Jack,
Are you able to share what slope crossovers are utilized in the new meta products? I've always been interested in Vandersteen speakers and the ideas of phase & time coherence. I recall a KEF Youtube video and if I remember correctly, you mentioned the potential of first order cross-overs. I believe there is also a Youtube video from PS audio where their speaker designer also toyed with the idea of first order cross-over networks. I write this as an owner of LS50s; R3s; and I have Ref 1 Metas on order although I'm tempted to go further up the line.
Hi
@MaggieRose,
The Reference META use mostly 2nd order electrical slopes. Time and phase coherence in crossover design is a topic that interests me too. However, a common misconception is that the crossover order is only due to the order used in the electrical filter. This is not the case, the true crossover order is the combination of both the electrical filters and the driver responses. This means that it's virtually impossible to achieve a true first order crossover and unfortunately means that this route to perfect time coherence is effectively closed.
Here's an illustration from the Reference 2014 white paper. Below we see the frequency response of an extremely good tweeter that is perfectly flat from 500Hz to 20kHz.
We now combine this with a perfect MF driver, with a flat response from 20Hz to 20kHz, and a 1st order electrical crossover network. The flattest response we can achieve is shown below:
The corresponding group delay is shown below and is not constant, indicating that time coherence isn't achieved. In conclusion, even with unrealistically good drivers it wasn't possible to achieve time coherence with 1st order electrical filters in the crossover.
The
white paper for the 2014 Reference series includes an entire section on the crossover order (starting on page 31) and gives some background to why we tend to use 2nd order crossovers.