will the HiVi M3N's work in the Needles?

[patrickm]

this is basically the same driver as the w3-871 used in the original needle but with a cheaper, flush-mount pincushion frame.
http://www.parts-express.com/pe/show...number=264-809

in a 10-minute test with the baffle clamped on, the M3N sounded pretty good, similar to the B3N... i didn't have the exact components for the dragon foal filters though. more to follow.

[nikbrewer]

so, any one have any hard box details yet? I know nothing of t-lines, so if someone can just draw up a side view with demensions, ill go ahead and build these

thanks

[Sibelius40]

so, any one have any hard box details yet? I know nothing of t-lines, so if someone can just draw up a side view with demensions, ill go ahead and build these

thanks

Here is the diagram.

[Wolf]

Here is the diagram.

The terminus length has changed, as patrick said.

He is also going to be reporting on his findings,
Wolf

[patrickm]

so, any one have any hard box details yet? I know nothing of t-lines, so if someone can just draw up a side view with demensions, ill go ahead and build these

thanks

nik - just use that diagram, with the following tweaks:

make the length of the slot port (made of the edge of "A" plus the length of "D" ) that the divider ("C") sits on about 1.6" long. this will shorten the port and raise the tuning frequency a little, which the model predicts will improve overall response.

you can adjust the length of "C" accordingly, so that its top edge ends up in the center of the cabinet an equal distance from the top, back, and front baffle. in reality, since you're only changing D by 1/2" or so, the difference in C's length will be minimal.

to answer Sibelius40's theory question -- you can have a small item jut out into the line with minimal difference, though it will have some effect in theory. in the Copper Project, marty used a couple 3/8" dowels as braces to hold the stuffing, and found no practical difference. those were in the middle of the line, and the round shape will help keep the air flowing pretty smoothly. you could probably have an edge jut out into this line, but i think based on what you propose, you'd be changing the line's taper (ratio of small end to large end areas) but not the slot port -- without changing D, the slot port is the same. what i think you could do is take a half inch off D, and change the positioning of C to change the length of the port by 1/2". mount on top of C for the base amount (1.6"), and mount flush with the bottom edge of D to add 1/2" to the port (2.1", close to the original). you would be changing the taper a little though. might be worth it for testing just to see.

[Sibelius40]

nik - just use that diagram, with the following tweaks:

make the length of the slot port (made of the edge of "A" plus the length of "D" ) that the divider ("C") sits on about 1.6" long. this will shorten the port and raise the tuning frequency a little, which the model predicts will improve overall response.

you can adjust the length of "C" accordingly, so that its top edge ends up in the center of the cabinet an equal distance from the top, back, and front baffle. in reality, since you're only changing D by 1/2" or so, the difference in C's length will be minimal.

to answer Sibelius40's theory question -- you can have a small item jut out into the line with minimal difference, though it will have some effect in theory. in the Copper Project, marty used a couple 3/8" dowels as braces to hold the stuffing, and found no practical difference. those were in the middle of the line, and the round shape will help keep the air flowing pretty smoothly. you could probably have an edge jut out into this line, but i think based on what you propose, you'd be changing the line's taper (ratio of small end to large end areas) but not the slot port -- without changing D, the slot port is the same. what i think you could do is take a half inch off D, and change the positioning of C to change the length of the port by 1/2". mount on top of C for the base amount (1.6"), and mount flush with the bottom edge of D to add 1/2" to the port (2.1", close to the original). you would be changing the taper a little though. might be worth it for testing just to see.

Thanks for the "theory" answer - from the pics you attached, I see what you are suggesting for the slot port in order to shorten.

The pic I attached is the Needles original design specs, but I am all in favor of modifying the enclosure to improve response. On the diyForum, another person who created a TL enclosure called the TABAQ posted some response graphs showing some rather imperfect results from the original spec Needle cabinet and TB driver (then more showing why his design was better). Clearly, one can get closer to perfection with this enclosure and I look forward to Nik's results.

[nikbrewer]

thanks, ill get to work on it tomorrow

[Wolf]

Keep us posted! I'm interested to hear how these "Dragonflies" turn out...

Later,
Wolf

[patrickm]

ok, some verdicts are in... and the winner is...

the M3N.
the bass is a little clearer, perhaps a few Hz deeper, and the top end is a good bit better than the B3N using wolf's and zaph's filters. I seem to recall tweaking zaph's filter at some point to try and give it a little more sizzle, but i still dont' think it was this good. kudos to wolf -- i may never buy another B3N...

the caveats:
-- both of my enclosures were clamped shut, and i can't guarantee they were perfectly sealed. but i did tape over the edges and tried a couple different clamping methods (without any noticeable differences in sound) and the M3N was still better.
-- my port is a 4"x1" slot port 2.1" long. these were already built, so i can't evaluate the 1.6" port from the model right now. but i'm already impressed with the upgrade to the M3N, so i can't imagine it'll work out worse than the default cab with a B3N.
-- i didn't have the exact dragon foals components. my 1mH was a p-core, and the 5.6 uF was closer to 5.79 uF. i tried to unwind the inductor from .10 mH to .095 to compensate.

obviously, there is not limitless bass, but it's impressive for a 3" driver. "in your eyes" from peter gabriel seemed like one of the better songs to highlight the better points of this design.

anyhow, hope that helps, nik. sorry for the delays. now i guess it's time to finish those rs150 TL's...

[Wolf]

and the winner is...

the M3N.
the bass is a little clearer, perhaps a few Hz deeper, and the top end is a good bit better than the B3N using wolf's and zaph's filters. I seem to recall tweaking zaph's filter at some point to try and give it a little more sizzle, but i still dont' think it was this good. kudos to wolf -- i may never buy another B3N

Thank you for your results and opinion! I like the M3N more than the B3N/S, and always have. Needs more taming, but just plain sounds better!

So- Who's building some 'Dragonflies'??
Thanks, Patrick!
Wolf

[Sibelius40]

ok, some verdicts are in... and the winner is...

the caveats:
-- both of my enclosures were clamped shut, and i can't guarantee they were perfectly sealed. but i did tape over the edges and tried a couple different clamping methods (without any noticeable differences in sound) and the M3N was still better.
-- my port is a 4"x1" slot port 2.1" long. these were already built, so i can't evaluate the 1.6" port from the model right now. but i'm already impressed with the upgrade to the M3N, so i can't imagine it'll work out worse than the default cab with a B3N.
-- i didn't have the exact dragon foals components. my 1mH was a p-core, and the 5.6 uF was closer to 5.79 uF. i tried to unwind the inductor from .10 mH to .095 to compensate.

obviously, there is not limitless bass, but it's impressive for a 3" driver. "in your eyes" from peter gabriel seemed like one of the better songs to highlight the better points of this design.

anyhow, hope that helps, nik. sorry for the delays. now i guess it's time to finish those rs150 TL's...

So you used Wolf's contour filter for this test of the M3N's, not Zaph's? Or both?

[Sibelius40]

Thank you for your results and opinion! I like the M3N more than the B3N/S, and always have. Needs more taming, but just plain sounds better!

So- Who's building some 'Dragonflies'??
Thanks, Patrick!
Wolf

I might...I have a pair of M3N's that are not being used right now. After looking at a finished pair of Needles, however, they reminded me a lot of a skyscraper in Chicago near my employer's office tower. I was going to call these Les Tours .

One other question, is there a significant enough difference in tuning or output between the slot port on these (using the 1.6" design) versus the round port on the Dragon Foals that might necessitate a modification of the contour filter? Or is the filter mainly designed to smooth out the upper freq range?

[Wolf]

The boost in his response and my box model are very similar, and should not be an issue to make a difference. Also- the contour I used does not include BSC, it's for the resonances only, as the tuning took care of BSC. The notches smooth out the hump in the midrange and the material resonance up high.



Patrick used the DF filter on his experiment, so just use the DF filter with the adjusted port tuning in the Needle boxes.

I'd like to hear these someday!
Later,
Wolf

[Sibelius40]

The boost in his response and my box model are very similar, and should not be an issue to make a difference. Also- the contour I used does not include BSC, it's for the resonances only, as the tuning took care of BSC. The notches smooth out the hump in the midrange and the material resonance up high.



Patrick used the DF filter on his experiment, so just use the DF filter with the adjusted port tuning in the Needle boxes.

I'd like to hear these someday!
Later,
Wolf

Wolf - that contour certainly does a good job at taming things down. I can't see because of the "Filter" button, but what is the upper-end frequency response on your sceenshot? Do these go up to 20khz?

Because of my work travel schedule for the next month, a quick build is unlikely. Perhaps in a few weeks I will have some time - if nothing else, it will give me time to think about the cabinet exteriors!

Scott

[PWR RYD]

It looks like all of those plot lines end at 20kHz.

[Wolf]

Yeah- power-rider is right. The plots stop at 20kHz, which means the FR is about 3-4dB down at 20kHz but only from about 18kHz. It's fairly extended.

That filter really does do a fine job.

I'm just tickled Patrick took this on, and feel it might be a very viable little project.

Keep us posted on your progress! Whenever you get to it....

Later,
Wolf

[nikbrewer]

well, i finally got around to building 1 box for my needles.

I ended up makinf the slot port 2" long, so we will see how that goes

1 question, in this inclosure, do you add stuffing, and if so, where?

I should have the cross over together soon, so i can report results as soon as i know about the stuffing issue

[Sibelius40]

I believe the stuffing is added in the line all the way to the bend, where it is thinned out in the turn. I thought there was a pic of this somewhere on this thread - if not, I can probably find one.

[patrickm]

stuffing is against the front baffle, dense at the bottom and thins out slightly as you go up. if you use little dowels, you can try taking it around the turn. i think i ended mine at the top of the divider.

[nikbrewer]

cool, i stuffed it up flushwith the bottom of the divider. I did not do it before construction, so i kinda just shoved it down with a stick, but i think it worked.

One is all done and i am listening to it and it sounds very nice.

Good job on that filter wolf. I did a big terminal that uses a 3" hole, so i fed the crossover in that way and screwed it to the divider.