-

Jump to content



Photo
- - - - -

-> Maelstrom/Tempest analysis!


This topic has been archived. This means that you cannot reply to this topic.
16 replies to this topic

#1 of 17 OFFLINE   DerrickW

DerrickW

    Stunt Coordinator

  • 177 posts
  • Join Date: Jan 26 2003

Posted April 20 2003 - 07:12 PM

My goal is to build a great subwoofer using one channel of my Parasound 1500A amp, which provides 315W at 4ohm. The Tumult and AV15 aren’t readily available, and seem to like more power. The Tempest is a forum favorite, but I think the Maelstrom may actually work better as you’ll see in this analysis using WinISD 0.50a6. The maelstrom is now $225 from Adire, and there is at least one example: http://www.acoustic-...tromsubpics.htm You can read the Adire vented white papers at http://www.adireaudi....plications.PDF You may also want to read the two recent threads on maelstroms, which don’t have any results posted.

My primary goal is sound quality. I will be moving into an apartment soon, so do not know the exact room dimensions (likely under 15x20 with several openings). I also shouldn’t upset the neighbors too much. I will probably use the system 70/30 music/movies. I will be using aura bass shakers in the couch, which may help the low bass feeling. My main speakers are -3dB at 28Hz, so for a lot of 2 channel music I may not even use the subwoofer (unless it sounds better than the Lsi15’s ported 8”). My HK525 has flexible triple crossover bass management, so after the sub is built I can explore this. I will be using a BFD to flatten response on the sub and shakers. Sonotubes are too weird looking to pass the WAF but gigantic boxes are ok.

All graphs are plotted with 315 watts at 2.5 meters (estimated seating position). The plotted designs are:

Tempest (blue)
Sealed Q= .707 Butterworth
122 Liters

Tempest (green)
Sealed Q= .577 Bessel
230 Liters

Tempest (black)
Vented: EBS alignment
340 Liters
tuned to 15.8Hz

Maelstrom (red)
Sealed Q= .577 Bessel
150 Liters

Maelstrom (orange)
Vented: EBS
320 Liters
tuned to 20Hz

Maelstrom (purple)
Vented: custom design
280 Liters
tuned to 17Hz

Maelstrom (light green)
Vented: custom design
340 Liters
tuned to 17Hz

Note that even though the Maelstrom driver is larger than the Tempest (18 vs 15”), the comparable Maelstrom designs aren’t always larger. Because of the design criteria mentioned, I’m guessing that EBS and Bessel designs will be best for tight “audiophile” sound.

#2 of 17 OFFLINE   DerrickW

DerrickW

    Stunt Coordinator

  • 177 posts
  • Join Date: Jan 26 2003

Posted April 20 2003 - 07:13 PM

These are the parameters if you want to verify:
Posted Image

#3 of 17 OFFLINE   DerrickW

DerrickW

    Stunt Coordinator

  • 177 posts
  • Join Date: Jan 26 2003

Posted April 20 2003 - 07:14 PM

Posted Image
The transfer function shows the following -3dB points:
Tempest .707 (blue) = 36.8Hz
Tempest .577 (green) = 36.8Hz
Tempest EBS (black) = 19.1Hz
Maelstrom .577 (red) = 57.1Hz
Maelstrom EBS (orange) = 42Hz
Maelstrom 280L/17Hz (purple) = 49Hz
Maelstrom 340L/17Hz (light green) = 50Hz

These numbers seem to suggest that the Maelstrom will have poor extension, but after looking at the further graphs and considering the use of a BFD to decrease higher frequencies I don’t think this is the whole story.

#4 of 17 OFFLINE   DerrickW

DerrickW

    Stunt Coordinator

  • 177 posts
  • Join Date: Jan 26 2003

Posted April 20 2003 - 07:16 PM

Posted Image
The transfer function phase plot is included to be complete, though I don’t know if anything is really helpful here.

#5 of 17 OFFLINE   DerrickW

DerrickW

    Stunt Coordinator

  • 177 posts
  • Join Date: Jan 26 2003

Posted April 20 2003 - 07:17 PM

Posted Image
The maximum power graph shows that all the Maelstrom designs can handle 315W above 20Hz, while none of the Tempest designs can! Uh oh.

#6 of 17 OFFLINE   DerrickW

DerrickW

    Stunt Coordinator

  • 177 posts
  • Join Date: Jan 26 2003

Posted April 20 2003 - 07:18 PM

Posted Image
Group delay is an indicator of the “tightness” of the subwoofer design. The sealed designs of course provide less delay below 30Hz than the ported designs. The EBS 17Hz tuned Maelstrom is the worst ported design (still very good as far as ported goes). The EBS 15.8 Tempest and the two 17hz Maelstroms are very close in performance. The sealed Maelstrom outperforms the two sealed Tempests.

#7 of 17 OFFLINE   DerrickW

DerrickW

    Stunt Coordinator

  • 177 posts
  • Join Date: Jan 26 2003

Posted April 20 2003 - 07:19 PM

Posted Image
This is the SPL graph. The EBS Maelstrom and 340/17Hz Maelstroms hit 104dB at 20Hz (2.5m distance). The 280/17 Maelstrom and EBS Tempest hit 103dB at 20Hz. The .707 Tempest hits 95dB, the .577 Maelstrom hits 96dB, and the .577 Tempest hits 97dB at 20Hz. This does not account for any room gain. Because I will be using a BFD to flatten out response, I think this is probably a better measure of low frequency extension than the -3dB points from the transfer function. The Maelstrom and Tempest are very close here, but there is more to the story. At higher frequencies the Maelstroms peak at 110dB and the Tempests peak at 106dB.

#8 of 17 OFFLINE   DerrickW

DerrickW

    Stunt Coordinator

  • 177 posts
  • Join Date: Jan 26 2003

Posted April 20 2003 - 07:19 PM

Posted Image
The cone excursion graph demonstrates p-p. The Tempest’s xmax of 16.4mm (32.8 p-p) is violated by all the Tempest designs. The Maelstrom’s xmax of 13mm (26 p-p) is never violated above 20Hz.

#9 of 17 OFFLINE   DerrickW

DerrickW

    Stunt Coordinator

  • 177 posts
  • Join Date: Jan 26 2003

Posted April 20 2003 - 07:20 PM

Posted Image
The max SPL takes into account both maximum power handling and maximum cone excursion. Therefore, the SPL graph above isn’t real. The individual curves above can never extend past these maximum SPL curves without damage to the woofer. By taking the limiting SPL of these two curves, here are my ‘adjusted’ SPL’s at 20Hz:
Tempest .707 (blue) = 94.5dB
Tempest .577 (green) = 94.5dB
Tempest EBS (black) = 103dB
Maelstrom .577 (red) = 96dB
Maelstrom EBS (orange) = 104dB
Maelstrom 280L/17Hz (purple) = 103dB
Maelstrom 340L/17Hz (light green) = 104dB

#10 of 17 OFFLINE   DerrickW

DerrickW

    Stunt Coordinator

  • 177 posts
  • Join Date: Jan 26 2003

Posted April 20 2003 - 07:21 PM

Posted Image
The impedance of the Maelstroms gets much higher, is this a problem?

#11 of 17 OFFLINE   DerrickW

DerrickW

    Stunt Coordinator

  • 177 posts
  • Join Date: Jan 26 2003

Posted April 20 2003 - 07:22 PM

Posted Image
The rear port air velocity graph is from some quickly made port designs. For the purple and green maelstrom, I made 1.5inx24in slot ports. The orange maelstrom is 2x24. The slot ports vary from 23-28in long, so should fit in these large enclosures which would be deep. Realistically these would need refining if I went with a ported maelstrom. For the black tempest, it has 2x 4” circular ports at 19.6in long.

#12 of 17 OFFLINE   DerrickW

DerrickW

    Stunt Coordinator

  • 177 posts
  • Join Date: Jan 26 2003

Posted April 20 2003 - 07:23 PM

Ok, so now I need some input. At this point it seems that the sealed .577 (155L) Maelstrom and the 280L/17Hz Maelstrom are the strongest candidates. I’m leaning towards the ported 280 Maelstrom over the 340 because of the similar performance and size difference.

1. Any errors in this analysis?

2. Is the high Maelstrom impedance an issue?

3. Do the graphs for cone excursion/maxspl/maxpower mean that these Tempest designs just won’t work with 315W of power?

4. What should I do for designing ports for the Maelstroms designs? The 280/17Hz Maelstrom with dual 4” ports shows 20.7in long with 18.5m/s at 20Hz. If I go 3 4” ports to decrease the m/s, suddenly the ports are 32.5in long, which may be harder to work around. If I go slotted, what do I need to know about avoiding walls, etc?

4. Which design would you pick?

Posted Image

#13 of 17 OFFLINE   DerrickW

DerrickW

    Stunt Coordinator

  • 177 posts
  • Join Date: Jan 26 2003

Posted April 21 2003 - 09:32 PM

Posted Image

#14 of 17 OFFLINE   Mike Keith

Mike Keith

    Second Unit

  • 324 posts
  • Join Date: Jan 24 2002

Posted April 22 2003 - 03:27 AM

I have 2 Maelstroms up and running right now, one in a 260L (gross) 24" sonotube, the other in a 340L (gross) Sonotube. Both are vented, and I've tried just about every combo of ports I can think of, multiple smaller ports just wont work with a 24" tube, and I dont think you could go much smaller than 24" with a Maelstrom. A 8" diameter port works best from my actual listening experiences, and in the smaller enclosure it requires a 33" long port which is longer than WinISD calculated. In fact all the actual port lengths needed to be a few inches longer than WinISD calculated, so keep that in mind.

For my preferences I like the smaller enclosure and definitely no lower tuning than 20Hz, I tried 18Hz and no likey compared to the 20Hz version. However if you go for a smaller (240-260L) version you will not be able to fit the port in most boxes.

#15 of 17 OFFLINE   DerrickW

DerrickW

    Stunt Coordinator

  • 177 posts
  • Join Date: Jan 26 2003

Posted April 22 2003 - 06:30 AM

Yup, I can't do a sonotube (WAF), but I still noticed the porting issues. I didn't research the PR or slot porting further as the Bessel sealed Maelstrom is just screaming at me "build me." I just pulled the trigger on a Maelstrom driver, and as of now I'm planning on the .577 design. The group delay is just so low, it is no wonder that the few people who have the Maelstrom have described it as being wonderful for music. The enclosure is also much smaller, which is good. Since I can't crank the volume up to extremely loud levels being in an apartment, I don't think I need the potential output of a ported design. Either way I'd be EQing towards a flat response, so the difference is really just in group delay and potential SPL, not the cutoff slope.

Remember that the 96dB at 20Hz I mentioned for the Bessel sealed Maelstrom is calculated at 2.5meters without any room gain. This should provide more than 100dB at the seating area, which should be plenty of headroom for reference listening.

#16 of 17 OFFLINE   Mike Keith

Mike Keith

    Second Unit

  • 324 posts
  • Join Date: Jan 24 2002

Posted April 22 2003 - 01:58 PM

The sealed should be very nice for music, and the smaller enclosure will definitely look better than a ugly Sonotube.

However you have to trade off a lot of FR on the low end with any sealed, I found that 255 L gave me a slightly lower Q (.5 with port plugged) with a slightly higher Fb (60Hz) than the 150 L Bessel. So I’m going for the larger enclosure and just plug the port for music and leave it open for HT which will give me a F3 of about 40Hz and well good below 20Hz.

Wish I knew how to post pictures on this forum, then I’d just show you my prototype.

Good Luck, you wont be disappointed with the performance of the Maelstrom they will rock the house.

#17 of 17 OFFLINE   DerrickW

DerrickW

    Stunt Coordinator

  • 177 posts
  • Join Date: Jan 26 2003

Posted April 22 2003 - 02:17 PM

You'll have to put the pictures on another website. Then you can use the img /img to have the picture loaded here.