2 tweeters vs 1 tweeter vs no tweeter

[shadowsfal5656]

I had a question regarding using 2 identical bass cabs, both with tweeters. I was hoping someone with an audiophile or acoustic engineering background or with similar knowledge could chime in and give their perspective.

There are many bass players out there that own matching cabinets (2×410, 2×112, 2×210, etc.) that many times, bass players will take both for when gigs require more volume. Alot of times both of these cabinets have tweeters. What are the detrimental audio effects of using 2 tweeters vs 1 tweeter vs no tweeter? I have read online that in general for audio you should not use 2 tweeters as this can cause something called "comb filtering" that negatively affects the higher frequencies that are essentially the reason you have tweeters in the first place. Likewise, this makes me think of something else.

Let's say you have 2 Ashdown 210 cabinets. I've read some of these cabinets have their crossovers set at 2khz. Now let's say you have just one tweeter on. Won't that mean one cabinet will be reproducing frequencies up to 2khz and the other one will be reproducing frequencies as high as it can, which will result in uneven audio wavelengths that can negatively affect your sound? If this is the case wouldn't it just be better to have your tweeter completely off?

The reason I ask is because I have heard bass players talk about the importance of having matching cabinets to not affect the sound. At the same time I have heard some of these people say it's better to just keep one tweeter on. So I was hoping that someone with better knowledge on this topic could give their perspective.

 

[michael_t]

Thanks for posting this. I was just asking myself the same question. Talk about perfect timing!

 

[ardgedee]

Won't that mean one cabinet will be reproducing frequencies up to 2khz and the other one will be reproducing frequencies as high as it can, which will result in uneven audio wavelengths that can negatively affect your sound?

No, you have to think of the whole speaker array as a unit consisting of a bunch of low frequency drivers clustered together plus (zero|one|two|...) high frequency driver(s). If using one tweeter provides a better balance of highs to lows than two tweeters, then that's what you should use. If two tweeters sounds better, use them both.

Any time you have multiple drivers you have to deal with comb filtering to some degree. You also have to deal with the interference from the sound bouncing off hard surfaces like floors and walls. The real world we live in is messy.

 

[agedhorse]

In reality, it comes down entirely to personal preference.

 

[Wasnex]

I am not an acoustic engineer. This is too the best of my knowledge/ability. I don't claim it's 100% technically correct.

Theoretically, you would only want one tweeter to avoid comb filtering. However, I think the best practice is to try it and judge for yourself. I have tried it a couple of times and preferred both tweeters on.

The operating range and spacing between tweeters determines the comb filtering. You get destructive interference across the axis of alignment, but not across the other axis. So if you have the tweeters aligned vertically, comb filtering occurs across the vertical axis, but not across the horizontal axis.

Comb filtering and lobbing are different ways to describe the same thing. Here are the patterns that occur based on wavelength VS spacing.
The symbol for wavelength is Lambda

Distance is the center to center spacing between drivers.

Here are the resultant patterns:

I think the most common crossover frequency for 2-way bass cabs is 3.5kHz, which has a wavelength of just under 1". You mentioned 2K, which has a wavelength of about 1.7".

When I stack my GG1288T UQ cabs vertically, I believe the center to center spacing between the tweeters is 23". This means the spacing between the tweeters is relatively large in comparison to the wavelength. AFAIK by making the spacing large in relation to wavelength you get lots of small lobes that are very close together. AFAIK, at some point they start to be become less noticeable.

Distance to Lambda ratio for my cab at 3.5K is ~23/1. The ratio at 2K would be ~23/1.7. The chart above only only goes to a ratio of 5/2 (2.5/1). Per the chart, this is the resultant lobing pattern at 5/2:

Spacing to achieve this pattern at 3.5K would be about 2.5". Spacing for this pattern at 2K would be about 4.25". At spacing of 23" there will be a lot more lobes and they will be closer together.

Additionally, longer distances between sound sources reduces the degradation caused by off-axis come filtering. Keep in mind the SPL from a point source drops off at 6dB per doubling of distance. If the array is aligned vertically, your ears will be closer to the top tweeter. Because the bottom tweeter is further away, the SPL will be lower and it will have less impact on what you hear.

Let's keep in mind that tweeters are directional. A lot of the tweeters in BG cabs have a 60 degree conical pattern.

Note the patterns cross more if the tweeters are close together (more interference).

Put some distance between them and the patterns cross less (less interference).

When possible, it can be useful to experiment with both spacing and splay between drivers to make the comb filtering less objectionable. For example I often ran a pair of 12 + horn PA cabs on each side of the stage. It generally sounded better to space them out to about 1.5' to 2.5' and adjust the splay between them, rather than tight packing them.

.

 

[agedhorse]

In reality, with the boundary reflections, comb filtering for bass is going to be inconsequential.

 

[Al Kraft]

With two matching cabs stacked or even one single cab I think the only comb filtering that you would experience is for people in audience seats where reflected sound associated with the venue acoustics would be involved. And that would occur with single driver or multiple drivers, because it would be a result of how direct and reflected sound waves combine at that seat in the audience.

Think for a minute about speaker columns like the old Shure Vocal Masters or latest Genzler bass cabs with multi-mid/tweeter arrays, or Duke's AudioKinesis C112T bass cab with a mid/HF driver in front and in the back. If having two or more HF/mid drivers together was a problem, these iconic and stellar products I mentioned wouldn't work at all.

At the end of the day the only question I think you need to worry about answering is how much high end do you want coming from your cab/cabs. In any real world situation where there is a possibility of reflected sound or with widely separated cabinets whose direct acoustic dispersion crosses paths you will get some degree of comb filtering.

Finally, if someone sits between multiple speakers all producing the same signal (with no direct or reflected dispersion interference) they might experience comb filtering due to "time of sound arrival" variance from those sources depending on any difference in distance from sources. This difference in time of arrival of the sound is quite similar to hearing a direct and reflected sound create comb filtering. While the speed of sound is the same for all frequencies there is a difference in degree of directionality of wave propagation between lower and higher frequencies that can also exacerbate a reflected sound problem in one frequency range and not in other frequencies.

 

[Silky Johnson]

Side question for agedhorse, in the Subway 212 and 215 cabs, are the horns/crossover louder than the 112 and 115 cabs to match having double the woofer? Always wondered this

 

[WRM]

Twice the woofer, twice the tweeter?... It is interesting that 2Xs only have one tweeter. I like tweeter, and it's a big difference between Bass and Skinny-string Cabs, but now that I think about it, guitar drivers probably have more top end than Bass drivers, so they don't need one?...

 

[agedhorse]

Side question for agedhorse, in the Subway 212 and 215 cabs, are the horns/crossover louder than the 112 and 115 cabs to match having double the woofer? Always wondered this

Yes, there are adjustments to the crossover to normalize the response between the 115 and 215 cabinets.

 

[shadowsfal5656]

These are all great answers, thank you so much guys.

 

[shadowsfal5656]

I am not an acoustic engineer. This is too the best of my knowledge/ability. I don't claim it's 100% technically correct.

Before I forget, I just wanted to tell you thank you for this explanation. This was so detailed and actually made a lot of sense. For you not being an acoustic engineer this was a very well explained break down of what I wanted to know.

On a side note, I appreciate your use of the variable lambda. This brought back memories/nightmares of having to take Calculus 3 in college lol

 

[-Asdfgh-]

I am not an acoustic engineer. This is too the best of my knowledge/ability. I don't claim it's 100% technically correct.

Theoretically, you would only want one tweeter to avoid comb filtering. However, I think the best practice is to try it and judge for yourself. I have tried it a couple of times and preferred both tweeters on.

The operating range and spacing between tweeters determines the comb filtering. You get destructive interference across the axis of alignment, but not across the other axis. So if you have the tweeters aligned vertically, comb filtering occurs across the vertical axis, but not across the horizontal axis.

Comb filtering and lobbing are different ways to describe the same thing. Here are the patterns that occur based on wavelength VS spacing.
The symbol for wavelength is Lambda View attachment 7282842

Distance is the center to center spacing between drivers.

Here are the resultant patterns:
View attachment 7282843

I think the most common crossover frequency for 2-way bass cabs is 3.5kHz, which has a wavelength of just under 1". You mentioned 2K, which has a wavelength of about 1.7".

When I stack my GG1288T UQ cabs vertically, I believe the center to center spacing between the tweeters is 23". This means the spacing between the tweeters is relatively large in comparison to the wavelength. AFAIK by making the spacing large in relation to wavelength you get lots of small lobes that are very close together. AFAIK, at some point they start to be become less noticeable.

Distance to Lambda ratio for my cab at 3.5K is ~23/1. The ratio at 2K would be ~23/1.7. The chart above only only goes to a ratio of 5/2 (2.5/1). Per the chart, this is the resultant lobing pattern at 5/2:

View attachment 7282850

Spacing to achieve this pattern at 3.5K would be about 2.5". Spacing for this pattern at 2K would be about 4.25". At spacing of 23" there will be a lot more lobes and they will be closer together.

Additionally, longer distances between sound sources reduces the degradation caused by off-axis come filtering. Keep in mind the SPL from a point source drops off at 6dB per doubling of distance. If the array is aligned vertically, your ears will be closer to the top tweeter. Because the bottom tweeter is further away, the SPL will be lower and it will have less impact on what you hear.

Let's keep in mind that tweeters are directional. A lot of the tweeters in BG cabs have a 60 degree conical pattern.
View attachment 7282861

Note the patterns cross more if the tweeters are close together (more interference).
View attachment 7282859
Put some distance between them and the patterns cross less (less interference).

View attachment 7282860

When possible, it can be useful to experiment with both spacing and splay between drivers to make the comb filtering less objectionable. For example I often ran a pair of 12 + horn PA cabs on each side of the stage. It generally sounded better to space them out to about 1.5' to 2.5' and adjust the splay between them, rather than tight packing them.

.

Some of the Bill Fitzmaurice designs have an array of tweeters. I've never heard one, though.

 

[el murdoque]

I've never approached this topic from this perspective. I've always come from a more practical viewpoint:
-If I'm running the tweeter full on when using one cab, I simply use the second cab with the tweeter full on.
-Likewise, If I'm using the one cab with the tweeter off, I use the second cab with - surprise - tweeter off.
-Most of the time, I use the tweeter at less than half, maybe about a third. When I add a second cab, I turn up the tweeter in the uppermost cab while leaving the bottom tweeter off.



Luckily, I've never used a cabinet with the tweeter around 3/4 up and added a second one, because that might prove difficult

 

[Avigdor]

I am not an acoustic engineer. This is too the best of my knowledge/ability. I don't claim it's 100% technically correct.

Theoretically, you would only want one tweeter to avoid comb filtering. However, I think the best practice is to try it and judge for yourself. I have tried it a couple of times and preferred both tweeters on.

The operating range and spacing between tweeters determines the comb filtering. You get destructive interference across the axis of alignment, but not across the other axis. So if you have the tweeters aligned vertically, comb filtering occurs across the vertical axis, but not across the horizontal axis.

Comb filtering and lobbing are different ways to describe the same thing. Here are the patterns that occur based on wavelength VS spacing.
The symbol for wavelength is Lambda View attachment 7282842

Distance is the center to center spacing between drivers.

Here are the resultant patterns:
View attachment 7282843

[...]

Spot on! So the important thing here is to look at the wavelength: At 2000Hz your wavelength is about 17 cm (half a foot), so for any realistic spacing between the tweeters you are in a realm where moving your head even a little bit will wash out the interference effects. Add in multiple reflections in the room (as agedhorse pointed out) and this is not something I would ever worry about in a real setting.

I've had great success demonstrating the above effect with two frequency generators finely tuned and speakers carefully placed at the front of a large lecture theatre with 200+ students. The frequency of 300 Hz gives a wavelength of a meter, which compares well to the average seat spacing. In a packed auditorium you can then ask people to stand up if the volume at their seat is loud (practical tip: give a reference tone to compare the volume to first). And then everyone can immediately see the resulting pattern in the room .

What I learned from that is that you really need to work on your setup, optimise speaker placement, minimise reflections, use pure tones (single frequency sine waves, no harmonics) only for the effect to be noticable at all. None of these conditions are fulfilled in a gigging scenario. You're good to go!

 

[MJ_Sotti]

People think way too much nowadays

 

[Pings Bass]

I'd say use your ears, but in my experience if the stage is so loud that you need two cabs, your ears will lie to you and nobody will care anyway.

 

[Jefenator]

I wouldn't be one of those guys telling you it's necessary to have matched cabs; I've been happily rocking a FrankenStack (Bergantino HDN210 w/ Bag End S-15D) on louder gigs since the pandemic. (The Bag End is tweeter-less so that makes it easy.)
Before that, I mostly used a single 2x12. Less cables to hook up, no need to worry about tweeter synchronization. (OTOH I had to schlep a bulky 75# box up 2 flights of stairs every week.)
If I had two cabs with tweeters, I'd just feel it out on the first few gigs. (It's not like a lunar landing module where someone will die if there's a flaw in the calculations beforehand.)

 

[Wasnex]

At 2000Hz your wavelength is about 17 cm (half a foot), so for any realistic spacing between the tweeters you are in a realm where moving your head even a little bit will wash out the interference effects.

Thanks for the correction . I looked at the wrong column on my spreadsheet and posted 1/4 wavelengths rather than full wavelengths. Wavelength for 3.5K is about 3.9" and wavelength for 2K is about 6.8".

Still relatively small in regards to a spacing of 23". You still should get more lobes closer together than this:

FYI using full wavelength, at a spacing of 23" this pattern should occur at a frequency of about 1.5K. 2K will look pretty similar.

If the tweeters were arrayed horizontally at this distance, you might hear some phasing as you turn your head from side to side. Probably not very noticeable with bass, but you might notice with full range program material.

If the tweeters are arrayed vertically, I don't think you will hear any phasing as you turn you head from side to side because your ears are horizontally arrayed. However, you might notice a change in levels depending on where your ears are in relation to the lobes and nulls in the dispersion pattern for a given frequency.

Keep in mind the distance between the drivers is constant, but the wavelength changes with frequency. Therefore the number and angle of lobes changes with frequency.


Again I think the best thing to do is try it. The spacing between tweeters will vary between cab designs. It's possible the spacing with some cabs would sound better with both tweeters running, while other cabs would sound better with only one tweeter on. Then the question is which tweeter to use.

 

[Al Kraft]

People think way too much nowadays

...or maybe not enough....for me it varies depending on what meetings I spend my day in.