DIY Speakers, worth the hassles?

The results you get are very good, if you have a good design, good drivers and crossovers, and execute it well. If you fall short in any of these areas, you have crap. I have designed, built, and lived with many DIY speakers for going on 30 years now. I have definitely had some pieces of crap. The ones I have now are glorious. I have even had ones where I salvaged the drivers and electronics, and burned the lousy boxes I made that sounded like crap. Then, I sold the rotten sounding drivers and hardware that someone had told me were good for my project. That was a 3-way B&W 801 clone. Probably lost about $300 on that one. Another fiasco was a horn loaded 3-way that didn't match up with driver sensitivities. A horrid mess. Then I got smart and started learning what I needed to know. Amazing how much better things started sounding. I made a nice pair of 2-ways for my mother's system that sounded alot like the old B&W DM601. Probably a little better on the bottom end. However, the Dalesford 1" dome tweeter was a little grainy, and I could't afford the ScanSpeak that I should have gotten. I had finally got my crossover and driver matching skills. I made a 12db/octave crossover on that one that was perfectly matched to the drivers and was pretty seamless sounding. That was 20 years ago. If I could give you some advice, I would say go with top quality drivers for the design you have in mind. If the drivers aren't up to the task, it won't matter how perfect you do everything else, you'll be stuck in mid-fi land. Also, keep it simple. A good 2-way mini-monitor is a good place to start. Make your own crossover networks. Don't buy some generic crossover and expect it to work like it should. Finally, if your effort sounds like crap, remember that TWL had his first few speakers sound like crap too. That might make you feel a little better. It's the first step on a long road. Good luck!

Clueless, I'm glad you had a good time at the Bash. Wish I could have been there, too.
Regarding the comments by Wiggens of Adire, on the driver quality issue, I'd take that with a large grain of salt. Remember this, the drivers are the parts that are radiating the sound for about 90% of the frequency range, with the cabinets being responsible for only the bass response tailoring, and providing rigidity for the driver. If the drivers aren't up to the task, no cabinet in the world will save them. If you have a mediocre cabinet, you will still get the quality of the drivers for most of the spectrum, but may have poor bass response. In this regard, I totally disagree with Mr. Wiggens. I also disagree with his assessment of the different order crossovers. Phase shift is a huge problem, and higher order designs have more of it, and more components to soak up power and obscure detail. The main problem with 1st order is slow roll-off associated with 6db/octave slopes. This contributes to intermod distortion, and wider overlap causes more likelihood of driver matching problems in the overlap area. Tweeters are also subjected to higher power levels than may be desireable. However, the steep filters in high order networks can be much more difficult to seamlessly match and introduce large phase problems which affect many listening parameters negatively. If I were designing a multi-way system, I would use a 2nd order cascaded network which is a good compromise in slope steepness, and remains relatively phase coherent with a 180 degree shift that can be largely overcome with polarity reversal of one driver, while using a minimum of circuitry. I believe that Mr. Wiggens point-of-view is skewed by Adire's product line of drivers that are not in the state-of-the-art category, and I'm not sure why he is of his opinion on high order vs lower order crossover networks. High order can be good for tweeter protection and intermod reductions, but IMO the negatives outweigh the benefits.