2-way bookshelf with TCP115 and ND25-FA
I decided to build my first speakers. I wanted to replace my existing PC speakers with something that had a bit more bass while not taking up too much space. Also, I don't want to annoy the neighbors too much. The goal was to create compact PC speakers (bookshelves) capable of reaching 50Hz. They are for nearfield listening, and 90% of the time I sit in the same spot at my PC. Over time, I began to realize it wasn't as simple as I had imagined.
The enclosure is printed on an FDM 3D printer because I have one available and lack woodworking tools. The enclosure is made of PLA. I chose PLA because it should be more rigid than PETG. The print has 6 perimeters and a 40% gyroid infill density. I printed it with a 0.4mm nozzle. The side walls of the enclosure are 7mm thick, and the front and rear panels are 15mm. Size of the speakers is 24*13*25cm.
I started by selecting the midbass. I chose the Dayton TCP115-4. For the tweeter, I went with the Dayton ND25-FA. Both drivers are cheap, making them an ideal choice for a first experiment. I calculated the bass reflex using WinISD. According to WinISD, the port length came out to 29cm, so I had to figure out how to fold it.
The whole setup is powered by a Wondom JAB4. It's a 4-channel amplifier with DSP, allowing me to run the entire system bi-amped. For the PSU, I temporarily used a laptop power adapter. I set the crossover frequency to 3kHz. I measured the frequency response and had REW generate an EQ file. The result lowered the overall volume but flattened the frequency curve.
My initial impressions were that it sounded great. However, after some listening time, I started noticing fundamental flaws. When listening at 70% volume or higher, air leaks around the connectors become audible. I'm not sure if the chosen connector type is ideal, and although having active speakers seems cool, it might be better to make both passive and keep the amplifier in a separate external box. The second issue is that 7mm walls are not very thick. I feel that the side panels resonate, especially when listening to human voices, which degrades the overall listening experience. I thought about modifying the design slightly to have a 4mm plastic wall, a 5mm gap, and then another 4mm plastic wall, filling the gap with fine sand. There would be ribs inside the gap. I got this inspiration from a Facebook group. Since I am already considering such radical changes, I am also thinking about replacing the tweeter with an SB Acoustics SB26STC-C000-4.