How good is the crossover in your loudspeakers?


 

I just watched a Danny Richie YouTube video from three weeks ago (linked below). Danny is the owner/designer of GR Research, a company that caters to the DIY loudspeaker community. He designs and sells kits that contain the drivers and crossover schematics to his loudspeakers, to hi-fi enthusiasts who are willing and able to build their own enclosures (though he also has a few cabinet makers who will do it for you if you are willing to pay them to do so).

Danny has also designed crossovers for loudspeaker companies who lack his crossover design knowledge. In addition, he offers a service to consumers who, while liking some aspects of the sound of their loudspeakers, find some degree of fault in those loudspeakers, faults Danny offers to try to eliminate. Send Danny one of your loudspeakers, and he will free of charge do a complete evaluation of it's design. If his evaluation reveals design faults (almost always crossover related) he is able to cure, he offers a crossover upgrade kit as a product.

Some make the case that Danny will of course find fault in the designs of others, in an attempt to sell you one of his loudspeaker kits. A reasonable accusation, were it not for the fact that---for instance---in this particular video (an examination of an Eggleston model) Danny makes Eggleston an offer to drop into the company headquarters and help them correct the glaring faults he found in the crossover design of the Eggleston loudspeaker a customer sent him.

Even if you are skeptical---ESPECIALLY if you are---why not give the video a viewing? Like the loudspeaker evaluation, it's free.

 

 

https://youtu.be/1wF-DEEXv64?si=tmd6JI3DFBq8GAjK&t=1

 

And for owners of other loudspeakers, there are a number of other GR Research videos in which other models are evaluated. 

 

 

bdp24

The thing that Danny Ritchie loves best is Danny Ritchie and far too many believe him.  But it sells, so bully for him.  Of course nobody here seems to remember that the best crossover is an active crossover.

Here's WHY active crossovers are so very much better than passive. A single loudspeaker driver is an inductor, and provides a frequency dependent, reactive load to an amplifier. Looking at the image here, the blue line on the bottom is the frequency dependent impedance curve for an SB Acoustics SB29RDAC Ring Dome Tweeter, and it typical of any dynamic tweeter. As you can see, it is anything but flat, yet it is listed as having a 4 ohm impedance. It's 4 ohms at about 1200 hz, but at 600 hz, has an impedance of nearly 10 ohms.

Now if you put a passive crossover circuit in front of it, you add capacitors, resistors and inductors, which then give you a frequency dependent impedance curve which looks like a Coney Island roller coaster. And that's just for a tweeter high-pass circuit.

Now when you add in mid and bass drivers, with high and low pass filters there... It's a real mess. But we're not done there yet. Nope. Many of your extreme hi-end loudspeakers add in equalization to their crossover designs, which makes that impedance curve even worse. This is very hard for an amp to properly manage. That's why people drop many, many thousands of dollars on things like Krell, Threshhold, Bryston, or Rowland Research solid state power amps.

Now when you use an active crossover, an amp channel only has to manage a single driver. There's no passive, reactive component in between the amp and the loudspeaker driver. Then you don't need a megabuck amp to deal with it.

Many of the best pro studio monitors are powered, with active crossovers and multiple amp channels.  All of the Linkwitz loudspeaker designs use active crossovers. His  designs have used both analog and digital crossovers. There are some digital crossovers that offer DSP EQ, which allows you to tailor the total system response for the room you are in. Then you're not just limited to whatever sound your speakers give you in the room you're stuck with.

The lowest cost active crossovers are typically pro grade, from manufacturers like Behringer, dbx, Rane or even Nady. There are many manufacturers. Some of the best known home audio digital crossovers are from miniDSP.  Even a Behringer active crossover is far superior to any passive crossover.

Another major benefit is that you can use much, much lower powered amps when you use active crossovers. A lot of power is wasted having to push through a passive crossover. You really don't need to push many watts into a tweeter or mid-range driver to get a lot of level out. You could even run a single ended tube amp on your tweeter, and a mid-level tube power amp on your mid-range driver, and a solid state amp for the bass driver. You have a lot of options.
 

Just as a point of reference -- there was a recent thread regarding ATC monitors where the ATC distributor stated that the active versions were superior to the passive versions.  Yet several ATC owners stated that they strongly preferred the passive versions.  Their consensus was the passives were more emotionally satisfying.  Go figure.

@russbutton Awesome detail, I wish more people understood how critical it is to learn how to use and active crossover to get the best out of the music, speakers, amps, and the room. I was an early adopter of the dspNexus from Danville Signal Processing and it is the first DSP to operate 8 channels at 192kHz/24bit and you can custom design the crossover with Audio Weaver. Design it to be as simple or as complicated as you like such as any traditional slope or go crazy with biquads like me generated through Room EQ Wizard and Multi Sub Optimizer. 
 

thanks,

Steve

@bdp24, my condolences. You are a brave chap attacking peoples' cheapo crossovers! What nerve? March onwards with the torch held high.

Why the Danny haters? Except for the uber expensive just about every XO can be improved. The parts quality can almost always be improved and there are those who will argue that a cap can not improve the sound and if it measures to spec and voltage rating it's good enough, don't fix summin that ain't broke. To those I say, you are quite right, no cap can improve the sound but I choose caps that do less damage. All components will degrade the sound.

An example: I built a pair of Zigmahornets that use the little fostex FE103 full range speaker. Zero XO but had no bass, it's only about 4". Can't get more bass so I tamed the mids down with a parallel coil, cap and resistor in one leg. This is not a XO as such just merely knocking the mids and top end down a little and did indeed provide more 'apparent bass' but it was soon tossed out because it killed the life of the otherwise great little thing.

That's what components do. Take the much vaunted LS3/5A BBC monitor with 2 dozen components to force the response to be smooth but to exhibit the famous/notorious BBC 'smile' so a little tilted up in the bass and top end, hence the smile. These were engineered for mobile recording studios where the speakers were about a foot from their face.  I just hear them as dynamically constipated.

I design speakers and place importance on accurate phase tracking at XO and if the drive units are chosen carefully with smooth roll off out of band then a simple XO will work and work well without excessive editorialization. Some manufacturers boast about how they have with many components achieved a flat response but get them home and the measured in-room response is anything but flat!

One of the worst offenders is the ubiquitous sand cast resistor, most of them with magnetic ends.

Now if you don't believe that a resistor can change sound for the worse then here is a cheap experiment:

Cheap experiment:  Most tweeters have an attenuation resistor in series ranging from 2 to 10 ohms, it doesn't matter. Lets say it is 4.7ohms, remove it and replace it with a string of 10 white abominations of 0.47 ohms soldered in series = 4.7 ohms, right? How do things sound now? But, but, but, yeah now you sound like a cheap Chinese two-stroke: