I theorize: "Vesuvius" (no, I will not be building this one)

[Wolf]

Just a teaser for now....
UM15-22 and an RSS460-PR in a 120ltr total bandpass yields F3 of 22/76Hz, and with 400W yields an output of about 110dB with no boost required, and Xmax is kept safe.

More later,
Wolf

[chrisn]

this for the Tempo? )))))

[isaeagle4031]

That was a name I was throwing around to use. This would be a beast!

[Wolf]

I figured the 12" might be called 'Krakatoa'... I've always liked that name...

Chris- I have been driving a Dodge Magnum since 2009, so that's a no.

Anyway, so here is the grit! The name came from the list of most devastating eruptions on record, Vesuvius being the top of the list when it devoured Pompei and Herculaneum. I figured this would be a devastating eruptor of sorts. When I saw @ibzdually thread about a bandpass with it, I had to see what it would take to make a bigger brother for the Overdrive 10. Being that the PR line only goes to 18" with the 460, I couldn't do a model with the UM18 and make this work. I felt the approximate idea of upscaling the box for a larger setup was solid, so I looked up the driver/PR dims and set out on a simulation.

The 15" driver will fit through the 18" through hole. The PR set the majority of the front chamber dimensions using cutout, Xmax, and basket depth for consideration. So I came up with an inside dim of 17" square, and the basket + Xmax came out to about 7.125" occupied. I went with 7.5" x 17" x 17", yielding a volume of 35.5 ltrs. Using the 3/4" thick MDF, this yields a 1.125" clearance to the front of the inner double baffle. Since the woofer will be recessed to the second of the inner baffle boards, the surround will only be about 9mm forward of the front inner baffle. The cone will be further back due to the dish shape, and therefore no calamity or collision should occur under operation.

Now that the volume of the PR chamber is approximated, I went to do a model for a 4th order ported-bandpass in Unibox. I usually set the ports as a normal vented box might be in terms of diameter so I don't mess up the graph with too many resonances. In this case I used 2 4" ports as an example. I used the driver's spec sheet information, and wired it as a series scheme for a 4 ohm load. I set the sealed box to about 3 ft^3, or 85 ltrs, and the vented to the preset 35.5 ltrs, hit 45 Hz for tuning, and BAM! it fell right into place. I really mean it too! I tried tweaking volume a smidge here or there, futzed with the tuning, and came right back to where I started in the end. The model said a loss of 0.8dB, so I'm within my usual range of -1 to +1.5dB in practice, and I felt it as a good compromise. It modeled at 110dB with this loss included using 400W, so I doubt that it will really be an issue in most cases. No boost circuitry is applied to the low-end of this model, and I would recommend a pro-amp at 4 ohms, or a plate with highpass below 20Hz to get the most out of this. Xmax is fine for the driver in this alignment too, so on to the next stage....

I modeled for the PR with the UM15 in a 35.5 ltr PR box in Unibox to check Xmax, and get an approximate about how much mass to add to achieve tuning. Using the theoretical specs of 500g Mms, it appears that adding 172g will achieve the required tuning of 45Hz. **I highly recommend running an impedance sweep to verify the Fb is actually 45 Hz if one were to build this. In the initial Overdrive project, the Fb required less added mass than what was calculated in theory to get the result intended. Since I will not be building this, I cannot verify the actual mass I use to get the required result.**

"But, the Fs of that PR is 14.7 Hz! How is it you have to add mass to get the tuning down to 45 Hz?", you might ask... When you put a driver in a smaller box, the Fb of the box goes up in reference to Fs, and the same thing happens with a PR. If the box will the size of my living room, the Fb would still be 14.7 or just a touch higher. Since I am IMMENSELY decreasing volume the PR will be in, it will climb to above the target Fb and require added mass to make it return to the Fb needed. In fact, if this chamber is not small enough to get the Fb high enough, then it will not be appropriate for the design goals. The lower the tuning, the lower the range of bandwidth in a bandpass can become, and this issue also likely will suffer in output.

For a 15" sub with 18" PR, this box is not very big at all! If you add bracing as you likely should, I recommend increasing the sealed volume for whatever you add. This includes the driver magnet, the braces, the amplifier used if in the box, and whatever else. The PR section really does not need bracing as small as it is. Both chambers are lined in the simulation. I recommend 1/2" to 3/4" open-cell foam around the perimeter of the front chamber inside, and the same for the rear.

And now the graphics that show what I modeled, as well as a diagram for the construction. I don't think I drew the PR to scale, but the rest should be. I am not showing the sealed box model I also did for verification of the UM15 in the box alone, but it is something I did just in case.

Later, and ask questions if you have them...
Wolf

[humphreyblowdart]

I figured the 12" might be called 'Krakatoa'... I've always liked that name...

I know I'm missing the point, but you should switch the names around. The eruption of Krakatoa was an order of magnitude greater than Vesuvius, with a Volcanic Explosivity Index of 6 vs. 5. That doesn't sound that impressive but it's a log scale.

[Wolf]

Krakatoa caused less damage, but was more explosive. If I were to rename Overdrive 10, it would be Kilauea; smaller profile, but still very active. These names fit the size vs destruction capabilities.

Later,
Wolf

[dst]

Krakatoa, isn't that east of Java?

[Wolf]

I'd have to go look it up again...
Don't know why I think about 'Stub a toe-a' when I do...
heehee,
Wolf

[daryl]

Hi Wolf!

Your excursion simulation using a vented alignment as a proxy will come out correct if you modify the specs of the driver so that Fs=Ftc of the driver in it's 85l sealed enclosure.

You would be reducing Cms (Vas) to about 62l while keeping Mms, Rms and Bl the same.

Also Qts=Qtc for the driver in the 85l sealed.

Then your driver and PR excursion curves will be correct at all frequencies.

The system output would simply be the PR spl curve.

There would be no need to do a separate bandpass simulation just ignore the driver spl and total spl curves.

At the rated 1600W program power for the UM15-22 you have 116db.

You should not be expecting a signal with full power below 20hz!

With 1600W input Xmax would not be exceeded until 31.5hz by the simulation that you have made.

Commercial subwoofer output is often rated at 63hz or so which is also not correct.

Normally I would use 40hz to rate a subs output and power requirement but a bandpass is a special case because the Helmholtz resonance is near that frequency and unrealistic high ratings would result.

Xmax seems to be entered into your simulation as 22mm but it's 19mm by PE's specs.

[djg]

[djg]

[Wolf]

Xmax seems to be entered into your simulation as 22mm but it's 19mm by PE's specs.

The 22mm is the 15% overshoot before distortion sets in. I set Unibox for this overdrive number.

Good info on the other points. Thanks, Daryl!
Wolf

[carlthess40]

Could you model this for this Dayton audio 12” ? RSS315HO-4 single 4ohm coil

[Wolf]

If you would like me to start another thread, I can help you with that. I have modeled a few of the other drivers from Dayton, but not that one, in a PR-bandpass just to see. These type of systems really take a certain set of qualities, and I was not getting the results I wanted from the RS210 drivers when I modeled them. I can see if it will work. I know the Ultimax 12 will likely work as well due to their type of specifications. Krakatoa will get modeled as well, and likely added to this thread because of the similarities. I'd like to keep other models separate though. I'll see when I get time to see if it'll work well or not.

Later,
Wolf

[carlthess40]

That would be great