My design philosophy

Especially going from a bass driver to a midrange driver can be implemented as a first-prder crossover. Even going from a very large woofer to a small widebander can be implemented as first-order and the resulting sound can be so cohesive that it's almost impossible to tell there is a crossover at all.

It still requires the drivers to be time-aligned, but the end result is worth it for me. It also logically requires a wide operating bandwidth from both drivers and puts extra strain on the midrange unit. This is why full-range drivers are generally best-suited for this task as they are made to handle large excursion.

The same resulting phase and step responses can be achieved digitally through DSP, but I have not heard an implementation that didn't leave something to be desired compared to using good quality crossover components.

For the midrange to treble transition it is much harder to achieve a cohesive transition and few speaker designers have tried to do so in the past. Even so those designs aren't without their own set of compromises. Notably the vertical off-axis response and the max SPL from the tweeter running into its limits.

The vertical off-axis response can be controlled using a coaxial midrange/tweeter combination. But when I tried this design with a very special coaxial driver, one where the midrange can be run without a filter and still not sound too sharp, the sound was ultimately lacking a lively and transparent quality that I am looking for. Simply put, the drivers were not up to the task.

I won't stop trying to mate a midrange and a tweeter, but for now the designs on this website are both built around very special widebanders.

The sad reality is that the better and also generally more expensive crossover components allow for a significantly more transparent sound.

This goes so far that I find sometimes better sound can be achieved through a combination of high-quality crossover components and middle-of-the-road drivers than through TOTL drivers and standard crossover components. You would be surprised at the quality of crossover components in even the most expensive speakers.

This is also where a lot of the allure and transparency of widebanders come from. Sometimes widebanders can be built without any filters at all and despite doing so still achieve quite good neutrality. When using high-quality drivers this is where the biggest bang for the buck can be had, in my opinion.

Sometimes components not necessarily built for audio can be of very high quality. Such components can be a real bargain as it avoids the "audio-tax" of pretty components made either by or for small companies.

I think damping is a necessary evil. It's always a compromise, but when done right and with the right type of materials the resulting sound can be much more transparent than it would be with either the wrong kind of materials or with the wrong quantity.

One thing that is hard to get right is the right density and placement of materials. A denser material often doesn't have the intended effect of being more effective at damping in the frequency band of interest and placing materials near the walls is actually the least effective position. In many ways the way I damped my speakers can be seen as backwards compared to what companies or DIYers like to do.

Materials are once again a compromise. Some are inherently warm sounding and some are inherently sharp sounding. Like with most other things, I tend to avoid sharp sounding materials, which is why natural materials are my preferred type. Sometimes a combination of materials is the best solution, too.

This is why one of the materials used inside the Hathor prototypes was Monacor MDM-3, an out-of-production material that combines synthetic fibers with sheep's wool. It had a very interesting sound to it and it's a shame that it's no longer in production and can no longer be bought since I bought all the remaining stock.

There are other synthetic fibers which have a very interesting sound to them, namely the Mundorf TWARON fibers. These are probably the most expensive damping fibers around, yet they could still become somewhat too sharp sounding for some drivers.

This is why I generally fall back on good old sheep's wool. However there are different types and forms of sheep's wool, each with their own application and I find it's often best to use multiple different densities and types of wool inside one enclosure, something that I've done inside the Hathor speakers.

Some people love the sound of silver plated copper in teflon. I despise it. Sorry.

I like the sound of tinned copper in other materials with a low dielectric constant.
Silver wires also have their application and especially combined with natural insulation materials can have a sound that is unlike anything else.

Despite what an accelerometer may tell us I believe that cabinet materials all have their own sound.

I like using natural materials and especially the front baffle and baffle-to-driver interface matters most here.

Unfortunately there's no one solution here for the most life-like sound and it depends on the both the floor and the speakers. I like using a combination or compromise rather of rigid coupling and decoupling.

Which really doesn't say much, but really there isn't a solution that works for every speaker.

The unfortunate truth is there are two conflicting goals here:

Frequency response linearity at the listening position and imaging.

For the most linear frequency response it's often best to place the woofers near the floor and use a high crossover frequency to the midrange units. Unfortunately this can mess up the directivity index and the imaging of the speakers.

I like to go for close driver spacing allowing a smooth directivity index and better imaging, with the downside of floor bounce in a single-position measurement being visible.
Measuring at many different positions it mostly averages out, however and the power response from such speaker designs can look better than it might otherwise.

This is also where special designs like a D'Appolito driver arrangement come into play.