1

Topic: The true horror of passive crossovers !

A while ago on this forum there was some discussion about loudspeaker crossover networks and the relative merits of the active and passive approach.  This prompted me to investigate passive crossovers further.

First, some background:-

Loudspeaker drive units are designed to be driven by a very low impedance, and the published frequency response plots from the manufacturer will always be shown with the drive unit driven directly from an amplifier, without any intermediate crossover network.  If the driving impedance is not low, the response will be adversely affected, and the amplifier will lose control over the speaker cone motion.  The degree of control is termed 'damping' and is defined by a 'damping factor', which is simply the ratio of the nominal speaker impedance (usually 8 ohms) to the driving impedance.  So for example, an amplifier having an output impedance of 0.5 ohms would have a damping factor of 16.

Amplifier designers generally strive to achieve a damping factor of 50 or more, and for good ones it can be several 100s.  But what happens when you insert a passive crossover network in between the amplifier output and the speaker drive units ?

The effect of the passive crossover :-

To quantify this, we need to establish the impedance of the crossover network output, since this becomes the new driving impedance for the speaker unit.  The plot below shows this impedance for a 2nd order passive Linkwitz-Riley crossover network with a crossover frequency of 2kHz.

http://hddaudio.net/wordpress/wp-content/uploads/2009/04/chris-2.jpg

As you can see, over most of the audio band the impedance is not very low, reaching a maximum value of 8 ohms at the 2kHz crossover frequency.  By calculating the ratio of this impedance to 8 ohms, we can establish the effective damping factor when the crossover network is in circuit.  This is shown in the plot below.

http://hddaudio.net/wordpress/wp-content/uploads/2009/04/chris-1.jpg

At 30Hz, the damping factor is a just about tolerable 34, but it drops sharply as the frequency is increased, dropping to less than 2 right from 500Hz to over 7kHz.  At the 2kHz crossover frequency the damping factor is just 1 !

So what does this mean ?

The results show that a typical passive crossover has an absolutely disastrous effect on speaker damping.  Over most of the audio band, the damping factor is very low, so the amplifier will have almost no control over the cone movement at all.  Were it not for the mechanical self damping in the drive unit suspension, the cone would be flapping around wildly.  As it is, the cone movement will certainly not be accurately following the applied signal voltage from the amplifier.

What about active crossovers ?

With an active speaker, the crossover network is connected at the amplifier inputs, and the amplifier outputs drive the speaker drive units directly.  In this case, the excellent damping factor of the amplifiers is maintained, and the cone motion is accurately controlled at all frequencies.  In the past, the main argument against the active speaker approach has been the cost.  But these days, power transistors are cheap, so there really is no excuse !

Chris.

2

Re: The true horror of passive crossovers !

This may explain the "Flat Earth" view which prefers a wild frequency response but used with a very simple passive crossover roll

Another view (not mine) is that active crossovers introduce distortions and colourations of their own, but my experience has always been that passive speakers (good ones, often of the "BBC inspired Thin Wall Box" persuasion) tend to "gentle up" the sound dynamics somewhat, losing something like 75% of the efficiency of the system.

I personally think that passive speaker systems can be made very well when the speaker manufacturer has absolute control over the crossover and drive units being assembled together, but there have also been expensive active systems marketed that are truly horrendous, any possible increase in quality being overshadowed by the desire to sell as much gear as possible...;)

Last edited by DSJR (2009-04-27 17:15:22)

Peace,

Dave

3

Re: The true horror of passive crossovers !

I must say that when you look into it as we have in considerable detail, you just keep asking yourself why anyone would even consider passive speakers, they are a horrendous bodge!

DSJR a passive crossover produces about 1% distortion and an active one 0.001%, so that's a no brainer too.

The problem probably is that non technical audiophiles are persuaded that in order to find true system synergy they must accumulate the different parts of an audio system from different companies who are best at each of these. It's arrant nonsense but it has give dealers and magazines scope for "building systems" and selling endless upgrades, something I suspect that enthusiasts will be reluctant to give up on.

I'm very glad that a non audio industry scientist has presented the facts because it shows clearly that our claims were anything but reckless and that the reaction from some quarters to ADM9.1s was simply ill-informed prejudice and ignorance.

I do agree that passive speakers can sound pleasant and there is a place for them, but not for hi end if we agree that the highest possible sound quality is the goal.

Ash

Manufacturer: www.avihifi.com / Contact: Email AVI / Blog: AVI HiFI

4

Re: The true horror of passive crossovers !

Except AVI passives..? roll

5

Re: The true horror of passive crossovers !

Chris,
Thanks for putting this in writing.
Does active has the same benefits in the time domain (phase, group delay)?

6

Re: The true horror of passive crossovers !

Roseval wrote:

Chris,
Thanks for putting this in writing.
Does active has the same benefits in the time domain (phase, group delay)?

To the extent that if the drivers are up to it, that they can conform more precisely to the theoretical ideal. Obviously if any equalisation is applied to correct for anomalies, then then although an active crossover will sound and measure better, phase will suffer as it does in passive speakers.

Tim Farney, a keen headphone user and regular poster on various Forums, had noted prior to his acquisition of ADM9.1s, that all passive speakers he's listen to had an emphasis in the up mid/treble region that he didn't like. ADM9.1s removed this objection completely as I understand it.

Ash

PS. Using steeper filters in Active designs helps give a far greater dynamic range than is possible passively, which means that they can replace very much larger speakers, especially if a sub woofer is available as an option for those with rooms that need it.

Last edited by Ashley James (2009-04-27 20:29:27)

Manufacturer: www.avihifi.com / Contact: Email AVI / Blog: AVI HiFI

7

Re: The true horror of passive crossovers !

Roseval wrote:

Chris,
Thanks for putting this in writing.
Does active has the same benefits in the time domain (phase, group delay)?

I agree with Ashleys response on this.

Active crossovers normally use a 4th order Linkwitz-Riley response, since this gives large attenuation in the frequency ranges where the drivers can 'misbehave', but also has quite a benign transient response due to the low Q (0.5).  Another property of this type of filter is that the output from the two drivers is precisely in-phase at all frequencies, so the polar pattern (and hence imaging) is stable.

Although it is possible to implement such 4th order filters in a passive crossover, component tolerancing is a nightmare, and you also need to accurately correct for the frequency dependent impedance of the drive units.  With active filters, on the other hand, it is easy to obtain very accurate 'textbook' responses.

Chris.

Last edited by cs (2009-04-28 13:04:04)

8

Re: The true horror of passive crossovers !

Tim Farney, a keen headphone user and regular poster on various Forums, had noted prior to his acquisition of ADM9.1s, that all passive speakers he's listen to had an emphasis in the up mid/treble region that he didn't like. ADM9.1s removed this objection completely as I understand it.

That sums it up pretty well. Even on very expensive speaker systems I hear a kind of glare in the upper mids that was completely absent when listening to the same material on headphones (Sennheiser HD580s). That glare is also absent on the 9.1s. In fairness, that particular problem is mitigated by most active speaker system. Only the most expensive active systems I've heard compete with the 9.1s overall, though. They are such little smoothies -- great clarity, wonderfully dynamic, beautiful imaging, and much more powerful than one has a right to expect from such a small package.

Tim

ON EDIT: By the way, I do not recommend listening exclusively on headphones for an extended period, then going to your listening room or hifi store and listening for the distortion described above. I can't stop hearing it now. It is a curse.

Last edited by tfarney (2009-04-28 12:06:42)

GrumpyOldArts.wordpress.com

9

Re: The true horror of passive crossovers !

The problem I have with headphones is that stereo is most of all STEREO.
Tim, any thoughts on this one?
http://bs2b.sourceforge.net/

10

Re: The true horror of passive crossovers !

This is interesting because I thought damping factor to be most applicable to bass.

i.e. an amplifier needs to be able to control the boom often found in loudspeakers.

Is this a graph for a 4th order active system?

And how might the graphs above change with amplifiers of different power?

Shake your Snaic with me.

11

Re: The true horror of passive crossovers !

Darren wrote:

This is interesting because I thought damping factor to be most applicable to bass.

i.e. an amplifier needs to be able to control the boom often found in loudspeakers.

Is this a graph for a 4th order active system?

And how might the graphs above change with amplifiers of different power?

Darren,

Both plots above are for a 2nd order passive system.

Yes, damping at bass frequencies is important because the cone motion is largest there.  But you can see that the damping factor is still quite poor even at low frequencies.  The biggest difference is at or near the crossover frequency, where the damping is almost non-existent so the cone/dome motion will be completely uncontrolled.

Changing the amplifier power won't really make any difference because the amplifier's output impedance will normally be very low (few milli ohms).  It will be swamped by the effect of the passive crossover network.

Chris.

12

Re: The true horror of passive crossovers !

Which is why we don't like Class D. Not only does it have much more distortion but it also has  filter on the output to remove the switching frequency, so you still have to contend with a low damping factor in an Active design.

Our friends at LS Design who've done Subwoofers for several companies found that the resistance in the filter kills control/power at low frequencies and also that the very powerful amplifiers have quite a high noise level. In our opinion low power Class D is fine for mobile phones, but not for hi fi.

Ash

Manufacturer: www.avihifi.com / Contact: Email AVI / Blog: AVI HiFI

13

Re: The true horror of passive crossovers !

But you know a certain amplifier manufacturer well used to bandwidth limiting will be developing new amps based on this technology and, of course, all the disciples will love 'em and treat them like the second coming..... roll

Last edited by DSJR (2010-02-19 13:44:49)

Peace,

Dave

14

Re: The true horror of passive crossovers !

We just got this as we often do and it makes the job worthwhile:
Goodmorning Ashley,

Thank you for shipping the speakers so soon. They arrived yesterday afternoon. My first listening experience is almost an emotional one, I am amazed with the transparency and the overall sound quality. I really think you achieved something extraordinary with this speaker design. Thanks again for your great service.

With kind regards,

Manufacturer: www.avihifi.com / Contact: Email AVI / Blog: AVI HiFI

15

Re: The true horror of passive crossovers !

Roseval wrote:

The problem I have with headphones is that stereo is most of all STEREO.
Tim, any thoughts on this one?
http://bs2b.sourceforge.net/

Yes. You simply don't listen to headphones for the same kind of imaging you get from speakers. They won't deliver it. Hardcore headphone listeners sometimes refer to the image cans create as a "headstage," because it takes what was mixed to lay an image out in front of the listener and puts it around (and inside) the head. It is not natural. Of course neither is the stereo image created for most studio recordings, so it is, in the end, a matter of degrees. Hifi is a series of compromises and trade-offs. When i put on my Sennheisers and lay back in my recliner, I trade that somewhat more realistic image for intimacy, detail, and the comfort of knowing that if my wife asks me to take out the garbage in the middle of Monk's Dream, I won't hear her. cool

Also, until the 9.1s arrived relatively recently, the Senns were much better than my speakers on all fronts except imaging. They were so much clearer and more resolving than my speakers that they made the speakers difficult to listen to. I no longer have that problem and the Senns are spending a lot more time hanging on their peg.

Tim

Last edited by tfarney (2009-05-02 11:54:55)

GrumpyOldArts.wordpress.com

16

Re: The true horror of passive crossovers !

Just found this Forum.I use a Crown K2 as the bass amp in an active system no noise or lack of bass control with this Class D amp.

17

Re: The true horror of passive crossovers !

I heard distortion only in the upper registers with ICE.
And to my amateur ears, it was subtle, not glaring by any means.
But I did hear it, in as much time as it took to switch amps,
as inaccurate as that pursuit would be.

It might be telling that B&O used all ICE amps in their first BeoLab model,
and in their subsequent model, used ICE for bass channel, but class A/B
for mid and high channels ...

First model:

http://www.bang-olufsen.com/UserFiles/F … _en_na.pdf

Subsequent model:

http://www.bang-olufsen.com/UserFiles/F … _en_na.pdf

18

Re: The true horror of passive crossovers !

There are issues with Class D for power amps, which means that the best analogue ones are quite a bit better, however there are plenty that are worse too!

Ash

Manufacturer: www.avihifi.com / Contact: Email AVI / Blog: AVI HiFI

19

Re: The true horror of passive crossovers !

As I understand it, there are two kinds of class D amps - digital and analog. At work, we have Pioneer Elite AV receivers which are fairly expensive, and use the B&O "Ice" amps. I think these are analog class D, though I'm not sure about that. I hear distortion in these amps, though it is not unpleasant. It manifests itself as a slight "warmth" (I hate that fuzzy term, but there it is) when compared, for example, to one of the high-current Yamahas of similar power driving the same speakers. It is extremely subtle, so much so that it could even possibly be my imagination. I also own a "class D" amp, a quite inexpensive Panasonic AV receiver I use as a headphone amplifier. It is digital, but I really can't comment on its distortion because the headphone section uses the digital front end, then converts to analog and amplifies with an op amp, not the digital amps. I've never had it hooked up to speakers good enough to hear its distortion, I'm afraid, though reports on the internet (for what they're worth) claim amazing clarity.

Tim

GrumpyOldArts.wordpress.com

20

Re: The true horror of passive crossovers !

Maybe this link is of use: http://www.hypex.nl/docs/allamps%20hypex%20layout.pdf

21

Re: The true horror of passive crossovers !

Roseval wrote:

Maybe this link is of use: http://www.hypex.nl/docs/allamps%20hypex%20layout.pdf

Oy...such a headache I have.

Seriously, I had to stop halfway and play some music.

Tim

GrumpyOldArts.wordpress.com

22

Re: The true horror of passive crossovers !

Anyone have these charts that Chris posted?

Most passive speakers I've seen have impedance charts that rise as you approach the crossover frequency.  What I can't make sense of is that this would suggest that damping factor is actually higher for passive speakers at the crossover frequency.  The assumption is that amp output impedance stays the same.

Does amp output impedance actually rise as impedance rises?  Just confused b/c the charts are missing.

23

Re: The true horror of passive crossovers !

Have to say that I was suprised with the 'accuracy' of the ADM9.1 active speakers when i compared them to my electrostatic headphones which are pretty brutally revealing.  Very close and much closer than many passive speakers I've heard.  Of course that may not suit some people as they tend to reveal bad recordings for what they are - but with good ones they are spooky ....

24

Re: The true horror of passive crossovers !

Links fixed.

Shake your Snaic with me.