood and boy did those cabinets resonate but produced a very deep bass response. I rebuilt those cabinets several times using MDF and optimized the design to the best of my ability at the time but later realized that there is only so much performance that you can squeeze out of some Bose clearance drivers (although I got lucky because they seem to have BSC built into them).
After completing my first design I began frequenting the speaker discussion boards, Madisound and Parts Express in particular. Along with reading some of the basic speaker books (Vance Dickason's Loudspeaker Design Cookbook for example) these discussion boards are a great resource for speaker design information and are the only source for information beyond the scope of most of the textbooks that are available. As a beginner project I decided to build a 250W Titanic 1200 subwoofer using a PE plate amp. This was a relatively easy project to complete because there isn't any crossover design work involved but it still sharpened my cabinet building skills (which still need some work even now).
I next began building some proven designs in order to get a feel for how a good set of DIY speakers can sound. However, I decided to make some minor modifications to some of these designs to suit my budget or driver availability. My first project was Tony Gee's Proteus speaker but I substituted the Seas 27TFFC tweeter for the original Seas Excel tweeter in order to save some money and have some fun tweaking the crossover. I also modified Tony's DD8 design by substituting the Seas L17RCY/P woofer. For both of these designs I initially tuned the crossovers by ear (which can be very tricky) and then later updated them by measuring the drivers and retuning the crossovers but only minor modifications were required. I'm still using the Proteus jrs. in my 5.1 surround sound system and my brother is using the DD8.1s as front channels in his 7.1 system.
At this point I began looking into some of the tools available on the net for designing speakers. There are a lot of great (and free) tools available on the net. My favorite is Speaker Workshop which has the capability to measure drivers (both T/S and frequency response) but some people find it difficult to learn especially when it comes to taking measurements. There are also a LOT of excellent tools available from the FRD Consortium who provide several Excel spreadsheets to allow you to model different aspects of speaker performance including baffle diffraction effects, enclosure bass response, vertical polar response, room response and several others. I highly recommend both of these resources and there are some short tutorials on how to use them under the Audiofiles section of my website. As of 2010, there are some updated versions of some of the most popular Excel programs from the FRD Consortium site found at this link Jeff Bagby's Loudspeaker Design Software.
I next invested in some measurement equipment and bought a Behringer ECM8000 microphone and UB802 mixer (don't get the UB502 because it doesn't have phantom power for the mic). I also built a Wallin Jig II for use with Speaker Workshop and sent my mic to Kim Girardin to get calibration curves generated for it which insured that my measurements were accurate. I learned how to use Speaker Workshop to take impedance and frequency response measurements and began building designs of my own. My first solo design was the Hyperion project which I built for a friend. At this point I was still not entirely confident with my crossover design skills and referred to Geoffrey Dillon's Metaphor design for guidance. I even "borrowed" his woofer filter because it sounded so good. The Calypso is another design where I "borrowed" a woofer filter (from Dennis Murphy) but the rest of my designs are completely original. The designs in the projects section are listed in chronological order with the most recent being at the top.
As I began building more designs I got a better idea what different drivers and crossover types sounded like. I also got better at designing crossovers as well. As you can see I began with two way designs to polish up my skills and then later moved on to three way designs. I tried to vary the designs a bit to round out my resume and get a better feel for different drivers and designs. In general the sound of my speakers has improved with every new project with a few exceptions.
It is important to distinguish the difference between a DIY speaker builder and a DIY Speaker designer as it relates to what type of projects you plan to take on. I chose to become a designer from the start which meant that I wasn't simply building established speaker projects that other experienced designers had devoted their time and knowledge into perfecting. This is a very important distinction to make for a new DIY speaker enthusiast because it is much more difficult to design a speaker than build an existing design and it is a good idea to first determine your goals when choosing a speaker project. If your goal is simply to build a nice set of great sounding speakers that sound as good or better than commercial offerings which cost many times more than a DIY speaker project, then I would suggest to choose an established speaker project and do a lot of research in order to ensure that you find a project that best meets your goals and has the highest probability to suit your listening tastes. However, if you plan to be in this hobby for quite some time and desire to design your own speaker projects, then I suggest that you follow a strategy similar to mine described above. I'll just add that the amount of knowledge and experience required to be a designer is much greater than that of a speaker builder.
Assuming that you plan to design speakers and make this a long term hobby, I strongly suggest that you follow a strategy similar to mine in terms of which projects to try first and which ones to hold of until later. In general, subwoofers are the easiest speakers to design and can provide experience in cabinet design and construction. I found that taking established designs and making minor modifications was a nice way of moving to the next step of speaker design as well as learn a bit about how other experienced speaker designers approach the hobby. Then when it is time to start from scratch with a new design, it is always best to start with a simple full-range single driver design or a basic two-way design. I strongly recommend against attempting a three way speaker design until you have a few two-way speakers under your belt because three-way speakers are exponentially more difficult to design and tune considering the amount of variables involved, especially with the crossover.
I've found a nice way to advance my design skills that should work well for just about any new DIY speaker designer. Start off with low cost projects to learn the basics. As I would complete a project I would try to sell it (that's the hard part) even if I only recovered the cost of the parts involved in the design. I would then use this money along with a little bit saved up since the previous project to invest in a new bigger and better project. As I advanced my projects improved and I was able to put more and more money into the next project while learning a lot with each project completed. It does get trickier to sell projects as they get more expensive though so make sure that your cabinet design skills are up to par.
Not only have I been building my own projects but I've helped out dozens of other DIY speaker builders around the world. I consider my specialty to be crossover design at this point and have helped a lot of people design and fine tune crossovers. Since most people don't have measurement equipment I've found the best way to design a speaker without this equipment is to use simulation tools along with manufacturer's frequency and impedance response data. The FRD Consortium tools are a great resource when it comes to determining what the final driver's response will be when it is mounted in the final cabinet. The key is to take the manufacturer's curves which are usually based on an infinite baffle with a given closed enclosure volume and determine the bass response in your target cabinet and baffle diffraction effects as well as generate the minimum phase information. Then this processed driver information can be used in Speaker Workshop or something like Jeff B's PCD (Passive Crossover Designer) (Jeff Bagby's Loudspeaker Design Software) in order to design the crossover. Even though I have full measurement capabilities I still use this method to pick out drivers and determine which ones will work well together prior to building the cabinets and buying the drivers. While this method (pure simulation) is the most effective way to design a speaker without the use of measurement tools, it is only as accurate as the data provided to the simulation which includes the accuracy of the manufacturer's data (T/S parameters, frequency response, impedance response, etc.) as well as the procedure of processing that data for use with a crossover simulator. Also, I cannot put enough stress on how important it is to actually listen to the speakers and make adjustments beyond what any simulation or measurements may tell you in order to arrive at a speaker design that sounds neutral and suits your listening tastes. Even with accurate measurments, a crossover simulation is only a snapshot of one type of speaker response and doesn't reflect what your ears can receive when the speakers are played in an actual room with the actual audio components providing the signal. So in any case, I always tell people to trust your ears before you trust any measurements and/or simulations because it is far too easy to try to adjust the sound on your computer and assume that it is telling you what sounds best when your ears are the only measurement device that can truly perform that task.
If you are interested in this hobby, please take a look at the projects that I have built, even those that may be obsolete which may have information about a certain design strategy or based on what I learned during the design of that project. I'm always open to positive criticism and am very willing to answer questions related to anything on my site, so please contact me if you want to (see the About Me section for my email address). Thank you for visiting my site, enjoy!
Legal DisclaimerAll of the information on this website is copyrighted and cannot be duplicated without my permission. The projects are for personal DIY use and are not to be built for sale or mass produced. I am not responsible for injuries sustained while building one of my DIY projects. These projects require skill at woodworking and soldering which can be difficult and dangerous for a beginner. Build them at your own risk.