Yes but you forget waveguide behavior in the equation: what happen to it if you lower the 'optimal' cross over point? Won't it loose ability to maintain directivity behavior? Is there other parameters at work which will be detrimental about soundquality? Like phase shift and acoustical cut off (from the horn design)?
I should have given a little more detail with my question. I'm only talking about the difference between the 1300Hz crossover point and a 1000Hz crossover point I would like to use. If I even decide to use a wave guide, which is looking less likely now.
I'm looking at the SEOS-12 polar response map and directivity looks great down to 1000Hz and OK down to almost 700Hz. The minimum recommended crossover point is listed as 950Hz. So my 1000Hz
But LineSource is suggesting 1300Hz as the crossover point, based on directivity of 12" woofer at that frequency matching up well with that of the wave guide.
So, what I'm saying is think about what we actually hear with the two different crossover points.
If we cross our 12" woofer at 1300Hz, off axis response begins to roll off a little below 1000Hz until the directivity pattern matches that of our wave guide somewhere around 1300Hz. Our on axis response begins to roll off a little below 1300Hz until the wave guide output is dominating.
What we hear at our listening position is off axis response rolling off between 1000Hz and 1300Hz and hopefully we don't hear any change in on axis response.
If we cross our 12" woofer over at 1000Hz, off axis response begins to roll off a little below 1000Hz, just like the on axis. But off axis will roll off a little faster than on axis since both beaming and the crossover filter are in play there.
What we hear at our listening position is off axis response rolling off between 1000Hz and maybe 1150Hz or so and hopefully we don't hear any change in on axis response.
I do understand that not all 12" woofers have the same diameter cone so actual frequencies for beaming will vary, and different crossover slopes will also change the roll off rates. But I tried to keep the example simple.
So why am I so worried about this? Because that 300Hz difference in mid bass driver low pass frequency makes a huge difference when selecting the 12" woofer to use.
We're already pushing our 12" woofer to the limits of the usable frequency range. LineSource described how trying to go below 100Hz on the bottom end, with 1300Hz on the top, results in audible IMD. So, regardless of manufacturer specs, we aren't going to find a single 12" driver that covers my desired 70Hz to the LineSource recommended 1300Hz.
But using our 12" woofer from 100Hz to 1300Hz, throws off well established guidelines for subwoofer integration, forcing us to do things like build 18" woofers into our mains. There's nothing wrong with that if you have the room for it and WAF is not a concern. But I already have a couple 15" subs that I like. If I replace them, I would like it to be with four 12" subs in smaller cabinets that I can more easily position around the room.
The driver choices for the example speakers LineSource has provided all seem to start with that 1300Hz crossover point, which I'm not convinced is audible better than a 1000Hz crossover point.
If we take the bandpass range we expect from our 12" woofer and simply shift the whole thing down, we should still be able to minimize IMD while using separate subs at recommended crossover points. So we give up 300Hz of upper frequency extension to gain 30Hz of low frequency extension.
If that shift of the bandpass frequency range doesn't audibly change our transition to a wave guide then I think we have a much more flexible mid bass module. It can integrate with sub woofers on the bottom end at a frequency that allows re-positioning of single or multiple subs an it can integrate with a SEOS-12 on the top end, if we want to do that. We also get a wide mid bass bandwidth to play with so our choice of mid range drivers is wide open. It also allows us to use a 12" woofer with more XMax so it can take some of that PEQ beating. Personally, I cut peaks much more than boosting valleys so I don't think I cause very much beating.
Anyway, all that above is my current thinking on this. 😀
-Chris
But using our 12" woofer from 100Hz to 1300Hz, throws off well established guidelines for subwoofer integration, forcing us to do things like build 18" woofers into our mains
Honestly i don't know from where do you come with that? Could you explain a little... i'm perplex.
The driver choices for the example speakers LineSource has provided all seem to start with that 1300Hz crossover point, which I'm not convinced is audible better than a 1000Hz crossover point.
I repeat what i said in previous post: you focus on the wrong way to see the problem: this is not the woofer which exhibit issues at 1000hz but the waveguide! (in case of seos12) 😉
In case you will have spare time another reading! 😀 : This is the first and second part of the same article:
https://www.grc.com/acoustics/an-introduction-to-horn-theory.pdf
http://audioxpress.com/files/kolbrek2885.pdf
Last edited:
Hold the woofers in there partner.
If I remember correctly I kind of gave some advice in an optimised bass hinge point.
There is sound musical reasoning and physiological reasoning behind that recommendation.
I thnk you listen to music right?
The greatest bass power in music is below 150 hertz. Kick drum is centered around 120 hertz. And what most people think of as being powerful bass is not much lower than 64 hertz with a great deal of louder higher frequency harmonics.
Why the magic 70 hertz for your largest movers of air?
Just because something get's mentioned a million times on the internet forums does not mean that it is true.
Think this stuff through.
Mark, please remember that this is far from the first speaker system I have put together. It's just the first one, inside the house, where I'm designing and building the main stereo channels from top to bottom. I purchased my first powered sub around 1998 or maybe 99. It was a little Energy S8, which is still downstairs and being used daily, but that isn't the point.
I've built and integrated multiple subwoofers into my home and mobile speaker systems.
70Hz is the highest I like to run a sub unless it is positioned in front of me. If the crossover is higher than that and the sub is to the left, right or rear, it starts to bother me as I can tell the sub bass is not coming from up front. I use 70Hz in my car and the sub literally disappears if I set the level right to integrate with my 8" mid bass. It sounds like all that bass is coming from up front. But turn off the sub and it's very obvious that it was making a substantial contribution.
Don't be the fictional lemming that just goes along with the crowd. Figure this stuff out for yourself, or ask people that can make a passing attempt to make some sense!
At times we are even right!
The last thing I want to do is offend you, or anyone else contributing to this thread. I probably know 10 percent of what you and some others do and couldn't begin to spec out and build my own drivers. But suggesting that I chose 70Hz as a subwoofer crossover point without understanding why I made the choice is, honestly, a little offensive to me.
-Chris
Honestly i don't know from where do you come with that? Could you explain a little... i'm perplex.
I thought LineSource explained it pretty well. With a 1300Hz low pass crossover point, you can't set the high pass crossover point lower than 100Hz without audible IMD.
The 100Hz crossover point is too high to allow remote location of a sub woofer without you being aware that it isn't in front of you. The 18" driver built into the mains addresses that problem by keeping the production of bass frequencies below 100Hz in front of the listener.
I repeat what i said in previous post: you focus on the wrong way to see the problem: this is not the woofer which exhibit issues at 1000hz but the waveguide! (in case of seos12) 😉
No, I was talking about the wave guide. What issues are there at 1000Hz with the SEOS-12? What I was saying is that their polar map shows directivity control to be excellent down to 1000Hz and not too bad down to almost 700Hz. And they recommend a 950Hz minimum crossover.
So it all seems to come down to whether or not a slightly different roll off rate, off axis, is audible or not.
-Chris
But suggesting that I chose 70Hz as a subwoofer crossover point without understanding why I made the choice is, honestly, a little offensive to me.
Well maybe taking offense is useful.
Useful if you wonder why?
Why would I say it.
And since this is not your first rodeo you may have an idea as to why it first of all can work, and work very well. And secondly why it will do what LineSource has been saying. Clean up your midrange.
You talk about using DSP. So think about how it can be harnessed to do something that is not done very widely in commercial products.
A steep slope crossover, and the proper driver placement within an enclosure will not allow you to tell where the source of the low bass is.
But you have to think past 2,3,4 order crossovers. Try 8th order and you will be able to do what I am suggesting.
At any rate.
I have no vested interest in this.
And this is looking like an attempt at a design redo from the 70's.
I grew up listening to Altec Lansing 604's, and Klipsch horn bass sections and have listen to Urei and JBL monitors at length.
They have their sound. And you may enjoy that type of sound.
I have zero interest in spending time pursuing a large driver format 2 way system in any way shape or form.
I know their sonic imprint very well.
And I have never thought they were musical in the least.
So have at er Chris and I sincerely hope you find and produce what you are seeking.
Then why didn't you just say that an extremely steep crossover slope might work to move subs around, without noticing their position, even when crossed over at a higher than normally accepted frequency?
I'm just trying to build a couple mid bass modules that give me flexibility for integrating different possible solutions for mid range and treble frequencies later. I couldn't care less if my box with a 12" woofer ends up being a design from 1942 or something cutting edge. Especially in this first try at it.
Until I get these mid bass modules built, I can't even experiment with steep crossover slopes at alternative frequencies, to see how it sounds to me.
It's doubtful that I will end up handing off from a 12" woofer to a SEOS-12 wave guide but it would be nice to have the option, as many people apparently think it sounds quite good. As much as I like my little Scanspeak 10f drivers for midrange down to about 450Hz, I somewhat doubt I would be satisfied with the compression drivers and wave guide.
Oh well, looks like this whole thing is going downhill fast. I really do appreciate the help and advice from all you guys up to this point.
-Chris
I'm just trying to build a couple mid bass modules that give me flexibility for integrating different possible solutions for mid range and treble frequencies later. I couldn't care less if my box with a 12" woofer ends up being a design from 1942 or something cutting edge. Especially in this first try at it.
Until I get these mid bass modules built, I can't even experiment with steep crossover slopes at alternative frequencies, to see how it sounds to me.
It's doubtful that I will end up handing off from a 12" woofer to a SEOS-12 wave guide but it would be nice to have the option, as many people apparently think it sounds quite good. As much as I like my little Scanspeak 10f drivers for midrange down to about 450Hz, I somewhat doubt I would be satisfied with the compression drivers and wave guide.
Oh well, looks like this whole thing is going downhill fast. I really do appreciate the help and advice from all you guys up to this point.
-Chris
Oh well, looks like this whole thing is going downhill fast.
Not as far as I can see.
But you take from this thread what you wish.
Some really good advice about different reasonings has been presented.
I've just agreed with the bulk of the existing opinions.
And misunderstood that you were not looking for a scratch 4 way system.
I think the time spent has been productive.
Just that as I said. I'm not interested in helping anyone design a flawed speaker system that I know can easily be bettered.
Keep in mind that what many people say sounds great is not always so.
I have listened to some very highly touted semi-pro designs that have been on the internet lovingly reproduced by scrupulously attentive gents.
I'd describe them as listening to nails on the chalkboard.
And then same drivers, in same box done by another gent that were so good I could listen to them all day long.
It seriously depends on what your frame of reference is.
How many acoustic concerts have you attended?
So you play any musical instrument?
These things radically change the aural perception of people who listen to sound reproduction.
I used to do custom design for one off systems back in the 80's.
I could relate some rather interesting stories.
A simple one is the client that was listening to a Michael Murray recording of a pipe organ on the Telarc label. He heard the brass reeds buzzing within the horizontal trumpets on the pipe organ.
He thought this was a design flaw. He'd never heard that on his other system! Same recording. The speakers must be the problem!
WOW!
So I had to go into a detailed explanation of how that particular sound is produced in the first place and then explain that he had a much higher resolution extremely low distortion system now that shows off good recordings at the best.
And shreds like a blender the not so good recordings.

Think raunchy saxophone on steroids and amped up just a bit. ( a whole bunch ! )
What issues are there at 1000Hz with the SEOS-12? What I was saying is that their polar map shows directivity control to be excellent down to 1000Hz and not too bad down to almost 700Hz. And they recommend a 950Hz minimum crossover.
NO no no! They recommend on DIYsoundgroup an absolute MINIMUM crossover frequency of 950hz! Once you take a look about 'good'* design using SEOS12 you'll see they tends to cut at 1200hz/1300hz range. And if you take a look at Horn.pl site they do recommend 1200hz as cut off frequency.
And guess what Chris: 900hz IS actualy half an octave (+/-) lower than the minimum recommended frequency range of 1200/1300hz! Sound familiar? 😉
What LineSource is talking about is in fact true. But i've heard enough system good and bad to know that it is not the first parameter i will try to optimize as others things can become highly objectable pretty fast to my taste (directivity mismatch for example). I think everybody told you to cut higher your sub as a possible solution and this is one of the tradeoff.
Polar map and plots don't tell the whole story: what about phase behavior? What about acoustic cut off of horn? This is the reason i talked to you in private to study the paper about LR and Bessel filter and their behavior for every parameters which come into play and theyr interaction with acoustical parameters...
*actually i don't know how good they sound as i never heard one but from a theorical point of view...
What i was asking in my previous post was from where have you seen 'common practice' about sub integration and such (as nescessity to use 18")? Thats it which make me perplex and i would like to know your source!
Last edited:
Then why didn't you just say that an extremely steep crossover slope might work to move subs around, without noticing their position, even when crossed over at a higher than normally accepted frequency?
Maybe because the tradeoff for home use are very higly objectable! Poor transient response... being one of them!
If you were going to use your system for a PA you could live with that.
But it seems kind of counterproductive to focus on sealed sub (and other box) to destroy it using brute force in filtering.
What i was asking in my previous post was from where have you seen 'common practice' about sub integration and such (as nescessity to use 18")? Thats it which make me perplex and i would like to know your source!
My source was LineSource explaining why the speakers he posted have 18" drivers built into the bottom of them.
If a 12" midbass has a 100-150Hz low frequency crossover, physically adjacent woofers are required to avoid confusing the listener's brain from shifting sound-stage queues. Well studied in the literature, and why THX set 80Hz as the highest remote woofer frequency.
I think maybe we're hitting some language barriers in this thread. I don't really know what to do about that but I think we're just going in circles at this point.
-Chris
Last edited:
Maybe because the tradeoff for home use are very higly objectable! Poor transient response... being one of them!
If you were going to use your system for a PA you could live with that.
But it seems kind of counterproductive to focus on sealed sub (and other box) to destroy it using brute force in filtering.
Not exactly.
It depends on what you have available in terms of filters.
FIR filters with a bit of tweaking is the best way.
Plus there are ways to do it otherwise.
My source was LineSource explaining why the speakers he posted have 18" drivers built into the bottom of them.
I think maybe we're hitting some language barriers in this thread. I don't really know what to do about that but I think we're just going in circles at this point.
Ok i just re read post from linesource and i don't see where he explained or specified the use of 18" drivers: he explained sub must be located near 'satellite' (mid/high which is true following the 1/4wavelength) but for the rest i don't see. Please Chris point me to the post you interpreted that way.
Not exactly.
It depends on what you have available in terms of filters.
FIR filters with a bit of tweaking is the best way.
Plus there are ways to do it otherwise.
You are right Mark (and this is what i use actually with my monitors as my dsp allow quite good fir). I was thinking about 8pole LR or the likes (IIR). I should have been more precise.
You are right Mark (and this is what i use actually with my monitors as my dsp allow quite good fir). I was thinking about 8pole LR or the likes (IIR). I should have been more precise.
You can also do it decently well with IIR filters and allpass filters to do some of the correction.
And there are still other ways yet.
Audio is many times different paths leading to the same destination.
Sometimes it is different path full stop.
Ok, i 've re read it ad i ask you to read post #97:
the 80 hz spec for THX is for the surround and the central channel. The L and R should be 'fullrange' frequency wise, and the sub (.1) channel is and independent one (you don't nescesseraly have info from l/r directed to it: it there 'just' for effect and this is why you sometimes hear about 'musical' sub and 'theater' subs).
With multichanel format the surround don't really need low end content (and usually whan they have they are directed to the 'sub' dedicated channel, which can happen with signal from L/R too if you want a more 'dramatic' effect or release them (L/R) during complex passage).
The THX is for pretty large room and more standardized one (as theater are more homogenous room than living rooms) but you cannot translate this to your home especially for music systems without altering parameters fc being one of them.
The description of bass contents for musical source given by Mark previously is spot on. Your choice to try moving the fc point using his advice or not.
the 80 hz spec for THX is for the surround and the central channel. The L and R should be 'fullrange' frequency wise, and the sub (.1) channel is and independent one (you don't nescesseraly have info from l/r directed to it: it there 'just' for effect and this is why you sometimes hear about 'musical' sub and 'theater' subs).
With multichanel format the surround don't really need low end content (and usually whan they have they are directed to the 'sub' dedicated channel, which can happen with signal from L/R too if you want a more 'dramatic' effect or release them (L/R) during complex passage).
The THX is for pretty large room and more standardized one (as theater are more homogenous room than living rooms) but you cannot translate this to your home especially for music systems without altering parameters fc being one of them.
The description of bass contents for musical source given by Mark previously is spot on. Your choice to try moving the fc point using his advice or not.
Last edited:
You can also do it decently well with IIR filters and allpass filters to do some of the correction.
Mark i will contact you about that point using pm.
I have pretty complete allpass in the Lake and don't use them because i don't fully understand how to have best use of them. Maybe you could help me! 🙂
Chris, about the SEOS i'm not trying to convince you about the nescessity of horn or other mid solution, i just want you to understand the point about the choice to be made.
I know i've flooded you under tons of infos but i think your understanding of horn behavior made you see the issues faced in a way which is not 'correct' (whatever it means). The link given about horn theory are a great global grasp about the problems faced in this situation. Don't hesitate to read them when you'll have spare time to understand what i'am trying to explain.
I know i've flooded you under tons of infos but i think your understanding of horn behavior made you see the issues faced in a way which is not 'correct' (whatever it means). The link given about horn theory are a great global grasp about the problems faced in this situation. Don't hesitate to read them when you'll have spare time to understand what i'am trying to explain.
Mark i will contact you about that point using pm.
I have pretty complete allpass in the Lake and don't use them because i don't fully understand how to have best use of them. Maybe you could help me! 🙂
I am not a filter guru.
Iam a filter plugger. As in I keep working at it until I get what I am looking for and with what sounds best.
Some simple tricks is use a reference that is neutral.
A good set of headphones is a great example.
Next be careful not to believe that you are continuously making the sound better.
Take a break and listen to something else and go back to your work of art a day or so later.
Sometimes I have done that and reset everything back to the original stored values and found that it was indeed the best filter combination.
Last bit of information.
If it sounds right but measures not right try moving your mic.
Or move your measurement outside.
All becomes clear when you rid your microphone of unwanted reflections.
The reason for the THX 80Hz frequency spec is that you can't remotely locate your sub at a higher frequency and not notice that the sub isn't in front of you. My experience is that even 80Hz is too high. I prefer 70Hz for the crossover frequency if I'm going to position my sub(s) away from the front sound stage, even with fairly steep slopes.
I have not tried an extreme slope that basically just turns off any output beyond the crossover point. I'm almost certain it would not sound good without the use of FIR filters and have my doubts about it even then.
I do have the ability to use FIR filters with JRiver Media Center but that doesn't help me until I have some kind of mid bass module constructed.
I haven't really even heard a good reason I would want to cross over higher than the normal 70 or 80 Hz max. Just some vague statements about "cleaning up the midrange". I'm not convinced that crossing over at 70Hz instead of 100Hz will make my midrange dirty to begin with.
I'm willing to be convinced of that but so far I've just seen a couple opinions.
I have not tried an extreme slope that basically just turns off any output beyond the crossover point. I'm almost certain it would not sound good without the use of FIR filters and have my doubts about it even then.
I do have the ability to use FIR filters with JRiver Media Center but that doesn't help me until I have some kind of mid bass module constructed.
I haven't really even heard a good reason I would want to cross over higher than the normal 70 or 80 Hz max. Just some vague statements about "cleaning up the midrange". I'm not convinced that crossing over at 70Hz instead of 100Hz will make my midrange dirty to begin with.
I'm willing to be convinced of that but so far I've just seen a couple opinions.
- Status
- Not open for further replies.
- Home
- Loudspeakers
- Multi-Way
- My attempt at 4-way digital active