Ceramic 4-way 1st ORDER (bang for the buck)

Carbon Fiber 4-way 1st ORDER (bang for the buck)

Greetings members!

The weekend was productive Fellaz! I have revamped the system, It was slightly inferior, and I have made NEW MODELS
The next five pages of my posts in this thread will be OUTDATED. THIS VERSION IS NEWEST.
!!***CHECK THIS LINK FOR GRAPHS-VISUALS FIRST***!!
https://www.diyaudio.com/forums/multi-way/334161-ceramic-4-1st-bang-buck-6.html#post5709858

This idea has spawned an 3-way OFFSPRING, which also does it, with a COST, COMPLEXITY, and SIZE CUT by a FACTOR OF 2.
!!***https://www.diyaudio.com/forums/multi-way/334161-ceramic-4-1st-bang-buck-6.html#post5710001***!!

I have mapped all the drivers seperately, and then have combined them into a TRUE 1st order electrical 4-way response graph.

The reasons behind this is that I realised the crossover points between the drivers must be -6db to have a flat response, therefore I have made assymetric crossovers to make sure Im as flat as possible.

The tweeter is already modelled with the series LCR notch in parrallel to the driver, but I havent added the effect of the Zobel impedence equalization networks which are in parallel to the cone drivers.
Thier job is to make sure that the higher frequency zones after the low-pass points to the cone drivers are rolling off at least as intended, if not quicker (meaning 9/db octave instead of 6db/octave). This will even out the bumped kinks you see on the red midwoofer graph, and the black midrange graph past thier low pass points.

Here are the crossovers:


1). 1" TWEETER
39$ SB26CDC-4 3.4 OHM
SB Acoustics :: SB26CDC-C000-4

1st order high pass CROSSOVER VALUES (7200hz -0, 5400hz -3db, -3800hz -6db)
C1 = 8.66 uF

5.5$ KZK WHITE LINE MKP CAPACITOR (WILL USE 8.2uF)
0.5$ KZK K78-34 BYPASS CAPACITOR (WILL USE 0.02uf)

SERIES NOTCH FILTER PARALLEL TO TWEETER (inductor resistance value has been subtracted from resistor value)
C = 33.93 uF
L = 1.57 mH (0.375 OHM)
Rc = 5.69 Ohms (INDUCTOR RESISTANCE INCLUDED)


2). 3" MIDRANGE
39$ RF-30QY-PRO 7.4 OHM
GHXAMP 3 Inch Midrange Speakers 8ohm 30W Neodymium 92DB Mediant Loudspeaker For Car Audio Upgrade 3 way speakers 1PC -in Combination Speakers from Consumer Electronics on Aliexpress.com | Alibaba Group

1st order high/low pass CROSSOVER VALUES: (450hz -6db, 675hz -3db, 900hz -0db / 1800hz -0db, 2700hz -3db, 3600hz -6db)
C1 = 33.65 uF
L1 = 0.44 mH

13$ KZK WHITE LINE MKP CAPACITOR (WILL USE 33uF)
0.5$ KZK K78-34 BYPASS CAPACITOR (WILL USE 0.02uF)
8$ VAJD AUDIO 0,19 Ohm 1,6 mm AIRCORE COPPER INDUCTOR (WILL USE 0,44 mH)

INDEPENDENCE EQUALIZATION ZOBEL CIRCUIT:
C = 11.8 uF
Rc = 9.25 Ohms


3). 6" MIDWOOFER
44$ STX 180.8 FCX v2 6.2OHM
W.18.180.8.FCX_v2 - Nagłośnienie STX

1st order low pass CROSSOVER VALUES: (337hz -0db, 450hz -3db, -675hz -6db)
L1 = 2.19 mH

16$ Z52-2,2/1,5 TOROIDAL IRON POWDER INDUCTOR 0,11 ohm (WILL USE 2.2mH)

INDEPENDENCE EQUALIZATION ZOBEL CIRCUIT:
C = 4.05 uF
Rc = 8.75 Ohms

4). 10" WOOFER
82$ STX W.27.500.8 FCX 5.9OHM
W.27.500.8.FCX - Nagłośnienie STX

1st order low pass CROSSOVER VALUES: (66hz -0db, 100hz -3db, -133hz -6db)
L1 = 9.39 mH

20$ Z52-9,1/1,5 TOROIDAL IRON POWDER INDUCTOR 0,22 ohm (WILL USE 9.1mH)

INDEPENDENCE EQUALIZATION ZOBEL CIRCUIT:
C = 25.19 uF
Rc = 7.38 Ohms

(COSTS: tweeter setup- 90$ / PAIR CROSSED, midrange setup - 121$ / PAIR CROSSED, midwoofer setup - 120$ / PAIR CROSSED, woofer setup - 204$ / PAIR CROSSED)
TOTAL: 535$ + zobels, using JANTZEN SUPERES.
If I wont have enought money, I will build cabinets but I'll leave the woofers out for some time. I'll loose only the lowest octave.

738848d1551090668-ceramic-4-1st-bang-buck-system-acoustic-response-electrical-zobels-modelled-png

attachment.php
 
Last edited:
Administrator
Joined 2004
Paid Member
Here's my take:
You have flooded this thread with too much.
You are building a 4 way and the woofers are only 8"? Can't say I've seen that, nor understand. Loosely, if you are going 4 way it's because the woofer is too big and the tweeter too small to make it a 3 way.
You are attempting a first order 4 way xo. Can't say I've seen that before either. They are hard enough to get right on a 2 way let alone a 3 way or 4.
How much background do you have in building and what is it that has brought you thus far? I think people need the questions to be asked more slowly and have you understand that you seem to be approaching this differently than others. That said, I'll step aside and see what response you get. You have had 200 views and 1 response thus far.
 
There are a few good discussion threads on why 1st order crossovers seldom produce good sound, especially with metal cones which have nasty cone break-up. Experts like Troels Gravensen use 3rd order ckt on aluminum dome tweeter and 4th order ckt on aluminum cone midbass (Troels SBAcoustics 61-NAC design).

A few plastic cone drivers (like 6.5" SB17MFC35-4) sound excellent with 2nd order circuits, and would be good candidates for 1st order circuits over a more limited bandwidth.

suggestion: use a simulator like Xsim or Vituix which use SPL+FRD+ZMA data files to compare paper, plastic, and aluminum cone drivers before you purchase parts. Square wave, step and impulse response, polar response, ... help grading.
========
Troels admits that his Ellipticor-3 requires selected drivers with minimal cone break-up and a slanted baffle cabinet in order to use simple 1st order passive electrical crossovers, and that the resulting acoustic slopes are not 1st order.

Troels: "There are very few true 1st order filter speakers in this world. Some may claim they do 1st order filters, but it usually comes with a grain of salt. The filter may be 1st order on the electrical side, but even so, the roll-off can be 2nd, 3rd or even 4th order from very simple filters - due to the drivers' inherent roll-off. And it all depends, because how far beyond the point of crossover must a driver go before we can call it a true 1st order filter? There are no rules for this, not even any general practice. My best bid on this issue is 2 octaves above and below point of crossover. Here the 18WE driver follows 1st order roll-off up to around 6 kHz, then a dip around 8-9 kHz and it stops around 15 kHz but never exceeds target profile. The tweeter is pretty much 1st order all the way from 15 kHz down to 800 Hz, where is starts rolling off 2nd order. "
 
Too much information. Why go with anodized drivers from aliexpress when for similar money you can get much better ones from true manufacturers? SB Acoustics`s ceramic line is not that far. And...looking at your post, I doubt you`d be successfull - you have to be massively lucky to get a 1st order crossover with a 1st order electrical. Indeed, most have more components than 3rd order.
 
Thanks for your responses guys!

1). I will be using double 8" woofers, with a total of 432 cm2 (around 13" worth of bass). What I win in this situation is: A). Transient response is better on smaller drivers, much less moving mass and therefore more detail. B). Less beaming on shallow slope crossover, allowing a higher crossover point C). Lower cost, especially on shipping.

2). I have settled for a 1st order crossover because Troels said you cant go wrong with phase, and I already have DIY experience in 12 / 24 db octave 3 ways, so setting a challenge for myself, in some terms.

3). In Troels quote Linesource posted about electrical 1st order having 1st order acoustic slopes, I have modelled my drivers acoustically flat (+-1db) not merely 2, but 3 octaves, or 18db to resonance peaks. That is more than his minimums.

Only the tweeter has an LCR curcuit, so its low end will surely be 2nd or 3rd order acoustically towards its resonance of 700hz, so I get additional power handling on the tweeter on a 1st order electrical.

4). The SB Drivers I have chosen have been thoroughly tested, and are perfect value in my book. As for the anodized chinese drivers i mentioned, those are flagship units in Chinese terms.
Yes, SB SS and SEAS may be more proven, but at least 2x the cost, and If i were to upgrade, i would do revelators or satoris, which are 4x the cost.

Being metal, they truly are the usual suspects for cone breakup, but as I said I'm rolling of 3 octaves below resonance.
I'm winning in detail, and bass punch, as we know ceramic cones offer more detail until you reach nasty breakups.

5). I would love to know more about abstracts idea of an active crossover. How can I use op-amps and low-cost parts? Tell me about this, really interesting. The goal here has cost efficiency, so this has my attention. I'm not saying I chose the best drivers, but for 5 units / 200$, show me better.

6). I have a real question guys. I understand I can convert-modify my 1st order electrical low-pass inductors on the midrange, midwoofer, and woofer to LCR circuits. I understand the woofer will swallow alot of capacitor, but this is more realistic on midrange and midwoofer. I simply stick to my inductor values and manipulate caps and resistors to achieve a notch filter if I get audible breakups on my setup without the LCRs.

In this case, the Zobel networks stay in place in any scenario, correct? Or im converting the inductors to full LCR, do the zobels go?
 
Last edited:
Also, I have noticed that the resistance of the L- Inductor component of the LCR circuit has to be accounted for in the total value of the R component of the same circuit, correct?

I will use series LCR circuit on the SB26 CAC -4 tweeter:
C = 34.44 uF
Inductor
L = 1.55 mH
Resistor
Rc = 5.977450980392158 Ohms

using a 0.39 OHM 1.55mH inductor, The Resistor has to be 5.977 - 0.390 = 5.607 OHM.

Can someone verity these series LCR calculations for the SB26CAC?
 
Last edited:
Dear MS,

I'm not sure where to even start with you: but here I go!

You need to do some reading on loudspeaker design.

Start with a simple 2-way, with known good data, frequency response data, impedance data (FRD, ZMA). Do some reading on how loudspeaker drivers are measured.

Software:

I would suggest Jeff Bagby's PCD for software, check out the windows based version, WINPCD.

Loudspeaker performance is predicated on having good data, then modeling that data properly, so, watch or read some tutorials on it. Let the drivers tell you where they should be crossed over (I do this with every project).

When you say 1st order, you need to clarify what that means, electrical or acoustical.

This project of yours is way too much with what skills you have. I would massively reduce your plans and try to produce a speaker with 2 drivers that actually achieves acceptable performance.

You can start modeling with known good designs to see what a good design looks like in the modeling software of your choosing.

Good luck.

Greg
 
I'm talking about 1st order electrical. I tried to be precise, Im challenging myself to get it right.. people have asked why 4-way in the first place, well I believe its easier to get right than a 3 way, even more than a 2-way because the drivers are actually playing full beams on a more narrow spectrum.

I understand your concerns, and I understand I wont have a true 1st order acoustical nor am I trying to. 1st order electrical gives minimum components. That was the design goals.

If you've bothered to look, the sensitivity of the drivers in intended frequencies is closely matched, and if you looked into the given paraments further, you'd notice a built in slope, with higher frequencies fading 1.5db minimum above midrange.

Anyway, let's keep this constructive.
 
I'm talking about 1st order electrical. I tried to be precise, Im challenging myself to get it right.. people have asked why 4-way in the first place, well I believe its easier to get right than a 3 way, even more than a 2-way because the drivers are actually playing full beams on a more narrow spectrum.

I understand your concerns, and I understand I wont have a true 1st order acoustical nor am I trying to. 1st order electrical gives minimum components. That was the design goals.

If you've bothered to look, the sensitivity of the drivers in intended frequencies is closely matched, and if you looked into the given paraments further, you'd notice a built in slope, with higher frequencies fading 1.5db minimum above midrange.

Anyway, let's keep this constructive.

Yes, I was putting you on a path of success with the suggestion of a simple 2-way..

I've been building and designing loudspeakers for about 30 years and your goal of designing and building a 4-way loudspeaker is very unrealistic. You need to learn how this stuff works, and it's cumulative. Study, design, test, model, and build your projects.

I will not speak for others, so I'll just say if you stay on this course I am out. If you want to learn how to design and build loudspeakers I am in.

Cheers,

Greg
 
2 way only really works with a titchy woofer that may as well be the mid bass with a added pair of woofers.
better to make that titcy midwoofer a not so titchy midwoofer. something like 6.5 inch and then add a 4 inch mid between that and the tweeter and thus have enough usable bandwidth on each speaker driver to make the dividing network simpler..

I can see the logic.