Finally! Omni 12 Tall Boy build

High power and fidelity from an easy to build cabinet.
Post Reply
Message
Author
User avatar
Seth
Posts: 2733
Joined: Thu Sep 10, 2009 8:06 pm
Location: Sacramento, California, USA
Contact:

Re: Finally! Omni 12 Tall Boy build

#91 Post by Seth »

Grant Bunter wrote: Wed Sep 21, 2022 4:09 pm Very comprehensive Seth!
Your target curve:
You trot it out reasonably regularly.
I don't think I've seen you explain to anyone what cabs it's based on, how many cabs, (subs or no) and in what environment.

I wish there was a target curve panacea, but there isn't, so fill us in on the details!
Thanks Grant. I appreciate the kind words.
You're questions contain so many great underlying points. Very tactful way of bringing them up BTW :thumbsup:
It's really impossible to cover everything, every time. Please, feel free to pickup where I left off and speak to some of the caveats.
In 19 minutes, this video covers what would probably take either of us 2-3 hours to type up. Perhaps a good place to start.

Last edited by Seth on Wed Sep 21, 2022 6:30 pm, edited 2 times in total.
Build in process - 2 WH6, one Alpha 6a loaded, one PRV Audio 6MB250-NDY loaded

Two 2x6 shorty SLA Pro's
One T39, 16", 3012LF loaded
Tall AutoTuba, 20" wide, 2x 8" MCM 55-2421
TruckTuba, 8½" wide, 2x 8" MCM 55-2421

User avatar
Seth
Posts: 2733
Joined: Thu Sep 10, 2009 8:06 pm
Location: Sacramento, California, USA
Contact:

Re: Finally! Omni 12 Tall Boy build

#92 Post by Seth »

Build in process - 2 WH6, one Alpha 6a loaded, one PRV Audio 6MB250-NDY loaded

Two 2x6 shorty SLA Pro's
One T39, 16", 3012LF loaded
Tall AutoTuba, 20" wide, 2x 8" MCM 55-2421
TruckTuba, 8½" wide, 2x 8" MCM 55-2421

User avatar
Seth
Posts: 2733
Joined: Thu Sep 10, 2009 8:06 pm
Location: Sacramento, California, USA
Contact:

Re: Finally! Omni 12 Tall Boy build

#93 Post by Seth »

CarterKraft wrote: Wed Sep 21, 2022 4:20 pm ...I think I even follow what you are intending (Famous last words haha)...
All crossovers begin rolling off before their designated frequency and are already down 3-6dB at their designated cutoff frequency, depending on the type of filter selected. That's why we move them well out of the way when EQing the individual drivers at a reasonable volume, then move them back to their recommended values. We want the individual drivers response to be as flat as possible in their expected frequency range before the crossover filters are applied.

I like to use LR filters because they sum to 0. Other than the summation shown in the pic below, you can also see how far ahead of the crossover point the signal starts to roll off.

Image


But, they only sum to zero if they're in phase. So, if you apply a 400Hz LR crossover (any slope) to your mid and woofer, and your mid measures 84dB on it's own at 400Hz, and your woofer also measures 84dB on it's own at 400Hz, and then you play them together you should now have 90dB at 400Hz... If they're 100% in phase at that frequency. If it only measures, say 86, I can plug that information into a spreadsheet I created to identify the delay needed to bring them into phase. Or, you can just tinker with the delay until you achieve the highest combined output.


Screenshot (209).png


You can see in the spreadsheet/calculator that the "84dB+84db=86dB at 400Hz" scenario indicates that they would be 102° out of phase and require a delay of 0.71ms to bring them to 100% phase alignment at that frequency. If the same scenario occurred at the mid/tweeter 2.5kHz crossover frequency, it would require only 0.11ms correction.

What the calculator doesn't tell me is which speaker needs the delay. Just have to try it on one and measure. If it's real close to 90dB, great, done. If not, apply it to the other driver instead.

Once you get all those steps done, it gives you a blank canvas to start with for your final system tune with the pre crossover EQ.

In all reality, you could actually skip EQing each driver and just match the output levels, adjust for phase, reference your RTA, do all of your EQing on the pre crossover EQ, and just try to make the RTA look like whatever target response curve you like. Could probably even do without the phase alignment (If the combined output at the crossover frequency is down a few dB from it's perfect summation +6db output, could just switch the filter type from LR to Butterworth to pickup a few dB... or just make up for it with EQ). But, since I saw the individual post-crossover EQ's and phase alignment doohickeys in your systems signal flow, I went ahead and gave you the whole shebang.

I imagine the extra pride and knowledge gained by going through all the steps would be worth the time spent for some people. And not so much for others. Most of the time spent getting the individual drivers response as flat as possible, the whole thing will likely take the better part of 2 hours or more for a first timer.

Once you get to that place, there's some other options (DSP functions) I'd like to discuss. Different ways to maximize the peak volume potential and/or battery life. I don't want to put too much on the table at once, we'll chat about it later.
Last edited by Seth on Wed Sep 21, 2022 11:49 pm, edited 7 times in total.
Build in process - 2 WH6, one Alpha 6a loaded, one PRV Audio 6MB250-NDY loaded

Two 2x6 shorty SLA Pro's
One T39, 16", 3012LF loaded
Tall AutoTuba, 20" wide, 2x 8" MCM 55-2421
TruckTuba, 8½" wide, 2x 8" MCM 55-2421

User avatar
Seth
Posts: 2733
Joined: Thu Sep 10, 2009 8:06 pm
Location: Sacramento, California, USA
Contact:

Re: Finally! Omni 12 Tall Boy build

#94 Post by Seth »

CarterKraft wrote: Wed Sep 21, 2022 4:20 pm For a raw volts test I just need to play a 400hz tone and check the max voltage before clipping?
Disable all crossovers. I use 60Hz personally. Just my habit. Use a Digital Multimeter set to read AC volts. Then yes, check the max voltage just before clipping.

Volts x Volts ÷ Ohms = Watts
Build in process - 2 WH6, one Alpha 6a loaded, one PRV Audio 6MB250-NDY loaded

Two 2x6 shorty SLA Pro's
One T39, 16", 3012LF loaded
Tall AutoTuba, 20" wide, 2x 8" MCM 55-2421
TruckTuba, 8½" wide, 2x 8" MCM 55-2421

CarterKraft
Posts: 255
Joined: Thu Oct 22, 2020 1:46 pm
Location: DFW

Re: Finally! Omni 12 Tall Boy build

#95 Post by CarterKraft »

From last weekends quick fast demo. Basically no EQ, minimal tuning to ear for gains.

https://youtu.be/gIldcVx0g68
Weldon Carter

User avatar
Seth
Posts: 2733
Joined: Thu Sep 10, 2009 8:06 pm
Location: Sacramento, California, USA
Contact:

Re: Finally! Omni 12 Tall Boy build

#96 Post by Seth »

Build in process - 2 WH6, one Alpha 6a loaded, one PRV Audio 6MB250-NDY loaded

Two 2x6 shorty SLA Pro's
One T39, 16", 3012LF loaded
Tall AutoTuba, 20" wide, 2x 8" MCM 55-2421
TruckTuba, 8½" wide, 2x 8" MCM 55-2421

User avatar
Seth
Posts: 2733
Joined: Thu Sep 10, 2009 8:06 pm
Location: Sacramento, California, USA
Contact:

Re: Finally! Omni 12 Tall Boy build

#97 Post by Seth »

CarterKraft wrote: Fri Sep 23, 2022 5:00 pm From last weekends quick fast demo. Basically no EQ, minimal tuning to ear for gains.

https://youtu.be/gIldcVx0g68
That's great Weldon :thumbsup:

Mids and highs are amazing. Sounded like the lows were struggling a bit, a little distorted. Could have just been the camera.

So, how do you feel it compares to the DiamondBoxx XL2 so far? Volume, clarity, bass...
Build in process - 2 WH6, one Alpha 6a loaded, one PRV Audio 6MB250-NDY loaded

Two 2x6 shorty SLA Pro's
One T39, 16", 3012LF loaded
Tall AutoTuba, 20" wide, 2x 8" MCM 55-2421
TruckTuba, 8½" wide, 2x 8" MCM 55-2421

CarterKraft
Posts: 255
Joined: Thu Oct 22, 2020 1:46 pm
Location: DFW

Re: Finally! Omni 12 Tall Boy build

#98 Post by CarterKraft »

The mid/top is amazing IMO.

The bottom does seem to fall off at that distance.
The one caveat is this was with the Dynamic bass filter in and ON but no idea what the 4 fields that contained parameters are.
It sounded really good at listening volume in my shop but it might be cutting output at higher volume.

My son and I are both really pleased so far.
I can't comment in regard to the Diamond Boxx just yet.
I haven't heard it other than one time for a minute or two.
I hope to do a AB comparison of them.
Weldon Carter

CarterKraft
Posts: 255
Joined: Thu Oct 22, 2020 1:46 pm
Location: DFW

Re: Finally! Omni 12 Tall Boy build

#99 Post by CarterKraft »

This explains the DB pretty good.
I'll have to mess with it after EQ/gain etc.

https://wiki.analog.com/resources/tools ... cbassboost
Weldon Carter

CarterKraft
Posts: 255
Joined: Thu Oct 22, 2020 1:46 pm
Location: DFW

Re: Finally! Omni 12 Tall Boy build

#100 Post by CarterKraft »

One last thing, I noticed this Sunday and didn't really understand it but it appears if I had a response file for the drivers it can run a Auto EQ.

Drinking from the fire hose...

https://wiki.analog.com/resources/tools ... cspeakereq
Weldon Carter

User avatar
Bill Fitzmaurice
Site Admin
Posts: 28620
Joined: Tue May 02, 2006 5:59 pm

Re: Finally! Omni 12 Tall Boy build

#101 Post by Bill Fitzmaurice »

Outdoors the lows aren't going to keep up as well as indoors, as they have no boundary reinforcement. With full PA it's not unusual to need twice the subs compared to indoors.

User avatar
Seth
Posts: 2733
Joined: Thu Sep 10, 2009 8:06 pm
Location: Sacramento, California, USA
Contact:

Re: Finally! Omni 12 Tall Boy build

#102 Post by Seth »

It seems like you're gonna have to choose between full bodied sound quality, maximum output, or a compromise between the two... or have multiple tunes that you can load depending on where you're using it. It would be great if you could just put a switch on it somewhere for that function, a tune switch. Me personally, if it were mine, I'd much rather it sound amazing with fat, smooth, luscious bass at a lower overall max potential volume than have it be amazingly loud and less full. It seems most of the Bluetooth speakers on the market today also choose SQ over SPL (more than likely, a balanced combination of both). They also apply a limiter. So, even at the maximum volume setting, they never go into distortion. That's my observation anyway. The target response curve I posted will give you a full sound, but will limit the maximum attainable overall volume before distortion occurs from the woofer amp channel running out of steam when there's more power left for the mid and high to continue increasing volume cleanly. (that's when a limiter is handy, keep it clean)

Which brings me to the next thing I wanted to discuss. I intended to wait until after it was tuned. But, since it's come up now...



The drivers used in the chart above aren't the exact combination you have, but the overall shape of the curve should be very similar.

The reason the overall output is/would be limited is; to achieve the curve I posted, the 50Hz region needs to be about 10dB above the average response from 100Hz to 16kHz. But, the speakers native response (chart above) at 50Hz is about 8dB below the 100Hz to 16kHz average response. Meaning 50Hz will require 18dB more volts/power than 100-16k when tuned to the target curve. The inverse is also true, 100-16k will require 18dB less volts/power than 50Hz. So, if given a signal with equal amplitude, and you have ±150 watts available for the woofer, the mid and tweeter are only going to require the equivalent of about 2.5w each. Doesn't seem all that impressive, leaving all that on the table, does it? Don't worry, music isn't equal amplitude and it should be plenty loud before distortion. But, what if you want more output? What if you want to take better advantage of the power available to the mid and tweeter and not sacrifice the bottom end to do so?

Psychoacoustic Bass tricks your brain into thinking it hears something that it doesn't actually hear by creating harmonics of a lower tone. Which could possibly double the overall perceived volume of your project speaker by removing the need to produce the signals that take a lot of power to produce and replacing them with harmonics in the portion of the spectrum that's more efficient. Maybe a little more than double perceived volume, up to 15dB, depending on how much of the bottom end you choose to keep natural.
Screenshot (210)2.jpg

This is your ticket to the best of both worlds. It's also the same trick tiny Bluetooth speakers use to make them sound like they have amazing bass for such a small speaker.

I've never used it, but I understand it and it should work very well. I plan to live through your experience and I'm curious to hear your personal impression.

The dynamic bass will be good to use too. But, it will (should) be just about the last thing you setup, after everything else is tuned and set. The Psychoacoustic Bass (SuperBass) plugin too, one of the last things to setup. Get everything EQ'd and sounding good first. Then we can start fiddling with the add-ons and enhancements.

The auto EQ thing... Sounds intriguing, but I don't have it in me to soak it all up and make sense of everything presented. Pretty damn cool though, the amazing stuff these DSP chips are capable of, isn't it?
Build in process - 2 WH6, one Alpha 6a loaded, one PRV Audio 6MB250-NDY loaded

Two 2x6 shorty SLA Pro's
One T39, 16", 3012LF loaded
Tall AutoTuba, 20" wide, 2x 8" MCM 55-2421
TruckTuba, 8½" wide, 2x 8" MCM 55-2421

User avatar
Seth
Posts: 2733
Joined: Thu Sep 10, 2009 8:06 pm
Location: Sacramento, California, USA
Contact:

Re: Finally! Omni 12 Tall Boy build

#103 Post by Seth »

CarterKraft wrote: Wed Sep 21, 2022 1:10 pm ...The Dynamic bass fattens up the bottom nicely for easy listening...
CarterKraft wrote: Fri Sep 23, 2022 6:30 pm This explains the DB pretty good.
I'll have to mess with it after EQ/gain etc.

https://wiki.analog.com/resources/tools ... cbassboost
It sounds like you understand it, but I wanted to touch on this real quick. If not you, someone will get something out of it.

The purpose of dynamic bass is to gradually increase the bass for lower and lower volumes. At near maximum volume the dynamic bass should have zero effect. But, at the lowest volumes it has the greatest effect. And here's why...

Equal-loudness contour


Image


The red lines indicate our perception of sound, measured in phons. Even though the span of decibel level varies quite a bit along those lines, we perceive all the frequencies at the indicated SPL as the same volume. Notice how they flatten out with increased volume.

At all volumes, our ears are less sensitive to low frequencies than mid and high frequencies. But, it's a gradient. At lower volumes, the lower frequencies need to be increased for music to sound the same as it does when it's turned all the way up.

That's why Dynamic Bass Boost one of the last things to set. Get the system to sound amazing at full output, then apply the Dynamic Bass to maintain that phenomenal experience and perceived frequency balance at lower volumes too.
Build in process - 2 WH6, one Alpha 6a loaded, one PRV Audio 6MB250-NDY loaded

Two 2x6 shorty SLA Pro's
One T39, 16", 3012LF loaded
Tall AutoTuba, 20" wide, 2x 8" MCM 55-2421
TruckTuba, 8½" wide, 2x 8" MCM 55-2421

Grant Bunter
Posts: 6915
Joined: Mon Nov 28, 2011 11:12 am
Location: Ilfracombe Queensland Australia
Contact:

Re: Finally! Omni 12 Tall Boy build

#104 Post by Grant Bunter »

Seth has been giving you heaps of info, what he thinks, and how to get there.
Mostly, I agree.

Having said that, in most subwoofer plans from Bill, the advice, when checking phase response at a crossover frequency to tops, is to "flip polarity" on the subs, and if doing so makes the subs louder, then leave it that way.
Polarity switching will change phase response by 180 degrees. And it makes sense to do that with subs, because the sub frequencies are not trying to be played by any other driver in the system or cab.
Seth talked about changing phase in the mids, using, in his example, 120 degrees via his calculator.
The issue with changing phase response by 120 degrees (or any amount) is that the phase response is altered throughout the entire mids band pass, not just the crossover point.
Since phase response is frequency dependent, that may introduce another requirement for alteration in the next filter.
Other than that, phase response between drivers should already have been identified and dealt with by the cab designer, ie Bill.
I think it's better to do that to subs or woofer only.

Also, be wary of altering gain in a particular bandpass, as that alters the acoustic crossover point (as distinct from the electrical one) meaning that testing phase response at the crossover point may no longer be accurate.
Example: set a 100Hz crossover point from woofer to mids (Electrical crossover point). Mids seem not as loud so you raise the gain by +6dB. Acoustical crossover point is now
err, I'll ballpark it at 93Hz. Which point do you test?

It may well turn out that a simpler method is to leave mids and HF band passes alone, and simply invert the woofer only, and see if it becomes louder.

I guess, comparison is your friend here. Save your confguration.
Start a new one that has simple 400Hz and 2.5Khz LR 24 for each high pass or low pass. Play.
Invert the sub. Play.
Recall your saved configuration, play.
Which one do you like the most overall?
Built:
DR 250: x 2 melded array, 2x CD horn, March 2012 plans.
T39's: 4 x 20" KL3010LF , 2 x 28" 3012LF.
WH8: x 6 with melded array wired series/parallel.
Bunter's Audio and Lighting "like"s would be most appreciated...

CarterKraft
Posts: 255
Joined: Thu Oct 22, 2020 1:46 pm
Location: DFW

Re: Finally! Omni 12 Tall Boy build

#105 Post by CarterKraft »

Thanks to both of you guys to share your knowledge and testing programs to help me dial it in..

Oddly enough Seth I think I can select crossover profiles with a physical Pot, that might ge useful for a SQ/SPL mode. Something easy enough to implement once it's tuned.

And to Grant's point phase testing should be easy to do as well, with the check box's.

Do you guys recommend a "test album" of your most common test tracks?

Edit: I'm hoping to just be able to purchase via download etc. an album of the most common test tracks used.

Double edit: I see there are tone generator apps in the play store, maybe that's a simple way to acquire the required tones?
Weldon Carter

Post Reply