• WANTED: Happy members who like to discuss audio and other topics related to our interest. Desire to learn and share knowledge of science required as is 20 years of participation in forums (not all true). Come here to have fun, be ready to be teased and not take online life too seriously. We now measure and review equipment for free! Click here for details.

2-way vs 3-way speakers

Chromatischism

Major Contributor
Forum Donor
Joined
Jun 5, 2020
Messages
2,916
Likes
2,159
Actually 300 to 3000 Hertz is the range designed into all human voice communication systems. The first examples are telephones and 2-way EMS radio systems. Personally I avoid any crossovers between 90 and 3500 Hertz for my optimal enjoyment...
There are so many variables in speaker design that while that may be good in your current system, it certainly can't be a rule. I now have a 2-way with a tweeter crossed at 1800 Hz (LR4) and it sounds better (there is some element of preference here) than when it was crossed at 2670 Hz. There are no issues with voices. The tweeter is now doing more work (wider dispersion than the woofer at those frequencies) and more of the spectrum is thus being shaped by the waveguide. With waveguides, lower is generally better to take full advantage.
 

richard12511

Major Contributor
Forum Donor
Joined
Jan 23, 2020
Messages
4,289
Likes
6,368
I do wonder sometimes about why companies choose the 2 way design when they don't have to, even sometimes to their(IMO) detriment. JBL is the first one that comes to mind. 708p and M2 are expensive enough that they could easily afford to go 3-way, but they still choose to go 2 way. For those speakers it works, I'm guessing it works mainly due to the awesome waveguide and DSP. However, some of their other designs, like the HDI center speakers, I'm thinking those could be much better with a 3 way design(think the centers are 2.5?).
 

antennaguru

Senior Member
Joined
Jun 16, 2021
Messages
318
Likes
258
Location
USA
There are so many variables in speaker design that while that may be good in your current system, it certainly can't be a rule. I now have a 2-way with a tweeter crossed at 1800 Hz (LR4) and it sounds better (there is some element of preference here) than when it was crossed at 2670 Hz. There are no issues with voices. The tweeter is now doing more work (wider dispersion than the woofer at those frequencies) and more of the spectrum is thus being shaped by the waveguide.
Fair enough. Prior to this speaker system I had 2-Ways with 2 KHz crossovers and enjoyed them very much until I built this set of OB main speakers with 3.5 KHz crossovers. The last speakers always gave a great reproduction of the recording, and these just don't sound like a reproduction anymore - and instead sound like I'm there at the actual performance. Same Infinity IRS stereo subwoofer system, plus extra push-pull subwoofer. These mains just get out of the way better. Probably more to do with the higher transparency and almost 10 dB higher efficiency than the higher crossover point...
 

richard12511

Major Contributor
Forum Donor
Joined
Jan 23, 2020
Messages
4,289
Likes
6,368
The advantage is: After you tell your wife you want a 3-way, you pause, and then when you explain it is a speaker she's okay with it.
What if she's ok with it before she knows you're talking about speakers?
 
Last edited:

richard12511

Major Contributor
Forum Donor
Joined
Jan 23, 2020
Messages
4,289
Likes
6,368
Thank you for proving my point! I listen mostly at 60 - 80 dB SPL, though some peaks get higher.
?...that graph proves the opposite of your point, unless I'm misunderstanding your point. That graph shows that the human ear is most sensitive in the 2.5-4.5kHz region.
 

antennaguru

Senior Member
Joined
Jun 16, 2021
Messages
318
Likes
258
Location
USA
Single driver, full range speakers not mentioned or being discussed either? ;)

No crossover at all... potential IMD and beaming though.



JSmith
There's no crossover like no crossover!

Unfortunately single full range drivers have a tough time coping with the upper frequencies and the mid-bass. I built a set of Short Line Arrays with 4 X 3 inch full range drivers, so no crossover there. Just a series/parallel driver configuration. They were very nice sounding with good subs, and went high enough to be enjoyable because they were small full range drivers. Using four gave higher efficiency and power handling.
 

Jdunk54nl

Addicted to Fun and Learning
Joined
Aug 5, 2020
Messages
760
Likes
776
Location
Arizona
I guess you don't listen to the human voice then. I like listening to that instrument...
I do. That statement has literally zero to do with what frequencies our ears are most sensitive at though. But I’m done with this conversation, it isn’t worth my time. You have the well studied Fletcher Munson and equal loudness contours. Very well researched. Either believe the data or don’t.
 

antennaguru

Senior Member
Joined
Jun 16, 2021
Messages
318
Likes
258
Location
USA
Sure, at zero dB SPL. Look higher on the graph where listening is mostly done in the 60 - 80 dB SPL range and tell me that 500 to 2500 Hertz doesn't matter.
?...that graph proves the opposite of your point, unless I'm misunderstanding your point. That graph shows that the human ear is most sensitive in the 2.5-4.5kHz region.
 

antennaguru

Senior Member
Joined
Jun 16, 2021
Messages
318
Likes
258
Location
USA
I do. That statement has literally zero to do with what frequencies our ears are most sensitive at though. But I’m done with this conversation, it isn’t worth my time. You have the well studied Fletcher Munson and equal loudness contours. Very well researched. Either believe the data or don’t.
I guess everyone's hearing is unique. Mine is most sensitive to Cello and Voice frequencies, and that region is where loudspeaker crossover points typically bother me the most. I actually play piano though, and my keyboard amp uses a full range driver - unless I break out my JBL Cabarets to fill a larger space.

The F-M curve was to do with something totally different, the need to boost lows and highs at very low volumes - NOT intended for normal listening volumes. Anyone still using a "Loudness" switch? Look at 60 - 80 dB SPL, which is more representative of normal listening volumes. BTW, do note that the horizontal axis of the graph is a log scale when you interpolate points in between the specifically marked frequencies.
 
Last edited:

raph

New Member
Forum Donor
Joined
Jul 23, 2019
Messages
1
Likes
1
However if you wanted to design a 3-way speaker that completely avoided crossovers within a reasonable range of vocal band, it would look something like this:
Bass: 20-100 Hz
Mid: 100-4000 Hz
Tweeter: 4000-20000 Hz

Most 3-ways (or all?) I've seen don't cross to the bass driver that low, nor do they have a midrange that can cross that low if they wanted to. If it can, it probably won't extend to 4 kHz.
Looks to me that Purify is trying to address this midrange band with their 4" "dedicated midrange" driver https://purifi-audio.com/ptt4-0x04-nfc-01/
The FR graph seems to show a pretty stable behaviour from 100Hz to 3.5KHz. It presents the lowest HD in that range compared to their other drivers. Or am I mistaken ?
 

Newman

Major Contributor
Joined
Jan 6, 2017
Messages
1,102
Likes
1,259
1635915702595.png

Easiest demonstration of this.
Being sensitive is one thing, but being sensitive to colorations and distortions could be quite another thing. I think you are conflating the two. Without evidence that they are in the same range, I don’t think your evidence is relevant.

Wracking my brain, I seem to recall something about 1000 to 3000 Hz for sensitivity to coloration and distortion, but would love to see if anyone can confirm this.

cheers
 

JRS

Addicted to Fun and Learning
Joined
Sep 22, 2021
Messages
702
Likes
613
Location
Albuquerque, NM USA
It is rather silly to imagine a speaker that has no crossovers within the human vocal range, considering that would encompass all of the harmonics and extend from 85 Hz to 8 kHz.

However if you wanted to design a 3-way speaker that completely avoided crossovers within a reasonable range of vocal band, it would look something like this:

Bass: 20-100 Hz
Mid: 100-4000 Hz
Tweeter: 4000-20000 Hz

Most 3-ways (or all?) I've seen don't cross to the bass driver that low, nor do they have a midrange that can cross that low if they wanted to. If it can, it probably won't extend to 4 kHz.
Don't forget Tim Storms who can hit 0.3Hz when he tries. Elephants are impressed.
 

YSDR

Active Member
Joined
Sep 5, 2019
Messages
103
Likes
130
You can spread the crossover frequencies so that the important vocal frequency range can be filter free.
Yes, it can be, but I don't think that's a very important aspect. Check Revel's top speaker's (or 99% of top of the line 3 or more way, from smaller home speakers to the most expensive studio loudspeakers) crossover frequencies for example:
"Four-way, high-order acoustic response @150 Hz, 575 Hz, and 2.3 kHz"
https://www.revelspeakers.com/produ...color=Mahogany-USA-Current&cgid=floorstanding
 
Last edited:

dasdoing

Major Contributor
Joined
May 20, 2020
Messages
1,952
Likes
954
Location
Salvador-Bahia-Brasil
There's no crossover like no crossover!

Unfortunately single full range drivers have a tough time coping with the upper frequencies and the mid-bass. I built a set of Short Line Arrays with 4 X 3 inch full range drivers, so no crossover there. Just a series/parallel driver configuration. They were very nice sounding with good subs, and went high enough to be enjoyable because they were small full range drivers. Using four gave higher efficiency and power handling.

here is a 4" that claims to go up to 15Khz. you can argument that the range aboce that isn't realy needed
 

jae

Addicted to Fun and Learning
Joined
Dec 2, 2019
Messages
582
Likes
654
Wracking my brain, I seem to recall something about 1000 to 3000 Hz for sensitivity to coloration and distortion
This is the approximate frequency range of the formants that are the most important in speech intelligibility, so I don't doubt that is the case.
 

abdo123

Major Contributor
Forum Donor
Joined
Nov 15, 2020
Messages
4,749
Likes
4,274
Location
Brussels, Belgium
In Addition to what has already been mentioned, in my opinion 3-ways are way less likely to have resonances in the midrange, specially when you're buying off a boutique or a cheap brand that does not measure with a Klippel or anechoic chamber.

Quasi-anechoic measurements usually have the lowest resolution at the mid-range, so companies are likely to miss out resonances in that range. Add a crossover into the mix and it becomes a struggle to find an affordable 2-way speaker with a decent mid-range performance.

However if you buy a Revel or a KEF (not the Q series) you're likely not going to face issues in the mid-range. it's mostly an issue with amateur designers with low resolution measurements.
 
Top Bottom