Who makes the lowest distortion speaker drivers

Good way to get the thread going again by stirring up a bit of controversy. :D

................

In my opinion the ringing of a crossover should be treated on equal footing with any driver resonances to keep it in context - look at the CSD of the total summed response and see how the crossover resonance compares in severity to the cone breakup resonances... if you can greatly reduce audible cone breakup resonances at the expense of slightly more ringing at the crossover frequency (but still considerably less than the driver resonances that remain) then I'd say that's a good tradeoff. You also have the possibility of using linear phase filters to improve matters while still achieving higher order slopes.

Of course you don't want to make the filters any steeper than necessary, because you are then making the ringing at the crossover frequency even worse for little gain in reducing out of band response of the drivers.

For this reason I don't think there's any reason to go beyond 4th order for crossovers except in exceptional circumstances, and even then I would question the choice of drivers - in most cases I think 5th order and higher is beyond the point of diminishing returns.

If you stick to wanting in-phase summing, which IMO is desirable, that basically leaves you with 2nd and 4th order acoustic crossover responses being optimal for most situations.

Hi DBMandrake,

I couldn't agree more - but i have a question. Have you ever managed to make the crossover ringing visible at waterfall plot of a finished loudspeaker or have seen that someone managed to record it ? Thus far i have not, nor have i managed to hear it.
 
My understanding is that the ring is actually a resonance in the response caused by too steep of a slope. Nature doesn't want a sharp bend so the response becomes underdamped at that point. On active system you can do a variable slope which is good. For a regular passive crossover 4th order seems to be about the point of diminishing returns.
 
Last edited:
You can do variable slope passive filters. Usual is 12dB/o at the crossover region and then make it dive outside the passband by additional notch or tank filter - i've done it lots of times. Magico does something like that. My point is that even very sharp roloffs (as seen in Joseph Audio) aren't making any ridges of delayed energy visible at waterfall plots. I'm not sure how steep you need to go to be able to measure and visualize a ridge.
 
Last edited:
You can do variable slope passive filters. Usual is 12dB/o at the crossover region and then make it dive outside the passband by additional notch or tank filter - i've done it lots of times. Magico does something like that. My point is that even very sharp roloffs (as seen in Joseph Audio) aren't making any ridges of delayed energy visible at waterfall plots. I'm not sure how steep you need to go to be able to measure and visualize a ridge.

The main problem with a variable slope passive filter is the impedance swing that happens after the first filter. That's what I've heard anyways. This is not as issue with active setup and much more precision is possible. I would be interested to see a "ring" show up in a waterfall.
 
If impedance swing will exist and will it be a problem or not depends on execution of filter, regardless if it is elliptical or regular.

Lots of impedance graphs here of Magico speakers mentioned without a visible issue:

SoundStageNetwork.com | SoundStage.com - Loudspeakers

In my opinion, there is too much talk about higher order crossover being bad because of resonance but i just don't see any proof of that being a measurable or perceivable issue in finished loudspeakers. I have experimented with higher than 24dB/o acoustic slopes but i choose drivers to get away with less - not because of ringing but for practical reasons.
 
Last edited:
I also have JBL 245x compression drivers, currently paired with JBL 2226 (15") woofers and 2245 (18") subwoofers.

It has occurred to me that JBL 4" voice coil comp. driver could play sufficiently low to cross directly to an 18" so yeah, I think it could work for a driver like the 2242 that doesn't break up too low.
The 2268 is quite a close replacement for the 2242 in most parameters so it should work too, I haven't looked at break up of the 2268 but I expect it's similar.
The more symmetrical dual differential voice coil of the 2268 should have lower 2nd harmonic distortion but it probably hardly matters in a domestic installation, should be practically inaudible for either driver at any tolerable level.
I think the main improvement of the later driver is simply that it's less massive, nice for a tour sound speaker but mostly irrelevant for your use.
So I would sell the 2242s and replace with 2268 only if the swap-over cost is small.

The AE is nice but wouldn't be my choice, unless maybe it has the Apollo enhancement.

Best wishes
David

I was looking at the M2 horn for a while, but its crossed at 800hz, and I assume it cannot cross lower. I would be concerned the 18" would be beaming too much at that frequency though.

JBL state max crossover frequency as 1khz though mind.
 
The single best post I have ever read on DIY forum

But the beam is still tracking the sine.

In a linear system the modulation of a sine wave does not change the frequency or create new frequencies.

I do understand the relationship of time and frequency and what the Fourier transform can show us. What can be seen in time should be seen in frequency. If there is something seen in one domain and not the other, somethings wrong.

Modulating magnitude of a sine wave in a linear system should not create “new frequencies” as were called side bands in a previous post.

Please show me where this is not correct.

I am also familiar with the AM and FM but we’re not talking about transmitting or demodulating RF down to audio frequency.

I am not here to argue. I have spent a lot of time in class’ and in the field learning what is and isn’t seemingly measurable and what measurements make a difference in sound system integration and tuning.

This whole side band thing is not in any of the Handbooks for Sound Engineers pages, never showed up in any of the measurement articles or discussions on ProSoundWeb, was never mentioned in TEF training classes or in the Heyser anthologies, or SMAART training classes or any other book I have read, I think this is the stuff of audio mad hatters or the well intended miss applying or misinterpreting test results.

I know those are strong words but several times I have attempted to drill down into a couple of subjects in this thread the point gets intentionally obscured I believe until there is no question to answer.

This sideband thing flies in the face of everything I have come to understand.

GedLee’s statements of what is and isn’t audible as far as distortions go is on some fronts hard to swallow. I am not going to attempt to do a study to challenge him but the Burning Down the House distortion paper is very suspect to me. Sorry Dr Geddes, no dissrepect meant, honest.

The brief discussion on a floor bounce DSP script to remove or call back floor bounce after the fact and then chasing that inverse script with yet another is theoretical insanity to me.

Direct answers instead of runaround discussions of why it doesn’t matter until the point is lost has not helped me a bit. A reference to a peer reviewed paperwould be a godsend.

I don’t think I am going to get anywhere without fighting for it so I should just bow out and be happy where I am.

Again thanks to anyone who really tried to help me gain some understanding.

Barry.

Here Here !
I agree 100% with you Barry.
 
In a linear system the modulation of a sine wave does not change the frequency or create new frequencies.

I'm afraid this is incorrect.

If you play 100Hz and 101Hz at the same time, you'll hear a tone varying in magnitude.

Beat (acoustics - Wikipedia)

Similarly, if you take a 100Hz tone and then vary the magnitude at the same rate as in the previous example (by turning the volume control quickly or whatever), a Fourier Transform will show 100Hz and 101Hz.

Chris
 
I'm afraid this is incorrect.

Chris

Yea, I don't know why that was repeated because it is completely wrong.

Modulate a 1 kHz sine at 100 Hz and there will be peaks at 900, 1000, and 1100. If it is amplitude modulated the sidebands will be in-phase, if it is frequency modulated they will be out of phase. If you see different heights of the sidebands, as I have, then you have both amplitude and frequency modulation.
 

AllenB

Moderator
Paid Member
2008-10-18 11:31 am
Trying to decipher this apparent misunderstanding. Playing a 1kHz tone at the same time as a 100Hz tone doesn't mean they are modulated. The lower envelope follows the upper envelope. To get them to heterodyne, and have the total envelope vary in amplitude would require a non-linearity.