Totem Acoustic Element Metal: The WOW factor!

Apologies the page on the Element Metal speakers didn't describe the no crossover aspect of the drivers hence my mistake. Enjoy your speakers klipschking!

Just came across this thread. Unbeknownst to me, I have been singing the praises of this exact pairing - Element Metals driven by an XA30.5. Using a Schiit Freya + preamp. The amp really does a nice job with these 4 ohm speakers. Agreed, percussion is fantastic. 

Per V. Bruzzeze of Totem, the Element series indeed employs the Torrent driver without a (conventional) crossover. The witchy magnetic motor assembly somehow accomplishes the frequency modulation.

Before (even before the Totem Metal) I always thought that "no xover" means exactly that "no xover", but now thinking of it, I realize that there is still xover even if the xover is in the magnetic element itself (at least based on Totem’s claim). It will invariably involve the losses of the electrical energy. You either do it by using an external inductor or by using the magnet. So the question is why energy losses using the magnet is less bad than by using an inductor? I suspect a lot of inductors, especially the garden variety type, have parasitic excessive high resistance and capacitance that retard the transient. I guess if you’re using quality inductors, it should be close to no xover as in the Totem Metal. Also, I also suspect the Totem Metal driver magnetic design also suffer from resistive losses but probably not to the same degree as an external inductor.

Another variable to consider is that speaker uses 1st order filter and it is true that there is some magic in the first order and one has to listen to it to understand why.

And of course V. Bruzzeze is a very good designer so he knows how to design a good pair of speaker and the "no xover" part is only a small percentage of why the speakers are good.

Also something to keep in mind is that the problem with xover is NOT because of the losses since because it is its job to create losses.  BUT it's the parasitic losses that is the problem.  So if you design a driver that meant for NO xover, but the driver itself has a lot of parasitic losses it would be just as bad as using external xovers.

I was thinking about the driver design and it may be as simple as extra winding on the motor inductor coil.  Basically you need a way to control the frequency response of the driver.  If you break down a driver, here are all the possible components in which you can/or cannot modify:

1. The driver diaphragm
2. The driver rubber surround
3. The driver spider
4. The driver motor magnet
5. The driver motor inductor winding

Of the five elements above, the only realistic components that can be modified for frequency response is either the motor magnet or the motor inductor winding.  

As for the motor magnet, I think it would be very difficult.  Although every magnet response will be frequency dependent, to engineer it to have the exact frequency profile you want would be very difficult and expensive.

So the likely option would be number 5 since it would be easy to optimize the inductor winding to the exact frequency response you want, so it is in effect of moving the external inductor from outside xover to internal of the driver.

Anyway, that's just my guess.