Then you haven't been reading the same thread that I have. We've had lots of personal opinions and preferences as to why some people prefer passive, but besides that active is superior if looked at in purely objective terms, which is more then enough for me.
I think it will be superior for my system as I'm entering the world of compression drivers, horns/waveguides etc. and am a noob with crossovers having only used FR drivers before.
There is no way I could reasonably & cost effectively play with passive crossovers at my level of understanding/competence with physical circuits. The chances against me hitting the right combo first time are HUGE and I'd probably damage something along the way. I can however play to my hearts content within the limits of my drivers & personal knowledge on something like the minidsp without much fear of wrecking a driver.
As for sound? That is to be discovered, but as I've not listened to a speaker with a crossover for a few years now, I wont be comparing passive with active, but rather active with nothing at all!!
I just couldn't make the jump so easily if it weren't for active solutions, so it wins hands down.....for me, for now
Not sure about "purely objective" but certainly from a "purely engineering" standpoint it's superior. That does not necessarily mean it sounds better. Unfortunately.
Superior engineering doesn't bring better results, hmmm... what a pity.
Or, that engineeing is still not fully optimized.
To name a few, in an active system, normally there's no passive component between amps and drivers, so their individual flaws or any mis-match between them would be more obvious - more clearly revealed.
For example, some combinations might still need zobel on the driver. Damping characters of amp and speaker alignment need to be optimized for each other. And it's to be remembered some physical issues of drivers can't be cured by filters... etc.
In some cases I've seen, people tend to be over-confident on active xover in the first place, believing it would be better anyway -- because the pros say so. But then, they don't pay enough attention on optimizing everything, like they do to the passive ones. Initial excitement turns to dissappointment and is given up early.
Just my 2c. (and I haven't followed this fast-growing thread very closely, so sorry if those have been brought up before.)
Or, that engineeing is still not fully optimized.
To name a few, in an active system, normally there's no passive component between amps and drivers, so their individual flaws or any mis-match between them would be more obvious - more clearly revealed.
For example, some combinations might still need zobel on the driver. Damping characters of amp and speaker alignment need to be optimized for each other. And it's to be remembered some physical issues of drivers can't be cured by filters... etc.
In some cases I've seen, people tend to be over-confident on active xover in the first place, believing it would be better anyway -- because the pros say so. But then, they don't pay enough attention on optimizing everything, like they do to the passive ones. Initial excitement turns to dissappointment and is given up early.
Just my 2c. (and I haven't followed this fast-growing thread very closely, so sorry if those have been brought up before.)
All this says to me is that the active implementations weren't up to snuff...
This is simply incorrect, the job of both the passive and active xover is to tailor the voltage presented to the drive unit so that its response follows are specific target. Even if there are flaws inherent in the driver, or if there is an inherent flaw between the amplifier-driver interface then it can be measured and accounted for with both type of crossover. This is somewhat easier to do with an active crossover then a passive.
How does this in any way make the situation worse for an active loudspeaker?
I think its more accurate to say that the vast majority of people on DIYaudio don't end up properly optimising their loudspeakers be they active or passive. Those that do have the equipment and skills necessary to do this passively, will almost certainly posses the ability to do this with an active loudspeaker too. They would never automatically assume that going active means they can be less stringent in their design methodology. The inexperienced might automatically assume that active is some sort of cure all, but then these people end up with less then optimal results with either active or passive,although usually end up with something better designed if they go active because it tends to be more predictable. This is especially more so with the miniDSP type products that are now available, these aren't my idea of high fidelity, but there is a lot that they do do a lot right.
Passive components between amp and driver help in covering up some flaws. The flaws might be in driver, might be in amp, or both.
Try active on compression drivers and horn without extra cautious, hell breaks loose.
And, although eventually driving signals on the drivers should be the same no matter they're shaped in line level or after power amp, there're still many differences in the results.
For example, HF noises and distortions of an amp can be attenuated by the passive LP section. So those only appear on the tweeters. But in an active system, they can appear on both tweeters and mids.
I remember someone brought up listening fatigue on active system, or the likes. They're probably relevent.
Try active on compression drivers and horn without extra cautious, hell breaks loose.
And, although eventually driving signals on the drivers should be the same no matter they're shaped in line level or after power amp, there're still many differences in the results.
For example, HF noises and distortions of an amp can be attenuated by the passive LP section. So those only appear on the tweeters. But in an active system, they can appear on both tweeters and mids.
I remember someone brought up listening fatigue on active system, or the likes. They're probably relevent.
Active systems can be tailored with slightly falling treble etc, you don't HAVE to go for academically flat just because you can.
I do wonder if stronger driver damping could increase some cone breakup mode effects as the braking forces are stronger for shorter times. This might cause higher harmonics to increase
I do wonder if stronger driver damping could increase some cone breakup mode effects as the braking forces are stronger for shorter times. This might cause higher harmonics to increase
Some, sure. Others are beyond reproach, and sound darn good, too. There are many flavors. But that wasn't what I was getting at.
As I've said many times before, I used to be an active crossover devoté, even a snob. For reasons of "It has to be better because impedance, blah blah, passive components, blah blah, damping factor, blah, etc." All the reasons that get trotted out on a regular basis. And because I've heard some stunningly good active crossovers - I still like them. But I can't say they sound more like music than an equal well designed and implemented passive crossover. I was crushed to realize that and it took me years to admit it.
So yeah, I understand all the arguments in favor of active crossovers, I even hear the advantages in well built ones. But I still can't say that from listening to one or the other approach that either has a clear sonic advantage. Active should, but I've not heard it.
I'll do the favor of preempting the "You've never heard a good active system" argument before someone goes there - 'cause I have. Transistor, tube, DSP. Great stuff. Ultimately better than the best passives? I don't hear it. Just different, that's all. And that's what I have to go by, not what "should" sound better because of XYZ, but by how things actually sound to me.
This is some support info for those talking about reduced IM in a bi-amp situation.
Clipping in a Bi-amped System
I am still very much on the fence... it depends, there are lots of arguments in favour of each technique
dave
Clipping in a Bi-amped System
I am still very much on the fence... it depends, there are lots of arguments in favour of each technique
dave
Amplifier IM will be reduced in a bi-amp system.
Driver IM may be reduced if steeper crossover filters are used.
If the passive crossover acts as a significant source impedance, it may act more like current drive and get better driver IM than active, see Distortion example: Vifa - Current-Drive - The Natural Way of Loudspeaker Operation
Current drive vs voltage drive is one way that "traditional" active systems may sound harsh
Driver IM may be reduced if steeper crossover filters are used.
If the passive crossover acts as a significant source impedance, it may act more like current drive and get better driver IM than active, see Distortion example: Vifa - Current-Drive - The Natural Way of Loudspeaker Operation
Current drive vs voltage drive is one way that "traditional" active systems may sound harsh
Great post, CLS!!
Exactly, that's how i see it too.
Regarding current vs voltage drive, in any case using a fully active setup you could precisely choose how to drive each driver. Note that there is also "mixed drive" where the transducer is voltage-driven below a certain freq, and current-driven above certain freq.
On one of the amplifier-design books there was an explanation of how most distortions diminished when the amplifier (think of your typical class AB transistor amp) was asked to drive a much higher impedance load.
So the benefits may not be exactly due to "current drive" but simply because of presenting a higher impedance load to the amplifier.
That's one of the benefits of Bi-amping (not the active version).
The amplifiers handling the full passband see an average load impedance that is higher than that presented by the "whole passive crossover".
Each amp sees only a high pass filter and treble driver, or sees a low pass filter and a bass/mid driver.
that could have been nice in some cases(many)
but can't say I have had that luck, quite the opposite
why not the active ?
Hopefully, nobody's seriously claiming it is a panacea. (Except down low - given the size and cost of passive parts required, the greatly increased power demands, and the utility of easily being able to vary levels to fit a room make active a no-brainer for the modal region and below.)
In a system where large delays aren't needed, with gain structures optimized for active and the same transfer function for both an active and passive crossover, I highly doubt there would be any difference whatsoever between the two, let alone any real superiority to either approach..
Academically, I think the NAO Note design, which has a hybrid approach (active bass - midbass, passive midbass-midrange-treble) is the optimal from a cost-benefit standpoint, which is why that's what I'm doing in my next system. Sonically I just don't see the penalty, and there are fewer electronics boxes in the room than with an all-active approach. Someone with the time and comfort-level to build amps, or someone using drivers widely differing in sensitivity, may see a different balance as optimal.
Keep in mind it's not just the amps. Amps with perfect S/N ratios will still amplify the noise passed though to them. So optimizing the gain structure all the way through to minimize noise is important.
Split a traditional two way speaker so that it has two sets of speaker terminals.
Parallel the terminals and you have traditional passive crossover feeding two equal impedance drivers.
For simplicity I am going to assume perfect driver behaviour in their own pass bands and that each has a single pole filter.
Back to the split version.
One pair of terminals feeds a capacitor in series with the treble driver.
The impedance seen by the amplifier starts at around 8ohms in the lower treble and rises slightly as the VC inductive reactance increases with increasing frequency. As frequency falls the series cap becomes dominant and the capacitance reactance increases with falling frequency.
A wideband amplifier driving this treble + high pass filter sees an average impedance that is significantly higher than the nominal impedance of the treble driver. I'll pluck at a guess and suggest the average impedance is 20ohms for an 8ohms treble driver with a 6.8uF seres capacitor.
Considering the other amplifier connected to the other pair of terminals. It is driving a Bass Mid 8ohms driver with a series connected inductor acting as the low pass filter.
As with the above amplifier it too sees an average impedance that is higher than the nominal driver impedance of 8ohms. Again I'll guess and suggest 12ohms.
The two amplifiers, in Bi-amplifying mode, each handle the full bandwidth but each sees an average impedance that is higher than the nominal 8ohms impedance of the speaker. The amplifiers have an easier job driving the load and as with all SS amplifiers they perform better when presented with a higher load impedance.
Parallel the terminals and you have traditional passive crossover feeding two equal impedance drivers.
For simplicity I am going to assume perfect driver behaviour in their own pass bands and that each has a single pole filter.
Back to the split version.
One pair of terminals feeds a capacitor in series with the treble driver.
The impedance seen by the amplifier starts at around 8ohms in the lower treble and rises slightly as the VC inductive reactance increases with increasing frequency. As frequency falls the series cap becomes dominant and the capacitance reactance increases with falling frequency.
A wideband amplifier driving this treble + high pass filter sees an average impedance that is significantly higher than the nominal impedance of the treble driver. I'll pluck at a guess and suggest the average impedance is 20ohms for an 8ohms treble driver with a 6.8uF seres capacitor.
Considering the other amplifier connected to the other pair of terminals. It is driving a Bass Mid 8ohms driver with a series connected inductor acting as the low pass filter.
As with the above amplifier it too sees an average impedance that is higher than the nominal driver impedance of 8ohms. Again I'll guess and suggest 12ohms.
The two amplifiers, in Bi-amplifying mode, each handle the full bandwidth but each sees an average impedance that is higher than the nominal 8ohms impedance of the speaker. The amplifiers have an easier job driving the load and as with all SS amplifiers they perform better when presented with a higher load impedance.
I had to add the (not active) because the stateside of the Atlantic have adopted a quite different definition for Bi-amping. We on the east side of the pond seem to universally see Bi-amping as using two amplifiers to drive the passive crossover (high and low pass filters) that is inside the passive speaker.
OK, got it. Thanks.
I don't know if average impedance matters much (maybe it does) but rather the impedance at a particular frequency. An amp driving a passive low-pass filter will see a very high impedance above the crossover point. Vice Versa the high pass filter. Is that better or worse for the amp? Maybe it depends on the amp. Certainly there will be little current flow outside the pass-band, but will the amp deal well with the very high impedance?
A fullrange passive might have a lumpy impedance, but at least it's somewhat even over the audio range. I really don't know how much of a difference that makes - does anyone here know?
I don't know if average impedance matters much (maybe it does) but rather the impedance at a particular frequency. An amp driving a passive low-pass filter will see a very high impedance above the crossover point. Vice Versa the high pass filter. Is that better or worse for the amp? Maybe it depends on the amp. Certainly there will be little current flow outside the pass-band, but will the amp deal well with the very high impedance?
A fullrange passive might have a lumpy impedance, but at least it's somewhat even over the audio range. I really don't know how much of a difference that makes - does anyone here know?
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.