I think I'd hear it if I got the balance wrong; I have a measurement mic, but I haven't felt the need to use it yet! I do wonder if I've stumbled upon the ideal speaker for conversion to active, as it just 'works' and, with the relative levels set properly, I can't hear any difference with changing crossover frequency or slope over quite a wide range, which must be a good sign.
As an experiment, I just 'activated' a pair of AR bookshelf speakers with (rotted foam) woofers replaced with some random woofers of the right size, and they sound atrocious whatever I do. It will be interesting to see what can be done with a measurement mic and inverse-ish impulse response to correct them - no anechoic chamber in our house, though, I'm afraid.
Generally I do not equate 'more bass' with 'better bass', as a rough rule of thumb I equate 'tighter bass' with 'better bass' but if one does not listen carefully this might be mistaken for too little or too lean bass.
One of the strong points of line level XO
I don't care how well the active XO is implemented, the statement i made is still true.
dave
When I first purchased my dcx, It was very underwhelming. The grain was obviously there. I pulled my CJ tube amp out of storage and hooked it up to the tweeters. The grain totally disappeared. The point I'm trying to make is that maybe hooking up the high frequency section of a speaker directly to an inexpensive solid state amp, totally exposes the amp's weaknesses(high frequency grain), and inserting a good passive will create a more forgiving environment for the high frequencies of that given amplifier. So maybe It's actually the passive that's subjectively "beautifying" the high frequencies(With "great sounding" caps?) instead of the "inexpensive" active that's "ruining" it..
Last edited:
Yes, but to what degree does the active XO have to be designed for you to consider it a non issue?
I'm still working to build one that is transparent. But usually i can design the speaker to live within the realities of a PLLXO
Opamps & Sallen-Key have failed to get out of the way.
dave
Me. Usually digital off my media server. Slowly all my vinyl is coming over (at 24/192)
dave
Given a digital system it's very easy to design an active crossover that in no way compromises the systems linearity. Typically in a digital system the reproduction chain would look something like this.
Digital source > DAC> Power amp> Passive crossover> Drive units.
An active version would look something like this.
Digital source > DSP > DAC> Power amp > Drive units.
I am of course assuming here that the power amps and DAC s in both are the same, but just more in quantity in the second. If a system is put together correctly in this way then you've not introduced anything detrimental to the signal chain that wasn't already in the first, except in the second you've removed the passive xover components.
This is my idea of comparing active with passive. If the analogue active crossover introduces a level of distortion or noise to the signal chain that could be considered significant, then imo it is poorly designed.
Digital source > DAC> Power amp> Passive crossover> Drive units.
An active version would look something like this.
Digital source > DSP > DAC> Power amp > Drive units.
I am of course assuming here that the power amps and DAC s in both are the same, but just more in quantity in the second. If a system is put together correctly in this way then you've not introduced anything detrimental to the signal chain that wasn't already in the first, except in the second you've removed the passive xover components.
This is my idea of comparing active with passive. If the analogue active crossover introduces a level of distortion or noise to the signal chain that could be considered significant, then imo it is poorly designed.
In my comparisons (FWIW) it's been tube or discrete transistor or opamp or DSP crossovers vs passive. DSP is by far the fastest and easiest to get you where you need to go, if there is much complexity at all. But it would be hard for it to beat a 3 element passive crossover. 
Horses for courses.
Horses for courses.
For an ultimate system this is what i'd eventually like to do. I have notes/sketches from 1988 with this config. I'm waiting for it to get cheap enuff at high enuff quality. Even a DEQX still only does 24/96. Note, my whole system cost less than a DEQX (at least in cash outlay).
dave
I have made a PC-based XO using studio gear (RME and DAD) with a plain vanilla tower PC. Software is based on a MIDI router called CONSOLE and a mastering-quality EQ VST-plugin.
With the proper hardware you can do any sample rate and resolution you want. I can do DXD (384/24), but why bother when 99,99% of my source material is 44/16 ...
With the proper hardware you can do any sample rate and resolution you want. I can do DXD (384/24), but why bother when 99,99% of my source material is 44/16 ...
I don't do windows, PureVinyl has (basic) XO software built-in, supports AU plug-ins, i have 4 outs from my DAC at 24/192, but i'm afraid my (free) media server will bog down if i try to make it do much work.At some point i wll try it thou. Since i mostly use no XOs, I'm actually more interested in a convolution filter to flatten FR & phase response.
dave
Passive XOs are fine the way people are describing them here, as perfect components. But they are not inductors have series resistance and this is one of the key points. I have read through this thread and it seems no one understands what the effect of damping has on a speaker.
When a speaker is moving it's generating a back emf, ths emf needs to be absorbed for the driver to follow the signal front he amplifier. It is absorbed through electromagnetic braking as the amplifier will effectively short the loudspeaker to the back emf but when you start putting lossy inductors in series with it this effect stops and the driver is poorly controlled.
Have you ever seen the trick of running a magnet through a copper tube and it takes a long time due to the eddy currents braking the magnet? Ths is the same principle.
Inductors are ultimately flawed. One due to the series resistance and two because you can use iron, ferrite cored inductors to lower the series resistance (not eliminate it) but then you start using non linear components which then in turn cause distortion especially at high levels.
So you can go through all of this or not have anything there at all except a couple of op amps with tiny levels of distortion.
Also I don't understand why people talk about crossovers with poor implementation, in this case it doesn't matter with crossover it is!
I have made up my mind I don't believe things for no reason I started this thread for the benefit of the forum users not for people to tell me I'm wrong!
When a speaker is moving it's generating a back emf, ths emf needs to be absorbed for the driver to follow the signal front he amplifier. It is absorbed through electromagnetic braking as the amplifier will effectively short the loudspeaker to the back emf but when you start putting lossy inductors in series with it this effect stops and the driver is poorly controlled.
Have you ever seen the trick of running a magnet through a copper tube and it takes a long time due to the eddy currents braking the magnet? Ths is the same principle.
Inductors are ultimately flawed. One due to the series resistance and two because you can use iron, ferrite cored inductors to lower the series resistance (not eliminate it) but then you start using non linear components which then in turn cause distortion especially at high levels.
So you can go through all of this or not have anything there at all except a couple of op amps with tiny levels of distortion.
Also I don't understand why people talk about crossovers with poor implementation, in this case it doesn't matter with crossover it is!
I have made up my mind I don't believe things for no reason I started this thread for the benefit of the forum users not for people to tell me I'm wrong!
Last edited:
Having seen what you have already implied about damping, you don't either. High damping factor is not a panacea. There are lots of situations where too low an output impedance can be too low for the system that follows it.
Matter of fact it can increase distortion, particularily with a feedback amplifier as the back EMF feeds into the front end of the amplifier creating some insidious distortions. If you had a current amp (very low damping) the back EMF (and BC heating, and more) effectivey dissappear. But you need to have a speaker system that is appropriately damped (or EQed). Most speakers aren't. Everything is a balance of compromises.
That brings up an advantage of an active system ... you can use an appropriate voltage amplifier to drive a more typical bass driver & a current amp on the top.
dave
and limited to other 'certain realities'
'we' often recommend active crossover to 'first timers' who want to design their own dream speaker
many will not have the patience or knowledge to make a passive work
and no, neither do I trust the commercial manufacturers to get it right either
and considering the above 'statement', I may have to admit that quite many speakers might sound better converted to active
but thats not the same as saying active sounds better
its just a simple matter of passive being difficult
but even so, despite that active per definition may not sound better as such, the conclusion could still be that it will
Any competent passive crossover has inductor DCR and capacitor ESR built into the design. They're integral to the performance, not disadvantages. This is why I shake my head when the component-swappers start slapping in fashion parts of supposed "higher quality" where the performance will actually degrade.
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.