If you go for diy route for measurement device read this thread https://www.diyaudio.com/community/threads/measurement-mics.406766/ about problems with calibration things and decide what you think is the most good and cheap option for you, one click solution or diy solution!
Thanks @savan for the posts, many interesting things.
Speaker brands that I have never heard of, you have the courage to work with them.
I don't understand in the simulator how should the trials be thought to reach the correct phasing quickly and well .
I assume not by chance, or through many attempts , isn't it ?
And how it looks visually.
(vertical dashed lines ?one for each driver to the color of the speakers graphics?
and should they be grouped close toghether?)
(indyboilermaker69)
"Ideally, you flip one of the drivers (from whatever way you pick) and there is a very deep (as deep as possible, theoretically to 0dB) and very symmetrical null in the response…. This will mean that the drivers with their crossovers are as in phase as you can get..."
Speaker brands that I have never heard of, you have the courage to work with them.
I don't understand in the simulator how should the trials be thought to reach the correct phasing quickly and well .
I assume not by chance, or through many attempts , isn't it ?
And how it looks visually.
(vertical dashed lines ?one for each driver to the color of the speakers graphics?
and should they be grouped close toghether?)
(indyboilermaker69)
"Ideally, you flip one of the drivers (from whatever way you pick) and there is a very deep (as deep as possible, theoretically to 0dB) and very symmetrical null in the response…. This will mean that the drivers with their crossovers are as in phase as you can get..."
This is not helpful. Drivers in phase is a good starting point for a single axis crossover only because the information you have is limited. Besides, getting phase precisely to the degree has no benefit.
Phase alignment is important, but secondary to response for me. And sometimes the phase is not perfect when the frequency response is, But as long as it's very close it's not really hearable, not even for trained ears. But when it's not at least close, it's very obvious for me now.
And no speaker has a flat phase, it's inherent to drivers to have a phase respone that goes a bit up and down, and that is not an issue, it's the abrupt change of phase (between two drivers) that causes issues with smearing sound and phase cancellation (partially mostly) which both distort the sound.
That is also why i love single driver fullrange drivers, the phase is consistant over the whole spectrum. With multiway it's a bit more difficult to get that right.
And no speaker has a flat phase, it's inherent to drivers to have a phase respone that goes a bit up and down, and that is not an issue, it's the abrupt change of phase (between two drivers) that causes issues with smearing sound and phase cancellation (partially mostly) which both distort the sound.
That is also why i love single driver fullrange drivers, the phase is consistant over the whole spectrum. With multiway it's a bit more difficult to get that right.
Victorian, since you have good combination of the enclosures with one speaker per each box, from first post, thats very good choice for your first sucesfull speaker! Why? because you can move each box back and forward! Thats now very good choice for putting all things together all that you read in this thread. That way you can fine tune everything, and also your crossower is now easier to build! Moving each box to their right position you can fine tune things and also you have a way now to make seccond order crossover easily, even fist order if things allow! But you must have measuring device! Without that there is no go. : )
It is very difficult to get the entire frequency range with low distortion from one speaker. It should be taken into account that the speaker distortion in the low-frequency range is transmitted to the entire frequency range it reproduces.
To get an even phase in three bands, you need to use active band division. At the moment, the easiest way is to use a computer with a multi-channel sound card or DSP. If there is a DSP, then getting an even phase from three bands is not a problem, which in turn allows you to use three speakers and significantly reduce distortion across the entire frequency range.
But no matter how you look at it, in my opinion, setting up three bands without a microphone is a very useless exercise.
Uriy-ch, this is getting boring. I was told already a million times that. I know that, but fullrange also have qualities that multiway don't have. Speakers is a matter of choices... btw DSP also got disadvantages, that are a dealbreaker for many. It's a matter of choices what is more important for you, not absolute laws.
But what i wanted to say, is that you can get the phase consistancy of a fullrange driver setup in a multiway if you phase align the drivers. It will never be as perfect as a single driver fullrange (for THAT aspect), but you can get very close.You can do that with dsp and with passive (but passive will be much harder and expensive to do in most cases).
But what i wanted to say, is that you can get the phase consistancy of a fullrange driver setup in a multiway if you phase align the drivers. It will never be as perfect as a single driver fullrange (for THAT aspect), but you can get very close.You can do that with dsp and with passive (but passive will be much harder and expensive to do in most cases).
In my active three-way speaker system I did not align the phase of the drivers, I simply used a crossover that does not twist the phase and used the frequency range of the speakers where they do not twist the phase, and this was enough to get the results shown below for the measured phase.
I just wanted to say that for digital systems there are no problems with phase if you use a crossover that does not twist the phase. This allows you to make a three-band speaker system with a smooth phase like a wide-range speaker.
@uriy-ch Even having dsp you need measurement device! Each speaker have their owm phase curve, unique! If you not align phase from all speaker e.g. bass with mid, mid vith tweeter... thing will be not in phase! Just go and measure your speakers, each driver, combine all phase graphs from your measurements and you will see that its not phase aligned! : ) What you can do with dsp is similar to what you can do with pasive crossower, booth reguire measurements! Without right measurements everything is wrong, even might be soo wrong. And there is no trials and errors method by hearing for example, its simply not possible to make crossower or to dsp thing the right way without real measurement, waste of time and money! Dsp is the easy way, but offcourse without measurement device its just waste of time, it might fool you, for example speaker will sound one day very nice and seccond day will sound terible, the similirar to those with pasive crossower. : ) The good speaker is a speaker that you can listen to all day from the morning to the midnight without his sound bothering you in any way, that speaker you can call good speaker.
Last edited:
Maybe of some help without measurements, just one of those freebie signal generator programs.
No. A passive crossover is limited in its capabilities, any passive filter of the second or higher order twists the phase at the cutoff frequency. And an active digital FIR filter or active subtractive filter, can suppress the signal amplitude without affecting the phase of the signal at the cutoff frequency, this property allows you to match three bands without worrying about the phase at the speaker crossover frequency. But one way or another, this does not relieve us of the need to use a microphone. Without a microphone, adjusting three bands is a lottery game. I mentioned the properties of active band division only in the sense that, if desired, it is not particularly difficult to get an even phase in three bands like in a wideband speaker if you apply digital signal processing.
In my opinion, adjusting three bands by ear is a very unproductive activity for any type of crossover, whether active or passive. With a microphone, you can get a better result on three bands than without a microphone.
You still need to phase align the drivers in the dsp, but there you use a delay of a few ms, based on physical location of the speaker.
https://purifi-audio.com/blog/tech-notes-1/time-phase-alignment-acoustic-center-lobing-etc-14
But with measuring the drivers in the box, from one point, you can calculate that in the filters of your dsp also. The way you make the crossover does not change the factors in play, it's just how you use them that changes. The advantage of a dsp is that you have more tools than with passive crossovers to deal with those factors, and with less side effects that have to be taken in account.
For sure!
You just left out the phase changes, yes, but you also left out the phase alignment of the spokes. Bass, mid, tweeter... etc, all have different phase curves, and if you just suppressed the amplitudes of the signal without affecting the phase, that's ok, ok, but you still haven't aligned all the speaker phases, it remains the same mismatch as before the correction, because you didn't touch the phases! You have done nothing but change the amplitude on each speaker. : )
You just left out the phase changes, yes, but you also left out the phase alignment of the spokes. Bass, mid, tweeter... etc, all have different phase curves, and if you just suppressed the amplitudes of the signal without affecting the phase, that's ok, ok, but you still haven't aligned all the speaker phases, it remains the same mismatch as before the correction, because you didn't touch the phases! You have done nothing but change the amplitude on each speaker. : )
Last edited:
It is impossible for all phases to coincide because they do not have the same phase, bass, mid and tweeter have different phases, all 3 diferent! Its realy an coincidence when you get all 3 speakers which is capable for first order filrer and to become fully phase aligned in the same time, you must be realy soo lucky to find that speakers on market capable for that thing. : )
For example https://www.humblehomemadehifi.com/download/Humble Homemade Hifi_Plutone.pdf one close to first order, nice speaker.
Or something like this https://www.diyaudio.com/community/threads/direct-digital-power-dac-ddpd.395974/ regardless of the fact that I didn't have a good measurement result on it for some currently unknown reason, but I definitely haven't heard a better one dynamics literally gave me goosebumps when I heard it for the first time! I still have plan for next with dsp, bought adau1462 few months ago and waiting to finish some project first before I start on next ddpd.
For example https://www.humblehomemadehifi.com/download/Humble Homemade Hifi_Plutone.pdf one close to first order, nice speaker.
Or something like this https://www.diyaudio.com/community/threads/direct-digital-power-dac-ddpd.395974/ regardless of the fact that I didn't have a good measurement result on it for some currently unknown reason, but I definitely haven't heard a better one dynamics literally gave me goosebumps when I heard it for the first time! I still have plan for next with dsp, bought adau1462 few months ago and waiting to finish some project first before I start on next ddpd.
Last edited:
You don't need 1st order for phase coherence,. You never have that 100% over the full bandwith with a multiway. But if the phase is coherent at the filter frequency zones, it's unhearable. And that is much easier with a higher order crossover than with a 1st order actually, but you need to know what you do, and design the filter to measurements, not in a fixed form. Most crossovers that i design are actually mixed order, often combingin 2nd order with higher order filters, just to align the phase, and correcting it further with resistors on the parallel leg of the filter to finetune it. and with DSP you can even do more with FIR filters...
Yes I know, it can't be the same for sure, all speakers have their own phase curve unique. Me too doing mixed order to align all phases. : )
Of course, for "complete happiness" you need to align the acoustic centers of the three speakers using a delay, in my understanding, introducing a delay into the channel for synchronous operation of the speakers is not a correction of the phase of the speakers, since the delay does not twist the phase of the speaker itself, it simply allows you to align the acoustic centers of the speakers, which will allow them to work synchronously.
There is always a lot of confusion with this phase, since one word has to describe different processes. For myself, I separated the delay from the concept of phase, maybe this is not correct, but it helps me get less confused when I adjust the speakers. One way or another, if desired, DSP allows you to get a phase in three bands such as one broadband speaker emits. This is what I wanted to tell you about, and everything else is just the subtleties of adjusting the speakers to achieve this result.
How you know that your acoustic center have the right phase angle? I know that when phase is not aligned right but amplitude is perfectly flat e.g. in +-0.5db, thats very hard to get the right listening position, very hard on multiway speakers, and there might help only measurement or trials and errors which might take forever time.
Edit: I think I understand what you mean. Delay might be similar to what can be done when having moveable boxes of all 3 speakers like from first post. Than I can accept solution. But without measurement its not possible to know.
Edit: I think I understand what you mean. Delay might be similar to what can be done when having moveable boxes of all 3 speakers like from first post. Than I can accept solution. But without measurement its not possible to know.
Last edited: