Fearful Cabs and Comb Filtering: Should I Mirror Baffles?

[fu22ba55]

In the beloved Fearful plans, there's a note about mirroring the baffles when building cabs to stack:

I'm planning another Fearful flatpack build, and this time, I'm doing two 15/6s.

I CAN'T DO THE NARROW, TALL STACK because I need to place a 24" wide tube head on top. So I can either place two 15/6s on their side, or build two 15/6 tubes (the wider variant for tube heads).

I am NOT including tweeters, so I'll only have a 15" and a 6" in each cab, but I still want to heed the comb filtering warning?

Do I need to heed the comb filtering warning?

---

Will anyone hear the difference between A and B below?

Will anyone hear the difference between F and G below?

Will anyone hear the difference between I and J below?
I prefer to have the ports stacked vertically on the right side when facing the cabs, but that's just an old OCD habit of mine.

---

I'm curious if comb filtering is less of an issue overall since I'm not including tweeters, but please tell me otherwise and let me know if one of the images above is the most sonically advantageous. I'm partial to I or J (the last two images if possible.)

I'm looking for the opinons of the wise, professional Yoda's of speaker design: the @agedhorse s and the @DukeLeJeune s if possible. Anyone lurking here who actually knows the maths.

(Or should I just buy two Hathor 1506s and be done with it.)

Thank you in advance for your wisdom and generostiy!

 

[agedhorse]

IMO, not as big of an issue as it’s often made out to be.

It’s part of the trade-off equation, something else may be more important.

 

[crazycloud]

I'm curious if comb filtering is less of an issue overall since I'm not including tweeters

No, it will still be an issue as comb filtering happens when 2 drivers in the same frequency band are physically separated.

Agedhorse posted as I was typing, and I generally agree - if it wasn't the case all 2x, 4x etc MI speaker designs would be very problematic. I don't like the effect and wouldn't design this way, but it's not like punters in the bar you're playing in will be throwing things at you and shouting that your stack sucks because of the comb filtering.

 

[6thplanet]

Agreed, in general, not an issue. My opinion, I'd do F or J, just to have the mids as high up as possible.

 

[jmlee]

I do have two Fearless 112 cabs which were built mirrored. I've tried playing through the two of them in mirrored configuration as per the photo below and also with the top one inverted so both mids/tweeters are at the top of the cabs--although at left and right, not vertically aligned. To me, the mirrored version seemed more focussed as I walked away from the rig. I've not done this on a gig, but the experiment tried before a couple of gigs was pretty convincing.

 

[fu22ba55]

To me, the mirrored version seemed more focussed as I walked away from the rig. I've not done this on a gig, but the experiment tried before a couple of gigs was pretty convincing.

This is great, Thank you for real-world experience. "Sounds better the further away you get." But I'm also thinking of balancing this advantage with the advantage mentioned by @6thplanet : having the mid drivers closer to my head:

I'd do F or J, just to have the mids as high up as possible.

Agreed. This is also a consideration. Having the two 6" drivers pointing at my knees is great for the audience, but not great for me.

My last Fearful was two 15Subs and a 2x6 headcase. Best sounding cabs I've ever heard, and the mids blew my head off (in the most positive way possible).

I think I'll go ahead and mirror one of the baffles, so I can choose between the two configs below:

 

[Passinwind]

This is great, Thank you for real-world experience. "Sounds better the further away you get." But I'm also thinking of balancing this advantage with the advantage mentioned by @6thplanet : having the mid drivers closer to my head:



Agreed. This is also a consideration. Having the two 6" drivers pointing at my knees is great for the audience, but not great for me.

My last Fearful was two 15Subs and a 2x6 headcase. Best sounding cabs I've ever heard, and the mids blew my head off (in the most positive way possible).

View attachment 7434196

I think I'll go ahead and mirror one of the baffles, so I can choose between the two configs below:

View attachment 7434197

View attachment 7434198

Good call, IMO. I built a number of 15/6/1 cabs using a variety of head cases and therefore was able to do a deeper dive into how different configs behave. My favorite ended up being two fifteens in landscape mode, ports vertically aligned, with one horn loaded 6/1 cab on top. Woofers closely spaced, highs at head level, closer to a point source feel, and horn loading the 18Sound six kills the gank and balances a single driver with two fifteens quite nicely. I was using slightly more uptown fifteens though, although not really too different sounding than KL3015LF.

Best of luck with your builds!

 

[fu22ba55]

My favorite ended up being two fifteens in portrait mode, ports vertically aligned, with one horn loaded 6/1 cab on top. Woofers closely spaced, highs at head level...

Passwind, can you draw a napkin sketch of the orientation you're describing? I can't picture it. If I build the standard (narrow and tall, non-tube) portrait mode 15/6 and vertically align the ports, then I would get image A or B above?

I didn't include options C & D below (portrait mode 15/6 with mirrored baffles) because I can't fit an SVT on top. But is this closer to what you meant? Or did you mean A or B above?

C is probably what Greenboy intended, and is closer to @jmlee s photo above, except the 12/6 cabs have the advantage of having the ports vertically aligned. I'd still have to go with A or B above to fit a 24" tube head.

 

[Passinwind]

Passwind, can you draw a napkin sketch of the orientation you're describing? I can't picture it. If I build the standard (narrow and tall, non-tube) portrait mode 15/6 and vertically align the ports, then I would get image A or B above?

I didn't include options C & D below (portrait mode 15/6 with mirrored baffles) because I can't fit an SVT on top. But is this closer to what you meant? Or did you mean A or B above?

C is probably what Greenboy intended, and is closer to @jmlee s photo above, except the 12/6 cabs have the advantage of having the ports vertically aligned. I'd still have to go with A or B above to fit a 24" tube head.

View attachment 7434228

View attachment 7434231

Like so, I meant landscape not portrait (now edited above):

The mid horn is a JBL replacement part for PRX cabs, I still have a spare one. Pretty big PITA to build with though, I had to fabricate a pretty janky mounting solution. I always ended up just using one of the 115 cabs, the BMS drivers are rated at 600VRMS each and the cabs are 3/4” Europly, so quite heavy. My favorite cabs ever though, if I didn’t have to carry them.

Out of your options I’d probably prefer B, but I like your old headcase format better. I sometimes used these two top boxes rotated and stacked on the two fifteens as pictured above, biamped. It got loud!

 

[Wasnex]

Generally we assume that reducing driver spacing is better.

The sound waves from a pair of drivers will interact across the axis of alignment. You could call it comb filtering or lobing; just different ways of describing the same thing.

In the context of lobing:

The size of the primary lobe and the number and angle of secondary lobes depends on the ratio of driver spacing (center to center distance) compared to wavelength (Lamda). With a speaker array, wavelength changes inversely with frequency but driver spacing is fixed. So the number and angle of lobes will vary by frequency.

(Lambda)

is used as the symbol for wavelength

Example for one specific frequency and driver spacing:

1125hz has a wavelength (Lambda) of about 12". If the center to center driver spacing is also equal to 12" you get this pattern across the axis of driver alignment.

A pair of 10s will probably have center to center spacing of about 12". So a pair of arrayed 10s will produce a lobe pattern similar to this at 1125hz. Wavelenght of 1125hz is ~12".

You would get the same pattern at 1125hz with a pair of 6's spaced at 12"

Mirroring the baffles does a couple of things.

• The distance between drivers is reduced.
  • This will cause the lobe pattern in our example to occur at a higher frequency.
• The mid drivers are aligned vertically.
  • This will narrow vertical dispersion and cause comb filtering/lobing across the vertical axis.
  • Horizontal dispersion will be the same as for a single driver.
  • Since are ears are aligned horizontally, you won't hear phasing. However, perceived tonal balance will vary depending where your ears in relation to the lobes in the vertical dispersion

Notice one of the secondary lobes is shooting straight at the ear. So he will hear this frequency.


But if he moves here, his ear is in the null between lobes.

The 18 Sound 6NM410 is a square frame driver and let's assume we actually can make the center to center spacing 6" (for the sake of our example).

Since we reduced drive spacing from 12" to 6", the same pattern will now occur an octave higher at 2250hz where wavelength is also 6".

However if we keep driver pacing at 12" you would get this lobe pattern at 2250hz:

Spacing = 12"
Wavelength/Lamda = 6"

 

[agedhorse]

Aside from the math, it's generally nowhere near as big of a deal as some folks make it out to be. Otherwise the classic 810 format would be a total bust!

 

[fu22ba55]

Generally we assume that reducing driver spacing is better.

Thanks Wasnex. I read your post more than once, and some of the maths actually did make it into my feeble caveman brain. This is why I purchase speaker cabs and don't design them.

@Wasnex please correct me if necessary in the conclusions I'm drawing from your post:

a) spacing drivers closer together produces LESS nulls between lobes, and the nulls/lobes produced occur higher frequencies
b) spacing drivers further apart does TWO things: produces MORE nulls and the nulls / lobes impact LOWER frequencies the further apart the drivers get?
c) the smaller the diameter of the driver, the less real-world wiggle room I have when spacing the drivers? (I've got more margin of error in real-world inches between two 15s than two 6s, simply because the 15s are larger?)

Are those three statements roughly correct?

So given the two images below: I (eye) and Iv2 (eye v2), are the following statements also true:

d) Option I (eye) is better since the mid drivers are closer together, and any nulls/lobing would occur at much higher frequencies than at Iv2 (eye v2)

e) The 3015LFs are actually 2" further apart in I (eye) than in Iv2, so technically, mathematically, there would be more nulls / lobing (for the 15s) but this is less of a concern because of the diameter of the drivers?

Is there any point where the drivers are far enough apart that the negative effects are mitigated?

So in Iv2... are the 6" drivers SO far apart that the effect is nullified, because the lobing would occur at frequencies LOWER than what the 6" driver can produce, because of the crossover?

 

[agedhorse]

Decreasing spacing increases the frequency that the nulls occur.

 

[MikeBass]

I’m gonna get dragged for this…..

Unless you’re playing in a very controlled room EVERY time, it doesn’t matter near as much as you’d like to believe.
Hell, the humidity changes and things sound completely different….

Stop over thinking it.

 

[DJ Bebop]

I’m gonna get dragged for this…..

Unless you’re playing in a very controlled room EVERY time, it doesn’t matter near as much as you’d like to believe.
Hell, the humidity changes and things sound completely different….

Stop over thinking it.

I remember reading that Phil Lesh supposedly had, at one time, an equaliser or something in his system that would compensate for changes in humidity to keep his sound consistent.

 

[agedhorse]

I remember reading that Phil Lesh supposedly had, at one time, an equaliser or something in his system that would compensate for changes in humidity to keep his sound consistent.

This was part of his Meyer controller, but it really didn't do anything meaningful for bass since it's primary application was for longer throw full range applications.

It's in all of the Galaxy controllers, but IIRC there were other models where it was added as a custom option. It compensated for HF losses due to temperature and humidity. Not perfect, but can be a helpful tool.

 

[DJ Bebop]

This was part of his Meyer controller, but it really didn't do anything meaningful for bass since it's primary application was for longer throw full range applications.

It's in all of the Galaxy controllers, but IIRC there were other models where it was added as a custom option. It compensated for HF losses due to temperature and humidity. Not perfect, but can be a helpful tool.

Thanks for confirming my faulty memory

 

[Wasnex]

Stop over thinking it.

IME the difference tends to be subtle but has an audible impact on performance. I am not aware of any reason he should not build a mirrored set and the time to think about is now; before the flatpacks are ordered.

 

[fu22ba55]

the time to think about is now; before the flatpacks are ordered.

This is 100% the point of this thread. Do I bother asking Leland @ speakerhardware to mirror one of the baffles for me. If he's down to do it, and it's not a hassle, I'll 100% mirror one of the baffles.

 

[gerryjazzman]

I have built 3 Fearful cabinets, a 15/6/1 and two 12/6/1 cubes with mirrored baffles so I could stack a 12 on top of the 15 or both 12's together. To tell you the truth, I never really cared for the sound of the stacked cabinets, I liked the sound of an individual cabinet with the mid/tweeter on top. Was always something weird about the sound. I thing it had to do with the fact that all the drivers were not aligned on the same axis which is done in the standard D'Appolito configuration used in some HiFi speaker systems. The Fearful layout doesn't work quite right with it's offset drivers.

Anyway, with that said, you could see if Leland will do mirrored baffles for you and go with that. I would consider wiring the crossover to optionally switch off the mid driver. That 18sound midrange is a screamer with somewhere around 102-103 dB sensitivity in the passband used in the Fearful. The 3015LF is a bit less, around 98 dB midband sensitivity up to 500-600 Hz where I believe the Fearful crossover is designed to crossover (I don't recall exactly, it's been years). You could stack the top cabinet with the midrange high (which is where I would prefer it) and the lower cab with the midrange turned off and avoid the whole scenario. Of course you have the option to try it mirrored as well and see what works better. In fact I think that was basically an option to run a Fearful 15 sub (3015LF only) with a Fearful 15/6 if you didn't need crazy midrange.

The standard Cheap but good 15/6 crossover actually included a switchable attenuator for the midrange in case it was too loud. If you go the route to switch off the midrange, be careful to put the switch between the jack and the INPUT of the midrange crossover section, not just switch the speaker which would leave an unloaded resonant circuit hanging on your amp output which would cause all kinds of havoc.