".5 Aligned" 4x10 cabs

[DukeLeJeune]

Doesn't this make for a darker cab now? Seems like one really needs mids to cut through, but now we have just half of everything from 475 and up.

Yes, darker overall but more uniform across the horizontal plane: Not only do we get the better dispersion of a single 10" woofer up high, but we transition to two 10" woofers side-by-side down low, so the blooming of the radiation pattern (in the horizontal) as we go down in frequency is slower, by about one octave.

Hopefully EQ can restore the tonal desired balance. Now if one were to build this type of system from scratch, woofers with a rising native response might make a lot of sense. For instance the Deltalite 10 looks like a good candidate, because of its rising response north of 1 kHz or so.

One thing I don't yet know is whether the phase rotation of the lowpassed woofers causes any significant response anomalies. Even if it does, the .5 alignment may be a worthwhile improvement is a lot of cases.

 

[Downunderwonder]

All of which is unnecessary if you run your 4x10 through a bandsaw and turn it into a pair of 2x10's which you stack vertically.

 

[JGR]

So after a couple days of mulling this over and reading the links posted, I've got this visualized in my head. All that's happening is that one vertical row of drivers has the highs rolled off in the 475Hz range. Doesn't this make for a darker cab now? Seems like one really needs mids to cut through, but now we have just half of everything from 475 and up. Yah, I get that it minimizes the comb filtering between the side by side drivers, now making for a more technically correct situation, but isn't that like cutting off ones nose to spite your face?

It's important to select the right drivers just like any design. If you pick something that has a rising sensitivity starting the the midrange, then you end up with a very balanced sounding cab.

Edit: just saw That Duke hit on this point above.

 

[Rick James]

Yes, I got alot of information for BFM! And most of it helped.

I asked him to comment on this thread. An excellent answer as always.
http://billfitzmaurice.info/forum/viewtopic.php?f=7&t=22368

 

[will33]

Sure miss Bill around here. I learn a ton from that guy.

I assume the inductor should still have equal or greater power rating than the sum of the drivers it's attached to. Would moving it north an octave probably allow the use of an air core, at least in a 410, reducing the saturation issue agedhorse was describing?

I like the idea of passing higher 12db rather than lower at 6.

 

[Coolhandjjl]

One thing I don't yet know is whether the phase rotation of the lowpassed woofers causes any significant response anomalies. Even if it does, the .5 alignment may be a worthwhile improvement is a lot of cases.

I assumed the woofers would be wired O-O-P to make up for that 1st order filter. Am I wrong here?

When you EQ the mids up they get to many more of your audience instead of just the ones on axis that is the hallmark of side by side tens.

Some heads, especially those with Fender style tone stacks may already be maxed on the mids in normal operation. Lots of folks use the Empress Para EQ pedal, but that's just having to add one more thing on the chain, risking adding noise and other unwanted junk in the signal chain.

So I do see where if building one from scratch selecting a driver with rising response is the way to go rather than modding an existing cab.

 

[will33]

I'm having a problem deciphering "O-O-P".

How would changing the woofer wiring make a steeper filter?

I'd think it would just change impedance, and thus part values, which could help with part cost.

 

[Rick James]

Sure miss Bill around here. I learn a ton from that guy.
I assume the inductor should still have equal or greater power rating than the sum of the drivers it's attached to. Would moving it north an octave probably allow the use of an air core, at least in a 410, reducing the saturation issue agedhorse was describing?

In all of the plans I have from him he never recommends air core inductors for low pass, always cored, with the main requirement being low DCR. From what I gather cored inductors have less problems with saturation than air core, and some testing I saw posted at the Madisound site backed that up. You'd have to ask him to get a better answer. He doesn't come here anymore but you can ask him on his forum.

 

[Coolhandjjl]

I'm having a problem deciphering "O-O-P".....

Out Of Phase

How would changing the woofer wiring make a steeper filter?

Adding a cap to the inductor would make it a 2nd order filter, now a 12dB/oct roll off. Is that what you were referring to?

 

[Fuzzbass]

Well, yes this is a definate possibility. There is definately a trade-off between brightness and uniformity, and some folks prefer one, some folks prefer the other. Pick what YOU prefer, what makes YOU happy, you are the one who needs to be satisfied with the results.

Welcome to my world... designing products for a wide range of players. Each player has personal preferences, and these preferences conflict. That's why there are so many choices out there for players. If something I design doesn't float a player's boat, there's plenty of alternative choices. The opposite is true as well.

It's not unlike when mixing FOH. I have to mix to make the majority of paying audience happy while pissing off the least number of folks who have "different mix tastes". It's the art behind the science, something that 30 years behind a console (my "other" night job) helps develop. The way I incorporate compromises in the mix, and the ability to prioritize what's really important and what's not each night is why I work so much. There is no perfect mix regardless of what some folks might profess. There is a range of acceptable mix parameters. Using those within the challenges of the room and the PA (and of course the damn stage volume) is what separates the wannabes from the pros. The pro engineer goes for a 95% night and gets there almost all of the time. The wannabes go for a 100% night but rarely get to 50%.

IMO, the same applies to players... I would rather see consistent, solid 95% sets than a band that's all over the place. Most paying audience would as well.

Ummm.... that's way too complicated. All I want to know is what's the best cab for metal.

 

[will33]

Out Of Phase

Adding a cap to the inductor would make it a 2nd order filter, now a 12dB/oct roll off. Is that what you were referring to?

Yes.

I must've interpereted your post wrong or something. Thought you were describing some way of steepening the filter by changing the woofer wiring, not adding a cap.

 

[will33]

In all of the plans I have from him he never recommends air core inductors for low pass, always cored, with the main requirement being low DCR. From what I gather cored inductors have less problems with saturation than air core, and some testing I saw posted at the Madisound site backed that up. You'd have to ask him to get a better answer. He doesn't come here anymore but you can ask him on his forum.

His plans that I do have (Jack 12 & 15, DR200, Wedgehorn, Titan39) the lowpass inductors are iron-core for their DCR and power handling. Uses air-core in higher frequency filters.

When we get to the deep end of the pool with stuff like saturation or insertion loss, reactive loads, etc. I'm in over my head.

 

[seamonkey]

I saw the JBL service manuals had schematics on their quasi 3-way. They use two low pass filters one for eash sub. There is a phase shift as would be expected between the drivers at some frequencies. It's not bad at all. Better than I though.

 

[DukeLeJeune]

Would moving it north an octave probably allow the use of an air core, at least in a 410, reducing the saturation issue agedhorse was describing?

Moving the rolloff frequency north won't change the amount of current that goes through the inductor, but it would allow us to use a lower value (which would cost less) of the same gauge. I'm under the impression that cored inductors resist saturation a bit longer than air core types, but that air cores saturate more gradually so they can handle peaks better. I always use air cores, largely because that's what my mentor (Earl Geddes) felt was best.

I assumed the woofers would be wired O-O-P to make up for that 1st order filter. Am I wrong here?

We'd definitely want the woofers to be in-phase down low, so that limits us to rolloff slopes that work well in-phase. Since the interaction we're looking at is between two drivers running acoustically "side by side", rather than "one above and one below" as with a normal two-way configuration, the phase relationship will be different from what we're used to.

Adding a cap to the inductor would make it a 2nd order filter, now a 12dB/oct roll off.

I think that might increase the ripple in the response as the lowpassed driver rolls in and out of phase, but won't know until I roll up my sleeves and get into it. Hopefully this weekend.

Incidentally, in this application adding the Zobel circuit I talked about makes the lowpass section "look like" a damped second-order filter. Imagine a second-order filter, and add a resistor on the shunt (capacitor) leg. Juggle the values and it's a first order with a Zobel.

 

[agedhorse]

I asked him to comment on this thread. An excellent answer as always.
http://billfitzmaurice.info/forum/viewtopic.php?f=7&t=22368

"Sure. First, a .5 alignment is when two identical woofers are used, but one of them is low passed. One reason for doing this is baffle step compensation. Below where the baffle is 1 wavelength in dimension response rolls off, eventually settling in 6dB down from nominal. If you use two drivers with one low passed at the right frequency it will give back that 6dB only at the lower end of the spectrum where you need it for flat response. This is especially handy with pole mounted PA cabs that are run full range without subs, and that’s where you used to see this technique employed a lot. You don’t see it so much anymore since nowadays subs are the rule.

Another reason for doing this is the one noted in the thread, to broaden midrange dispersion and eliminate high frequency comb filtering when drivers are mounted side by side. Above where the two driver centers are more than 1 wavelength apart comb filtering causes high frequency response to change as you move across the sound field. Placing the drivers side by side also results in the midrange dispersion angle being more than halved compared to one driver. The cure for both those concerns is to low pass one driver, or in the case of a 4x10, 6x10 or 8x10 one of the vertical sets of drivers.

In the thread the poster low passed one vertical set of drivers in his 4x10 with a first order filter. A second order filter would work much better, and the added cost would be minimal, as the required capacitor isn't expensive. A higher knee frequency could be used as well, lowering the cost of the inductor.

One potential problem is if the drivers are series/parallel wired in a 4, 6 or 8x cab. If you don't know how to wire the cab up correctly you can really make a mess of things.

Why .5 alignments aren't used by the electric bass cab industry in general comes down to cost, pure and simple. At the manufacturing level the cost of the filter is perhaps $15, but manufacturers hate to spend even one buck on anything that the consumer can't see."

Excellent? Maybe. Accurate? no so much.

How can this be especially handy on pole mount PA where there is NO pole mount PA that uses dual 15" drivers? It has never been used by the pro audio industry on pole mount PA spaeakers as almost all use a single LF driver, and I have never seen this technique used on dual 12's because they appear almost exclusively in pro oriented products with lower crossover points. The smaller diameter and closer spacing reduce any beaming effects to areas that are insignificant in the basic system's response senistivity.

I can say without a doubt, based on products that I have been involved with, that cost is not the issue. The challenge is that the tradeoff to most players is that the benefit does not outweigh the drawback of feel, and when you ADD IN the additional cost, it just reinforces the already clear decision thatthe majority of the marketplace has made.

I do agree, however, that with the appropriate choice of drivers, a very good product can be designed with this technique. A product that might appeal to some players more than others. there is no perfect approach, this is but one possible approach that might appeal to a player.

 

[Coolhandjjl]

....We'd definitely want the woofers to be in-phase down low, so that limits us to rolloff slopes that work well in-phase......

Bad terminology on my part. I should have wrote something like "should the low passed woofers be wired in reverse polarity to keep them 'in-phase' if using just an inductor."

 

[agedhorse]

Bad terminology on my part. I should have wrote something like "should the low passed woofers be wired in reverse polarity to keep them 'in-phase' if using just an inductor."

shouldn't make any difference as the result will be (in theory) a net 90 degree phase shift.

The problem is that this assumption does not take into account any electro-acoustic contributions, so in practice you will meet to try both to see if one way works better than another. May not matter one bit in practice.

 

[Coolhandjjl]

My bad...90 degrees, your right. I'm getting old. Need that cap to get the full 180 for a reverse polarity wiring.

 

[DukeLeJeune]

My bad...90 degrees, your right. I'm getting old. Need that cap to get the full 180 for a reverse polarity wiring.

Actually even with the cap in place I don't think the phase of the low-passed driver would rotate all the way to 180 degrees until you got north of the "crossover" frequency. So you'd still want the drivers all wired in-phase, otherwise they'd cancel one another out down low.

I checked on what happens in the phase domain with and without the Zobel. I think you're arguably slightly better off without it, as this is an application where we want to minimize phase rotation.

 

[agedhorse]

Actually even with the cap in place I don't think the phase of the low-passed driver would rotate all the way to 180 degrees until you got north of the "crossover" frequency. So you'd still want the drivers all wired in-phase, otherwise they'd cancel one another out down low.

I checked on what happens in the phase domain with and without the Zobel. I think you're arguably slightly better off without it, as this is an application where we want to minimize phase rotation.

Agreed.