Yes, that's still a crossover. You get the best results when the low frequency driver is well behaved at the upper end and got a smooth drop off and got no or low break-up resonances, same for the tweeter. Same goes for 6dB crossovers need easy to handle drivers too unless you use a suction/absorption filters for resonances and flattening the impedance peaks.
You need an electrical part on which the voltage drop can happen, so on just drivers parallel these filters won't work.
If you want to build a speaker without a crossover, use a small fullrange driver for the upper part of the frequency range, with a sealed enclosure to limit the excursion. That could work well up to medium volume. A fullrange driver for such a speaker is the Visaton FRS 5 X, sounds very good, got a good dispersion and on top of it is really cheap (~10-15 bucks).
You need an electrical part on which the voltage drop can happen, so on just drivers parallel these filters won't work.
If you want to build a speaker without a crossover, use a small fullrange driver for the upper part of the frequency range, with a sealed enclosure to limit the excursion. That could work well up to medium volume. A fullrange driver for such a speaker is the Visaton FRS 5 X, sounds very good, got a good dispersion and on top of it is really cheap (~10-15 bucks).
There isn't a fixed definition, but not necessarily. Usually all that's meant is that there is an electrical circuit of some description, be it passive or active, that modifies the transfer functions of one or more drive units in order to achieve a desired set of results in transitioning from one drive unit to another.
Last edited:
Parts will be following the applicable laws and the results will be in accordance with it. Of course, audiophiles might think of you as being a certain type, you know, but they are well behaved beings after all.
Naturally, you may be testing us and/or our patience.
I can't imagine a speaker play well, with no cross-over of some sort 😕
And crossing from a 8" or 10" to a 1" dome... with no waveguide.... why?
Indeed, a speaker can sound less "open" with a poorly made filter/cross-over. If EQ makes something sound bad/worse... maybe you are doing something that is out of bounds with the laws of physics 😉
And crossing from a 8" or 10" to a 1" dome... with no waveguide.... why?
Indeed, a speaker can sound less "open" with a poorly made filter/cross-over. If EQ makes something sound bad/worse... maybe you are doing something that is out of bounds with the laws of physics 😉
Thanks adason and Scottmoose and all, for your helpful replies. This really speeds up the learning process. Please, no worries, it's a hobby and hope we can all enjoy together.
Yes, the Seas kits looks about right except -6dB is 60Hz?. I guess their crossover compensates for the 89 / 95 dB difference in sensitivity. It reminds me of a smaller Hartley-Luth 2 way speakers which had an 18" woofer and tweeter.
I only listen at 75dB in a 15 x 15 room but want bass good to at least 40Hz
Yes, the Seas kits looks about right except -6dB is 60Hz?. I guess their crossover compensates for the 89 / 95 dB difference in sensitivity. It reminds me of a smaller Hartley-Luth 2 way speakers which had an 18" woofer and tweeter.
I only listen at 75dB in a 15 x 15 room but want bass good to at least 40Hz
digitalthor, EQ can flatten out the response curve, but I believe the Behringer EQ caused phase shifts which made the speakers unlistenable. I used a Denon UD-M31 mini system and the treble and bass boost caused no audible difference.
Anechoic. Like the Dynaco A25, it's designed for use near room boundaries so has a well-damped LF alignment (resistively vented / 'aperiodic') to prevent excess; in practice you should get usable output to the low '40s. You can get more inherently with a (significantly) larger box.
Then maybe a small set of neutral speakers, would be a good choice for you.
It does not have to be expensive. Some inspiration here:
Heissmann Acoustics | Kits | Speaker developments
I use a similar design, with a DXT, Dayton RS125, two SB23NRX and 4 subs. Extremely transparent and relaxed sound at both low and rather loud levels.
The trick is to use drivers combined with a cabinet design, which have a FR that is easy to fix with EQ. For low listening levels, I see no reason to chose the bigger drivers, that you mention. I tried the Behringer.... and for sure it will kill all sound qualities to obtain a flat FR - if you are not careful, since it's auto-feature is brutal to sound quality, when seeking the flattest FR. Do remember that most auto processes do not know any better than what you measure.... so measure wisely and take your time to interpret the data.... or else you are easily chasing a black cats tail - which is actually never there to begin with 😉
It seems you are trying to avoid phase shifts at all cost. People got different perception and preferences but I think you are trying way too hard to keep one single thing 'perfect' but take on a whole lot drawbacks and problems with it.
"No crossover equals no phase problems" does not work! Two different drivers working at the same frequency range will shift in phase to each other even without any crossover. Moreover, you'll get interferences over a broad bandwidth because of that and that's exactly what you want so badly to avoid with 'no crossover'. You are sacrifycing low distortion, dynamics, linearity and resolution over just one thing you probably weighting way too much. With a crossover you keep the range both are working at the same time controlled and (if you develop the crossover properly) will reduce the phase shift or even eliminates it but that's what you have to work on when designing a speaker respective crossover.
Everything is a compromise in audio reproduction, you have to find what's ideal for you and I don't want to force you to use a crossover or what type you have to use. But about 80-90% of the sound of a speaker does not depend on the drivers but instead on the crossover, it's the most important part! I promise you, with your crossoverless project you will get only a fraction of what the drivers are capable of with a well developed crossover, but most likely you'll experience a shipwreck with that plan.
My suggestion: Take a simulation program (ie. Boxsim), play around with drivers, switch drivers around (no matter if it's the drivers you actually want to use), try parts/crossover and see what the phase, frequency response/linearity, impedance, dispersion etc does, look at what it does before deciding to invest money and time in something that won't work.
When a speaker has a wiggle in its frequency response, it typically has a matching wiggle in its phase response. This is called minimum phase behaviour, a mathematical term stating that the phase response is linked to the frequency response. There can also be "excess phase", which is phase shift that isn't due to frequency response problems.
Now, the Behringer DCX2496 or DEQ2496 are both minimum-phase EQ units. That means there's a phase shift applied when frequency response corrections are applied. Here's the neat bit: the phase shift brought in by EQing often reverses the speaker's original minimum-phase wiggles.
ie, by EQing the frequency response flat, you're often improving the phase response.
However, this isn't always the case. Some issues require FIR processing to fix, which is where you can alter the phase and frequency response independently of each other. The price there is latency. In my PA system (used for live sound, so low latency is important), I have a flat frequency response from 40Hz-18kHz, and a flat phase response from 600Hz upwards. The system latency is 2ms, which I can live with.
If there's no audible difference, they're broken. EQing should have a very obvious effect.
Chris
Correcting frequency response often improves phase response so long as it's the speaker that's being corrected without room influence.
Why are the tweeter and woofer so far apart on the Seas A26? Or most 2-wyas for that matter. I thought they should be as close as possible. Especially due to the higher frequency of the x-over.
I agree x-over is the most important part. Who knows which speaker out there has one that is properly designed? I've heard 20K speaker that do not get it right.
There are many ways to look at a problem. Passive crossover can get the best out of the drivers but add their own issues. Active done right is better. There are convincing arguments for infinite slope and also for 1st order. Sound dispersion and room effects. Theoretically, I like 1st order the best. But first order have more room interaction problems. Who can say ACTUAL, in-room of a first order will sound okay after adding in reflections? Unless you anechoic the room.
If only listening at 70-75dB, why suffer performance issues caused by designs that rate speakers up to 95 dB and above?
With hundreds of 2-ways out there, how do you know which one is best? If you rely on your ears alone, you can go crazy auditioning. Makes me want to just give up.
The Denon eq'd but had no audible phase shifts or changes to sound quality. I had someone else in chat confirm they had the same experience.
The question with the A26, for example, is which sounds better at 500-2000 Hz? A 4" midrange driver or a 10" woofer or a 1" excel tweeter?
Thanks so much for the help and enlightened insights.
There are many ways to look at a problem. Passive crossover can get the best out of the drivers but add their own issues. Active done right is better. There are convincing arguments for infinite slope and also for 1st order. Sound dispersion and room effects. Theoretically, I like 1st order the best. But first order have more room interaction problems. Who can say ACTUAL, in-room of a first order will sound okay after adding in reflections? Unless you anechoic the room.
If only listening at 70-75dB, why suffer performance issues caused by designs that rate speakers up to 95 dB and above?
With hundreds of 2-ways out there, how do you know which one is best? If you rely on your ears alone, you can go crazy auditioning. Makes me want to just give up.
The Denon eq'd but had no audible phase shifts or changes to sound quality. I had someone else in chat confirm they had the same experience.
The question with the A26, for example, is which sounds better at 500-2000 Hz? A 4" midrange driver or a 10" woofer or a 1" excel tweeter?
Thanks so much for the help and enlightened insights.
You could get them about ~3,5-4cm closer if you mount the tweeter overlapping to the bass-mid driver and then route the bass, which takes off the overlap of the tweeter front plate. But that's indeed the closest on the baffle. I would not recommend that but it's helpful when using a waveguide.
You could mount it in coaxial position in front of the bass but that brings a lot other problems with it.
The price was never ever a measurement for sound quality. Most 20k speakers got just ~1k in drivers in it anyway.
If you mean by that, the room interaction is because of the dispersion pattern, then that's right. If not then there's a big misunderstanding either on your side or on mine.
I think your problem is a completely different thing. The dispersion pattern can be controlled by a wave guide and reflections can be countered by acoustic treatment and/or placement of the speakers in the room. If you like 1st order best then it's likely you want the homogeneity. That can be achieved by different methods. The 1st order xo got a huge overlap from the bass to the tweeter, so the sound is 'transitioning' from one driver to the other by simply mixing them together. For most drivers that's theoretically a solution but resonances, distortion and decay ridges often make it impossible. Unless you've got extremely tame, unproblematic drivers it's going to be a problem. Your plan for a no-xo speaker is even more problematic. Looking at the frequency response, you actually need not only some kind of equalizing but the impedance measurement on the datasheet hint the H1471 got a big resonance at 700Hz and the cone break up is at ~500Hz, which might suggest it's probably a driver better suited for a 3-way but will not shine on a no-xo speaker.
Because the distortion of the tweeter at or below its resonance will already start at that level? And besides that, the interference will happen because of the phase differences not because of the level? Others and also I've already told you. The results won't change just because you asked again. 😉
There isn't a "best" speaker. It's always a compromise. And it depends on your room, listening preference and, after all, your taste.
That does not 'prove' the absence of phase shifts, it likely had just a soft slope on the tone filters or they were active at frequencies the human ear is not very sensitive to phase differences (the human ear can't detect absolute phase anyway, only phase differences).
The 4" midrange. Or a similar sized fullrange driver. Then the woofer, then the tweeter. In that order.
Attachments
Thanks for the wealth of information.
"Looking at the frequency response, you actually need not only some kind of equalizing but the impedance measurement on the datasheet hint the H1471 got a big resonance at 700Hz and the cone break up is at ~500Hz, which might suggest it's probably a driver better suited for a 3-way but will not shine on a no-xo speaker."
So you are saying the H1471 should not be used below 500Hz because of cone breakup? I am not seeing it on the FR
"Looking at the frequency response, you actually need not only some kind of equalizing but the impedance measurement on the datasheet hint the H1471 got a big resonance at 700Hz and the cone break up is at ~500Hz, which might suggest it's probably a driver better suited for a 3-way but will not shine on a no-xo speaker."
So you are saying the H1471 should not be used below 500Hz because of cone breakup? I am not seeing it on the FR
No, but it requires care on the crossover to fix it. Depending on the width of the speaker, you need to compensate the baffle step too. Before you ask, the baffle step is an effect caused by the reflection on the edges of the baffle and its frequency depends on the width (on the vertical (top/bottom edge) too but that can often be neglected).
The cone breakup starts at 400Hz, where the angle measurements jump up from each other (7-8dB!) and the cone doesn't move as a piston anymore but instead breaks up in differently moving areas of the cone. It's not perfectly clear to tell from the measurements but a 8" driver the normal beaming starts further up, usually slowly around 1-1,5 and for most projects with a mid-woofer they aim at ~2k where the dispersion is still wide enough to cross over to the tweeter. It could be a resonance too but if you want to use this driver, you have to measure it yourself anyway because of the problematic resonance at ~700Hz (clearly visible at the impedance) which likely poses a problem you have to deal with. At this price I'd expect better performance in the important midrange, I'd use that driver only for a 3-way or FAST speaker.
A quick search shows a thread here in the forum. I haven't read it completely because I'm lazy 😛 and I'm confident it's explained reasonably. 😉
You don't need to calculate the baffle step yourself, the simulation software can cover it usually very well.
You can influence the behaviour by the baffle geometry/width, asymmetrical placement of the drivers on the baffle, by chamfered or rounded edges, by waveguides (for hifi not reasonable for a woofer though because of its size) or absorption (ie foam on the baffle).
Last edited:
Would a passive x-over in a loudspeaker also create phase shifts same as an equalizer?