lilmike

You'll most likely get this built and fully tested before I get anything done with my projects. Thanks for sharing the renderings and concept.

I know. I've waited like two whole days already...

Oh you tease...

I'll add a second to Ricci's post. I typically measure big drivers at about a watt applied, so 2V for a 4R driver, 2.83V for an 8R driver. My DATS and WT2 aren't quite enough for the big drivers, the Clio Pocket does OK though. The Clio manual recommends the lowest drive level that yields a clean signal. The headphone out on the UCA202 is almost enough power, but the crosstalk killed it as a contender, Pity, it was widely available and only $30. I haven't tried the headphone outs on the other Behringer USB interfaces I have. Maybe one of these days... When I need an amp, I use an old single-channel amp I got from a buddy now, it is an old TOA. Overkill for the low level stuff and it weighs WAY too much to move around, but it was free and does the job. Be sure to use a good meter and watch voltages when using an amp, it's way too easy to send more signal than the sound card can handle. Back to back Zener diodes and/or voltage dividers are a good idea. Yeah - my last post was ambitious... It has been damn near 6 months and I still haven't had the time to pull final specs and design final boxes. Worse yet, I am rapidly working my way through my stash of plywood building other projects.

Without accurate driver specs and Thiele-Small parameters there is no way anyone can design or predict anything.

I've done some rudimentary testing with other drivers and found that I could put a lot more power to them in the tapped horn than in a free-air situation. The driver I was working with (Shiva X2) would make mechanical noise (tapping) at really low frequencies at about 100 watts applied free-air. The same driver in a tapped horn was able to take 500 watts without any noises. Not totally scientific, and certainly not what I have planned for my future efforts, but at least it showed that there is some loading on the driver below the horn passband and at least some difference between the displacement predictions and actual performance in that case. I have also found that my tapped horns tend to measure a little flatter at higher output levels than they do at a watt. In the case of a ported box with an undersized port, the excursion increasing as the port compresses makes total sense to me. In the vast majority of cases the ported box is larger than a sealed box would be for the driver in question and the port essentially becomes a large leak once compression sets in and the port is no longer functioning as a proper resonator.

Sorry for the late reply, just getting back online after a vacation. The 4th order recommendation was based on simulations of the performance of lower-order filters. The lower-order filters do not provide sufficient protection from overexcursion below the horn's tune. Once at 4th order, things look OK, higher than 4th looks a bit safer, but as SME mentioned, higher-order filters come with other issues, like ringing and phase issues. That was all done in simulations, which are predictions of large-signal performance based on a set of small-signal parameters that do not scale linearly as signal increases. My experience is that excursion tends to not be as much of an issue as Hornresp predicts. Usually, there are warning signs before the bad noises happen, and so far, I've not killed any drivers in tapped horns (though, I don't set out to do so). I haven't had a chance to actually measure the excursion of a driver in a tapped horn yet, but I hope to be able to do that before too much longer. I do have most of the gear and software, just need to get it all configured. The highpass recommendation was my effort to try to make these cabinets as bulletproof as possible for other people. I would hate to have someone kill the driver in their newly-built sub while showing it off in a demo because I didn't recommend a highpass to protect the driver and minimize distortion. I run 20 Hz at 4th order in my system, and I definitely twist the volume knob to silly on occasion. The sub has moved itself out of the corner several times. So far, only warm voice coil aromas, no bad noises or other complaints from the sub. The rest of the system was running out of gas at that point, but I am working on that.

Thanks! Timing was just about perfect for my projects. I hope to get the sub designs finalized in the next couple weeks, then get some sawdust made before year end. Will drag out the measurement gear and get some impedance sweeps of my drivers as soon as I can, then see if the models still look OK. Neither driver has all that high of an inductance (Le/Re of 1.03 for the 15s and 0.47 for the 18s).

for reference... https://youtu.be/hcOwz99GS4c I'm right there with ya. Definitely have experienced the "fizz" with cars, even some proper subwoofers...

Admittedly, I oversimplified the statement regarding equal peak height above. The relationship between the driver parameters and the box tune dictate the relative size of the two peaks. What matters most is that the peaks you're measuring are a decent match to what your simulation software is predicting. As far as the grille spacing? Yeah, make sure things won't hit. X-mech plus a few mm is a great start, I'd look at the thickness of nominal pieces of plywood or other scrap stock. Make sure the grille does not rattle. Foam weatherstrip and lots of screws.

Yeah, that lower peak is a bit "soft", that generally means losses or leaks (or, as I learned, a really useless port). With a typical reflex box, we want the peaks to be about the same height.

Possibly. Looks like it is something where you could try it and see. Heck, just cut a piece and set it in place for testing. Won't be 100%, but will give you an idea.

PM me. I can help you work through it. They're pretty basic. Might be as simple as L and R channels swapped.

Three years.... Weak.... I am still rocking the OG MicroWrecker (since 2013)... No finish, not even sanded, still has big PL boogers on it.

Well, that sucks. Still, for what they cost and what they do, 6 years isn't too bad, mine's actually coming up on that point too. I'll be sad when it dies, but I've gotten my mileage out of it.

Ummmm, yeah, actually I'm pretty sure I do have some power on tap. Nowhere near that level though. DAYUM man, that's more wattage there than some tours run....

Sweet, this will be perfect. I can enjoy the movie without a sub and know that I am not missing a damn thing. (No subs yet because I am still trying to scrape the cash together to buy my drivers...)

Those lifting straps? I've tried them, they're OK for some things like couches and beds, but not something I would use on a sub. I have a little bit of experience moving big subs. We used an appliance dolly when we had to get the pair of LilWreckers down a flight of stairs and into a basement theater. Definitely took two of us, and I'm not a small guy. LilWreckers are ~32 cubic feet and about 300 lb if I recall. Google appliance dolly images You should be able to rent or borrow an appliance dolly without too much trouble. I moved the F-20, MicroWrecker, and a number of other unnamed tapped horns with my 2-wheel dolly, but I didn't have to climb too many stairs. No matter what, using some straps to secure the sub to the dolly is a fundamental. A second person can be a huge help too. Of course, if you have another stout person handy (looking at you, @Ukko Kari) , you just pick the cabinets up and carry them, like we did with the AlpineGeists.

Yeah, the Woofer Tester 2 is pretty nice. I am pleased with mine, I have used it for years. I shared my version of the REW impedance jig that the OP's gear is based on for cases like this where that sort of an investment may not be warranted. It might not be perfectly exact, but it is a measurement of the impedance of the drivers in the cabinet, which has a LOT of value, even with the crosstalk issue. It's pretty clear that the tune is closer to 30 than to 40, which was his target. Odds are, when he takes the tune up closer to 40, the crosstalk will be less severe and the results might look a little better. I have no direct experience with the specific gear he is using, I do have its little brother though (Xenyx 302USB). It's not used in my measurement rig due to the crosstalk. EDIT - my bad - OP is using a UCA202 for the USB input and source. Those definitely have the crosstalk issue we're seeing here. I have plenty of those and know the crosstalk issue very well.

Looks like the sound card has the dreaded Behringer crosstalk issue. There's a strong rising response below 20 Hz. That may mess with phase.

Nice work man, nice work. Very impressive results. Far better behavior in the impulse and waterfall than I expected. There you go, raising the damn bar again...

I'll second Ricci here, not sure that 2R load is a great idea. Sure, it is a ported cabinet, so the average impedance is much higher, but that minima at 40 Hz might cause issues. Have you measured impedance of the actual build yet?

That sucks. Glad nothing else got fried.

That's exactly where I am. I have no idea where they're going with this one.

I nearly grabbed it when I saw it on the shelf the other day. Sounds like I should be glad that I passed.