If tweeters are better suited to highs, small speakers to mids and large woofers to bass, why don’t we have an array of ten say ten or twenty gradually increasing speakers from tiny tweets to ENORMOUS woofers, each with a unique crossover at a specific frequency they perform best at?
We really seem to have settled on three sizes. Tweets, miss, and subs. Why is that the case?
Surely more speakers handling less varied frequencies means a better response, right?
You must log in or register to comment.
That’s exactly what my speakers aren’t!
Xovers are not invisible. Less is more.
What are the downsides to using a crossover? My understanding was is simply clipped the signal off above or below the relevant frequencies.
Umm, no. Every crossover point is actually a range of frequencies over which both drivers are operating. They must operate in phase and in a time alligned manner. The crossover components themselves introduce phase changes that are frequency dependent, so this is anything but simple.
And even if you get all that perfect, the drivers interact with each other as their wavefronts propogate. Those wavefronts have different dispersion patterns, and so their reflected energy is not balanced in the same way as their direct energy.They interact inside the cabinet as well.
The perfect speaker would be a single driver. This is part of the allure of panel speakers, although the PERFECT speaker would be a single, point-source, infinitely small driver.
The signal goes through a lot of components.
Contrary to popular belief, capacitors impact the sound. And twenty capacitors impacts the sound more than two.
Price and size.
There are 4, 5, 6 way designs and higher. Drivers usually are designed to cover a decently wide frequency response, so often two or three drivers can handle the audible range from 20Hz to 20kHz.
Things get a lot more complicated when you add more drivers. The crossover, impedance matching, level matching, driver spacing, driver interactions, baffle reflections, cabinet dimensions, time alignment - so many things need to be considered when trying to use multiple drivers in a speaker design.
Kind of making things unnecessarily complex, without being objectively better than a simpler design. Simplicity is the hallmark of good engineering, IMO.
I personally have a 4-way speaker setup, but have heard great systems using single drivers as well.
Yup. Less is more.
Check out these babies
https://www.youtube.com/watch?v=xKhcfXY8jFQ&ab_channel=PaulMcGowan%2CPSAudio
Mother of god that was awesome to watch. 1.5 tonnes each?!
This is exactly what was in my head with x-overs for every speaker.
I’m beginning to understand why it’s not so practical.
Also the upgraded model has 12,000 watts of amplification just for the woofers.
JFC. That’s a ridiculous amount haha
You want to minimize the number of crossovers. So it’s a trade off between the bandwidth of the drivers and minimum number of crossover points. Two crossovers is a pretty good compromise.
You want the least amount of components to avoid phase shift and distorsion etc. and the ideal speaker is a point source that plays all frequencies. You would have a real weird acoustic center if you had 10 or 20 drivers all with unique frequency ranges.
It’s matter of cost, and compromise. Crossovers are not infinitely steep, so there will be multiple drivers playing the same signal, resulting in interference in the dispersion pattern. That interference pattern is undesirable, as it creates directivity issues on the speaker, and this is why drivers that play the same signal are placed vertically close together, as this makes them merge better and points the crossover null lobes up and down rather than sideways, where they might point to someone’s listening spot. So your hypothetical speaker would be very tall, most likely, and come with lots of directional lobing.
2-way is the smallest it can get, but it is somewhat a compromise in harmonic distortion at midrange, or the bass extension, depending on how large the woofer is. 3-way is likely the sweet spot, the smallest design that realize both low harmonic distortion and extended bass response, in my opinion. If midrange and tweeter are coaxial, that is close to ideal, e.g. see Kali IN-8, various KEF designs, or Genelec The Ones series. Then, 4-way and 5-way may be good if you need absolutely the full bass range down to sub-20 Hz, or very high SPL, or something, but I doubt that kind of design matters all that much for home use.
I have 10" woofer and 1" metal dome tweeter on my living room’s speaker system, thus it is a 2-way. The large woofer gives in-room F3 bass to around 25 Hz, though the woofer and tweeter are placed quite far apart, and thus the crossover lobes are sizable and start near the acoustic axis, and thus you can’t allow much vertical displacement from the acoustic axis if you want to enjoy flat on-axis response. Therefore, longer distance is ideal. From measuring it, I can plainly see that the woofer is struggling to keep harmonic distortion below about -50 dB, until the crossover happens at about 2 kHz. As soon as it crosses over to the tweeter, harmonic distortion drops below the room’s noise floor.
Look at really expensive flagship speakers like the Wilson Audio ones, and you will see they have several sized speakers in each “case”.
But we are now in the hundreds of thousands of dollars field.
Answer from an audio engineer here. We look at audio reproduction in octaves, and those octaves don’t translate in frequency response linearly, rather they translate logarithmically. A gradual increase in driver size and subsequent narrow frequency crossover would result in what is called comb filtering, and excessive comb filtering at that. What that means to the listener is a loss in various frequency bands at different registers that will change based on where you are listening from due to a phenomenon called phase cancellation. This results in an incomplete soundscape or sonograph.
Hope this helps.
I think this is the most interesting question I’ve ever asked and I am loving every response learning more and more.
Thanks for taking the time.
The reason is to do with physics and the engineering of drivers. To reproduce low frequencies at standard levels in a room a driver needs to be large to displace enough air. For high fidelity in the home this tends to be around 10-12" or several smaller drivers. As the frequency rises the cone is going to cease moving as one and start resonating which degrades sound quality. A driver can typically cover a frequency range of about a decade with a flat response between a low frequency roll-off in output due to being too small and a high frequency degradation due to being too big and the cone, surround and suspension resonating. In order for the resonances to start at a higher frequency the cone needs to be smaller. Fortunately at higher frequencies physics requires less air to be displaced to be equally as loud as low frequencies. So a smaller driver that doesn’t reproduce the low frequencies enables us to cover another decade at higher frequency.
This leads to a high sound quality speaker requiring around 3 driver sizes to cover the 3 decades from 20-20kHz. Other details to do with overlapping crossover regions, smooth radiation patterns, controlling room modes at low frequencies leads to the optimum being more like 4 in the form of subs distributed around the room for the lowest couple of octaves and 3 way mains.
Having said that a small and modestly priced 2 way with 6.5" midwoofer and 1" tweeter can often do a pretty good job. It won’t play loud enough cleanly, go deep enough cleanly and will have some audible degradation due to resonances but often not to an intrusive extent.
If you spent $20 per driver on ten drivers, you’d be in for $200 per speaker in just drivers. The crossover will also be very large and expensive.
You’ll get vastly better results from a $100 woofer and $100 tweeter with a simpler crossover and a smaller box.
A lot of the really high end speakers will have those build in. https://www.mcintoshlabs.com/products/speakers/XRT21K
The cone size isn’t directly proportional to the frequencies that it can produce. Since most audio companies try to use common parts for better pricing/margins they will gravitate to common sizes. That is mainly because that is what is manufactured. The difference between a 1” tweeter and a 1.35” tweeter is super minor because they can pitch produce the same frequency range. But a 1” tweeter will cost less than a 1.35” because 1.35 isn’t used anywhere.
Because it turns out 2-way is the simplest and they sound pretty damn good.
But the type of thing you’re taking about certainly exists. They’re called line array speakers.
81 drivers of 4 different kinds: https://www.audiogon.com/listings/lisb2efc-mcintosh-xrt2-1k-flagship-line-array-loudspeaker-rarely-seen-for-sale-full-range