MTM TL/Horn Geometry Experiment

I feel like I’m crashing the party.

Due to midlife crisis and a desire to relive my youth I’ve decided to build some ridiculously cheap speakers that tick all the boxes… all two: looks and performance. The answer is yes, if you asked, I’m lying to myself.

So, I stumbled on some 8” bass looking like drivers that look like they belong in a car and whose brand is invented and inconsequential. These speakers are likely unavailable to yourselves in the West. I measured them and they are suspiciously like Visaton WS 20-E (8 Ohm) but they aren’t exactly. I’ve paired them with my trusty Audax 1” textile dome which is sensitive but has poor power handling just to make things more interesting and difficult. It’ll be an MTM arrangement because two woofers are better than one… at least so I am told.

So, crossover needs to be high and luckily these 8” drivers carry well all the way into 5k where they start to roll off rather quickly and don’t have any weird breakup.

So far, so good! But that’s a lie. I’ve already experimented with closed box and vented box versions without much thought and just a willingness to risk cheap chipboard and they turned out good enough… particularly the vented version but I wasn’t satisfied because there is never enough bass. I will not speak of the rear loaded horn experiment. Here we run into physics and its problems for salt of the Earth people like me and you.

Regardless of driver selection and enclosure (or lack of enclosure) there is a pesky hard limit called Xmax. The enclosure and properties of the motor and attached bits dictates its movements in low frequencies in a dramatic way. Regardless of an ability to get a great response in the low end at 1W/1m it’s not necessarily going to play out well at elevated listening levels - Xmax will be hit and even if not, distortion will abound due to the diaphragms excited lower frequency movements combined with the rest of the typical frequencies.

Closed Box.png

^sealed box, 99dB @ 40Hz, why even wakeup in the morning?

vented box long port.png

^vented with a long port, 102.5dB @ 40Hz... getting better

vented box short port.png

^vented with a very short port, 107dB @ 40Hz... but I want 110dB! Life isn't fair.

In a sealed enclosure Xmax is easy to hit and reduces max SPL dramatically. In a vented box this limit isn’t so bad but still not great… turning the port into a literal hole in the wall did wonders and increased maximum SPL (This is the version I liked best so far).

Now I’ve come up with some geometry which is sort of like a transmission line with inverted short horns. I am skeptical of this (it comes natural to me) but I’m in the process of making a prototype to at least see if it’s viable. In simulation it exhibits high sensitivity to geometry change which makes everything awful… adding to the difficulty IRL and my suffering. Of course, there is the issue of dealing with the nasty hump but that’s resolvable. BTW, the box is about 20% larger than the vented box so very untypical. Also, just to make he point, it's possible to get louder but at the compromise of more volume and I want to keep the boxes similar to the vented version size - it's a compromise.

transmission inverse horn.png

^weird transmission line, allows much greater maximum SPL.

You might be thinking by now, ‘but Gary, you’ve lost the plot and why do you want so much SPL?!’ Well… it’s more about reducing driver movement in the low end. A lower maximum SPL indicates a great deal of unwanted movement. Of course, there are many other issues that come along with what I’m doing but I’ll try one thing at a time.

More to follow in the next few weeks - don't get too excited yet.
  • Like
Reactions: 1 users