54L Sealed DPL12 Test Results

[swissy]

Last night I tested my DIY 54L DPL12 w/ 16oz of polyfill. The results were encouraging and frustrating.

Parameters:

Sine waves from 20-70 @ 5 Hz intervals until 40 Hz

In corner of room with SPL meter @ 1 meter away

Phase set to 0

Crossover set to ~90 Hz

Sub Volume @ 40-50%

Amp PE 250W

Yamaha RX-V2300 set to -44 (Found 30 Hz started to bottom out at -40.5)

Pioneer DV-440 for movie and test signals

Results:

dB Freq

95 20

104 25

102 30

102 35

111 40

112 50

103 60

103 70

DPL12 bottomed out at 30-35 Hz before any other frequency. A little strange since it played 20 Hz well. Played several tracks on Avia and calibrated the sub. I had to turn it down a little during the asteroid chase in Episode 2. The second sonic charge hits the 30-35 Hz range while braking asteroids and sub was bottoming out.

Conclusions:

Probably should have tested this in the center of the room or outside to try to match the response curves. In addition, the strange shape and size of the room (as seen in attachment) makes me think the ported might give me a better response. However, for best results, I think a second sub might do a good job of pressurizing the room. Budget does not allow for this, however. Sub volume must be set at 25-30% for audio calibration. Receiver cannot exceed -30 without affecting sub performance. This is the Avia calibrated volume for my system. However, actual listening will not exceed -35. I may attempt to move sub to a different location, but this corner has always been the best location.

What do you think? Should I expect more? Why would it bottom out at 30-35 Hz?

[fabulousfrankie]

----------------

On 11/19/2003 9:36:00 AM swissy wrote:

Last night I tested my DIY 54L DPL12 w/ 16oz of polyfill. The results were encouraging and frustrating.

Parameters:

Sine waves from 20-70 @ 5 Hz intervals until 40 Hz

In corner of room with SPL meter @ 1 meter away

Phase set to 0

Crossover set to ~90 Hz

Sub Volume @ 40-50%

Amp PE 250W

Yamaha RX-V2300 set to -44 (Found 30 Hz started to bottom out at -40.5)

Pioneer DV-440 for movie and test signals

Results:

dB Freq

95 20

104 25

102 30

102 35

111 40

112 50

103 60

103 70

DPL12 bottomed out at 30-35 Hz before any other frequency. A little strange since it played 20 Hz well. Played several tracks on Avia and calibrated the sub. I had to turn it down a little during the asteroid chase in Episode 2. The second sonic charge hits the 30-35 Hz range while braking asteroids and sub was bottoming out.

Conclusions:

Probably should have tested this in the center of the room or outside to try to match the response curves. In addition, the strange shape and size of the room (as seen in attachment) makes me think the ported might give me a better response. However, for best results, I think a second sub might do a good job of pressurizing the room. Budget does not allow for this, however. Sub volume must be set at 25-30% for audio calibration. Receiver cannot exceed -30 without affecting sub performance. This is the Avia calibrated volume for my system. However, actual listening will not exceed -35. I may attempt to move sub to a different location, but this corner has always been the best location.

What do you think? Should I expect more? Why would it bottom out at 30-35 Hz?

----------------

Something doesn't add up here...as you go down in frequency excursion demands increase and there are no exceptions to this when it comes to sealed enclosures. It doesn't make sense to bottom out at a higher frequency. Was the level of the sub changed between the tones?

Also, both of the Star Wars dvd's are some of the most bass intensive dvd's I've ever heard so it's not really a surprise given that you have it in a decent size room. If you switch to a vented enclosures excursion demands decrease around the tuning point and you'll usually be able to extend further down before your woofer bottoms.

[swissy]

Here is the procedure I used:

Set receiver to calibrated volume for other speakers (-30) and sub amp to 50% played 20Hz and recorded the readings. Then I moved up to 25, 30, ect.

When I reached 30Hz, the sub made a slapping sound. I turned down receiver until it went away. Then turned the receiver up slowly until slapping sound came back (-41) then decreased 3dB to (-44). Then I took all of the readings from there starting over again at 20Hz.

Here is my first thought after some time to think about it:

The polyfill I used was the type for stuffing pillows. I then secured it, so it did not fall, with ¼ thick roll polyfill. I stuffed this between the braces and made a floor out of the roll polyfill. Hence, the polyfill sits on top of the roll style. I suppose it is possible that this is negatively affecting the volume of the box. If I model the box so that it gives a net volume of 25 L, I get a response curve that matches that of my readings. Then I check the power response and @ 30-40 Hz, it shows the displacement being limited by the driver instead of the thermal limit. This would account for the driver bottoming out at the louder levels.

If you model a 25 L box, you would see a response curve with a large hump in the response. Exactly as the data taken shows.

Ill try it without the polyfill and see what the response is then.

If any of this sounds crazy, please slap me upside the head and knock some sense into me. If I did not hear it, I would not believe it either.

[fabulousfrankie]

The uneven response you're getting is not due to the subs design, it's due to your room acoustics. The best way to deal with this is experimenting with placement. The best way to do it is to measure the response at different locations and pick the one that gives the flattest response. How much polyfill did you put in the enclosure? As long as the polyfill is packed very tightly and it's not touching any moving parts of the driver or blocking the pole piece from behind it's probably not hindering performance. Also make sure to leave adequate space for ventilation of your amp if it's built into the enclosure.

Are you sure it was bottoming you heard? Could you describe the sound as best you can. Whenever I bottom a sub it's sounds so nasty it makes my spine tingle. Tom V describes the sound like a hammer hitting a frying pan, does it sound like this to you?

One other thing, since you seem to be dissatisfied with the level of bass you're getting, see if you can put the sub closer to you. Looking at your diagram, I'd experiment with the sub on the left wall to the left of the couch facing the tv. Consider this, for every 2M you move the sub closer to you you gain 6dB. That is the same increase you would get if added another 54L sealed DPL12 and colocated the subs.

[swissy]

I wouldnt say Im dissatisfied with the level. I did not mean to give this impression. I was just curious about what seemed to be an anomaly. I tried it again and was able to get 100% turn on the gain knob @ 20 Hz for a reading of 112 dB. However, a 30 Hz signal would bottom out at 30% gain turn.

The sound is two magnets slapping together. I jumped the first time I heard it. I am positive it is bottoming out.

Again, its not the level, and I know that the room acoustics will give me a bump in the response at the 30-45 Hz range. I just cannot understand why the driver would bottom out at 30 Hz and not 20 Hz.

I did find that one of the tinsle leads was hitting the spider. This solves the rattling issue but not the bottoming out one.

Thank you again for your help.

[fabulousfrankie]

It definitely sounds like it's bottoming the way you describe it.

Have you contacted Adire for any ideas?. Unless something is defective, I don't know what to tell you.

[swissy]

I did contact Kyle from acoustic visions and we are working on figuring it out now. He thinks it may be the amp but I'm working on some theoretical models to see if I can figure it out. I am also testing the amp for a dc offset. This may lead to the answer.

I did notice that there is a point at which I turn the gain knob up on the amp that the cone jumps about half or more of its excursion in. I slowly turned the knob until it made the jump. There was a point that just made the thing go crazy. The 20 Hz signal looked good when I did this, but the 30Hz signal had a LARGE jump. This is why I am testing for a DC offset.

[audionuttt]

I just modeled that driver and enclosure for you. Since Adire didn't post BL I derived it based on the TSP's. (13.7 Newton amps, wired in paralell to arrive at their TSPs). With a 54L enclosure, unstuffed, Fb of the box (IE impedance peak of the driver, the driver/box combo's most resonant frequency) is 32hz. Stuffing will bring this down a little, but not by much. If you don't find anything else, this is likely the reason the driver bottoms at 30hz.

[audionuttt]

Forgot to attach screenshot, sorry bub. It's "theoretical" model of DPL-12 in 54L unstuffed at 250 watts PIN. Screen show's SPL/Excursion/Z.

[fabulousfrankie]

----------------

On 12/5/2003 2:26:10 PM audionuttt wrote:

I just modeled that driver and enclosure for you. Since Adire didn't post BL I derived it based on the TSP's. (13.7 Newton amps, wired in paralell to arrive at their TSPs). With a 54L enclosure, unstuffed, Fb of the box (IE impedance peak of the driver, the driver/box combo's most resonant frequency) is 32hz. Stuffing will bring this down a little, but not by much. If you don't find anything else, this is likely the reason the driver bottoms at 30hz.

----------------

But no matter what as frequency decreases the excursion increases, so it doesn't make sense that it's bottoming at 30Hz and not at lower frequencies.

[audionuttt]

----------------

On 12/5/2003 2:41:37 PM fabulousfrankie wrote:

But no matter what as frequency decreases the excursion increases, so it doesn't make sense that it's bottoming at 30Hz and not at lower frequencies.

----------------

In the theoretical world, absolutely true. Dropping an octave requires 4x the volume displacement to achieve the same output. Period. If a speaker were flat to say 20hz, and were generating 90dB @ 20hz and 40hz, the excursion required at 20 is 4x that at 40. Take a good look at those SPL charts though, and even with the likely moderate amount of room gain he gets at 20 relative to 30 and 40, he's still down -10dB at 20hz.

In reality, the drivers falling BL with excursion, rising compliance with excursion, power compression(thermal and excursion related), the cabinet/driver combo's Fb, increasing compliance relative to displacement (decreasing freq), and a myriad of other variables come into play. Virtually EVERY time I've designed and built a new prototype, the various design software I use has NEVER accurately replicated the realworld at high levels. Most programs generate excursion plots based on RMS rather than peak (don't ask me why I don't write them. I've had the authors of mine modify my programs to show peak rather than RMS) and NONE of the currently popular programs accurately model non-linear effects of compliance/inductive/BL distortion or power compression. Even best of them at most can only give an IDEA of what happens in the real world.

Fb of a box IS where the driver box combination is at it's most highly resonant. Remember a measured "impedance curve" is motional impedance, not static. The reason for the rise in impedance at Fb (sealed) is that is where the cone is MOVING the most. If he's bottoming at 30hz on test tones, and cannot 'find' another reason, that's likely the cause.

[swissy]

Audionuttt-

Want to hear something even more strange (this shows my lack of understanding). When using the scope I used the BL of the Shiva (10.8) and found that around 30V should have done me in @ 20Hz. This was two weeks ago so my info may be a bit off, but I do remember measuring 60V and not bottoming out. After 60V the amp started clipping so I just shut it down.

Please dont take these numbers as fact. This was a couple of weeks ago. I do know, however, that I did reach 60V at 20Hz. Even with a newly calculated BL of 13.9 (very close to yours) I shouldnt have been able to do this.

BTW the amp did not have a DC offset. I knew this probably wasnt the problem, but I was out of explanations.

I did notice that I was excursion limited below ~35Hz but again I would think a bottom @ 35 would produce a bottom @ 20Hz. I guess this may be one of those things that Im going to have to let go of and just be happy with. Again, I am _not_ saying the driver preformed poorly, quite the opposite. I really enjoyed the experience and look forward to testing it in an 85L Vented tuned @ 20Hz.

[audionuttt]

----------------

On 12/5/2003 4:26:20 PM swissy wrote:

Audionuttt-

Want to hear something even more strange ~~ I shouldn’t have been able to do this.

I did notice that I was excursion limited below ~35Hz but again I would think a bottom @ 35 would produce a bottom @ 20Hz.

~~ I really enjoyed the experience and look forward to testing it in an 85L Vented tuned @ 20Hz.

----------------

60v into a 3.36 ohm dcr driver is roughly 1kw. 1kw is an ENORMOUS amount of power, and dumping that into a 2.5" diameter coil will IMMEDIATELY heat that coil significantly. I wouldn't be surprised if you weren't doubling or even quadrupling the drivers DCR after that run. Next time you bench it, measure DCR COLD, then RIGHT AFTER dumping 60v into it. You'll be amazed, this is the very definition of thermal power compression.

Looks like 85L and 20hz Fb vented should be fun :-) Check file uploaded.

[swissy]

I do not dare try it again. The only reason I kept the power coming was my curiosity took over faster than my common sense. The only fault I find with your calculation is that 20Hz will probably have higher impedance than 3.36 because it was pretty close to the tuning frequency of 35Hz. Therefore, my guess is the impedance was probably around 8 ohms. For a total of ~450W

Besides, I could not get the tinsel lead to stop banging on the spider, or perhaps a lead was loose in the cone. There was a tapping at all but the lowest volumes. Therefore, Dan at Adire sent me a replacement. They are definitely professionals there. He was great with answering my questions and did not even sound too mad when I told him the gamut of tests I ran (I am sure he was fuming inside though!)

Thank you for the responses.

[fabulousfrankie]

Did Dan give you his ideas as to why it was bottoming?

[swissy]

He didn't have a good explination for why it would @ 30 and not below. He just guessed it was the amps rumble feature, but I think he was just throwing that out there. He was more concerned that I had tried to find the frequency responce, and in the prosess bottomed it out.