AVS › AVS Forum › Audio › DIY Speakers and Subs › Hey guys...we need a little rallying here...
New Posts  All Forums:Forum Nav:

Hey guys...we need a little rallying here... - Page 115

post #3421 of 7506
3/31/12 at 12:06pm
Quote:
Originally Posted by bg40403 View Post

The speaker bwaslo is dreaming up should be able to be mounted with a hook and pivot motion. Hook one edge in a positive way to carry a significant amount of the weight, then pivot up to final position and make the connection which secures it in place. It can and should be designed so one person can do it, unassisted.

Good idea. A hook toward the listening position to place the thing and pivot on and support the weight, then another chain or something in the back to point it and pull the back to the ceiling.

(Do any of this at your own risk and ingenuity, BTW -- I'm not responsible if one of these falls down and clobbers anyone or anything!)
Reply
Reply

AVS Top Picks

post #3422 of 7506
3/31/12 at 12:36pm
Here's the schematic for the box shown a little earlier, and for which the distortion plots were done. It seems to work (at least in PCD) for either the Eminence DeltaLite 2512 or the Eminence Delta 12A. Measurements here are with the 2512, at about a meter, indoors, on the tweeter axis, 1/6th octave smoothed. Woofer and waveguide are on the same mounting plane, so if you inset one very much, inset the other by about the same amount (or you can vary the insets to direct the vertical lobe a little). BTW, on these designs I'd recommend getting the tweeters up away from furniture height for use, to minimize near reflections off of it all -- sounds better, I use ear level.

A relatively simple crossover (though it has my finicky tweaks on the waveguide). Parts Express part numbers are shown on the schematic as well.

For the 33uF capacitor, an inexpensive NP electrolytic will work fine, but if you plan to have the speaker passed down through generations of your descendents, you can use the $13 film cap instead.

Horizontal Curves, 0 to 90 degrees in 7.5 degree steps:

The dark blue trace (the 7th trace) is 45 degrees.

Horizontal Polar Map:


As Erich mentioned, the woofer hole was cut too low, so CTC is about an inch and half further than it has to be. Anyway, here are Vertical (Up) curves to 30degrees. Flat response is a bit narrow in the vertical, but that should be better when the drivers are moved closer together.
Reply
Reply
post #3423 of 7506
3/31/12 at 1:14pm
Quote:
Originally Posted by bwaslo View Post

Has anyone ever done ceiling-mounted L and R front speakers?

Ceiling no, but with music I found that having my mains up high (dual concentric tweeters ~5' up) did not adversely affect stage height. Add a picture and the effect should be less.

FWIW, I think you may have hit on a serious advance with the horizontal shaded array under the waveguide. I can't see why one with good drivers (like the Aura NS6) wouldn't be the generally preferred configuration for domestic use. The form factor makes properly identical front mains easy to effect, they offer good directivity control, null issues are smaller, and while ultimate spl is perhaps lower than with a top tier 12, they're still plenty efficient and capable for domestic use.

Nice work sir!
Reply
Reply
post #3424 of 7506
3/31/12 at 1:15pm
Quote:
Originally Posted by bwaslo View Post

Here's the schematic for the box shown a little earlier, and for which the distortion plots were done. It seems to work (at least in PCD) for either the Eminence DeltaLite 2512 or the Eminence Delta 12A. Measurements here are with the 2512, at about a meter, indoors, on the tweeter axis, 1/6th octave smoothed. Woofer and waveguide are on the same mounting plane, so if you inset one very much, inset the other by about the same amount (or you can vary the insets to direct the vertical lobe a little). BTW, on these designs I'd recommend getting the tweeters up away from furniture height for use, to minimize near reflections off of it all -- sounds better, I use ear level.

A relatively simple crossover (though it has my finicky tweaks on the waveguide). Parts Express part numbers are shown on the schematic as well.

For the 33uF capacitor, an inexpensive NP electrolytic will work fine, but if you plan to have the speaker passed down through generations of your descendents, you can use the $13 film cap instead.

Horizontal Curves, 0 to 90 degrees in 7.5 degree steps:

The dark blue trace (the 7th trace) is 45 degrees.

Horizontal Polar Map:


As Erich mentioned, the woofer hole was cut too low, so CTC is about an inch and half further than it has to be. Anyway, here are Vertical (Up) curves to 30degrees. Flat response is a bit narrow in the vertical, but that should be better when the drivers are moved closer together.

Are these better crossed @100? What do they look like down to 80 or 60? Whats the -3db point?
Reply
Reply
post #3425 of 7506
3/31/12 at 1:23pm
Quote:
Originally Posted by NicksHitachi View Post

Are these better crossed @100? What do they look like down to 80 or 60? Whats the -3db point?

That would depend on the box tuning. The crossovers were just done with "test mules", 2cuft+, no ports. To use either of these woofers in a box like this, you'd probably want to port them. Down below a few hundred Hz, you're in the territory where Thiele and Small (and the room placement!) tell the story, not the crossover. Put the models into WinISD (or other sim program) and come up with a box size and tuning you want to go with. I'd aim for something you can crossover to at around 80Hz. The ports should be contributing mostly below the crossover point, mostly just minimizing the cone movement and picking up the response near crossover point.
Reply
Reply
post #3426 of 7506
3/31/12 at 1:28pm
bwaslo,

I'm in the process of trying to make sense of polars. Would you mind explaining the dip at about 2.8 kHz? In the FR curves it looks like a 2-3 dB dip but in the polars it seems to be about 6 dB. Is that the room, crossover, or CD? I'm just trying to correlate the two - if there is a correlation at all - and make sense of it. Thx
Reply
Reply
post #3427 of 7506
3/31/12 at 1:40pm
Quote:
Originally Posted by Gooddoc View Post

bwaslo,

I'm in the process of trying to make sense of polars. Would you mind explaining the dip at about 2.8 kHz? In the FR curves it looks like a 2-3 dB dip but in the polars it seems to be about 6 dB. Is that the room, crossover? I'm just trying to correlate the two - if there is a correlation at all - and make sense of it. Thx

The 2.8kHz dip is a waveguide/driver thing. There are soft peaks at around 2kHz and 4kHz (tweaked down in the crossover), the dip is what is between.

When reading the polar, draw an imaginary horizontal line at the radiation direction angle of interest (if you want the response at 45 degrees, look for 45 degrees at the left axis). The colors that line goes through describes the response there. The white lines mark the dB levels that are identified on the right axis. Decibel levels are relative to the highest level shown on the chart (in dark RED, which will be the "0dB" reference).

Basically, for perfect constant directivity, you would see horizontal color bars across all the frequencies, with red in the center, and other colors as you go off-axis. For perfect omnidirectional (which I don't think would be a good thing, since the sweet spot can't be as wide), you'd see red everywhere. Real world CD speakers give narrowing at high frequencies (the red bar gets skinnier) and at some low frequency, the graph goes red from top to bottom where the pattern goes omni. Goodness is determined by how flat the color changes are across frequency, though gradual narrowing toward high frequencies isn't altogether a bad thing.
Reply
Reply
post #3428 of 7506
3/31/12 at 2:24pm
Thank-you very much! Appreciate the detailed explanation. Nice to have confirmation I've been reading them right. It's the correlation between what I'm seeing and what's causing it that is my problem currently. I've got some group buy drivers on the way and starting to feel the pressure to up my knowledge base a bit! Your posting of the crossover schematics is really fantastic and a big help to guys like me looking to start learning how to build and design crossovers. I have a couple of speaker building and active/passive crossover design books on the way and lots of reading to do!
Reply
Reply
post #3429 of 7506
3/31/12 at 2:56pm
Graph shows 1w and a 2nd order 80hz highpass. It can take 200w before xmax with the 80hz highpass.
-3db @ 68hz
Attachment 242327
LL
Reply
Reply
post #3430 of 7506
3/31/12 at 3:19pm
Is there a plan to work up a generic HF section for the SEOS 12 and DE250 (and clones?) similar to the eWave "standard" crossover? This would be helpful for those of us that plan to use non-standard woofers. Hint: calling it a standard design will minimize the the requests for help designing custom XOs.
Reply
Reply
post #3431 of 7506
3/31/12 at 3:44pm
To check using the Delta12A, here it is measured using the same crossover as given above, along with the curve (in red) using the 2512.


Pretty close, probably a good way to trim down some cost if needed, if weight doesn't matter. Call it the "DeltaHeavy"
Reply
Reply
post #3432 of 7506
3/31/12 at 3:52pm
Quote:
Originally Posted by mobeer4don View Post

Is there a plan to work up a generic HF section for the SEOS 12 and DE250 (and clones?) similar to the eWave "standard" crossover? This would be helpful for those of us that plan to use non-standard woofers. Hint: calling it a standard design will minimize the the requests for help designing custom XOs.

Well, ok. I hereby pronounce the HF section of the above crossover to be a "standard".

It's not really quite that straightforward, though, since the differences in depth can cause some problems near the crossover point (sometimes fixable by inverting tweeter phase, sometimes best done by shifting position of the waveguide forward or back). Same was true with the original eWave standard crossover. Of course, compared the measured responses of most of the legacy designs that get eWaved, "close enough" on depth offset will still be great improvement.
Reply
Reply
post #3433 of 7506
3/31/12 at 5:04pm
Quote:
Originally Posted by bwaslo View Post

To check using the Delta12A, here it is measured using the same crossover as given above, along with the curve (in red) using the 2512.

Pretty close, probably a good way to trim down some cost if needed, if weight doesn't matter. Call it the "DeltaHeavy"

At half the price, i dont see any appreciable difference in the response!?
Reply
Reply
post #3434 of 7506
3/31/12 at 6:08pm
Quote:
Originally Posted by NicksHitachi View Post

At half the price, i dont see any appreciable difference in the response!?

Maybe some of that is because they aren't ported or cranked up at higher volumes yet.


Quote:
Originally Posted by bwaslo View Post

To check using the Delta12A, here it is measured using the same crossover as given above, along with the curve (in red) using the 2512.
Pretty close, probably a good way to trim down some cost if needed, if weight doesn't matter. Call it the "DeltaHeavy"

You mean we call it "Heavy-D".
Reply
Reply
post #3435 of 7506
3/31/12 at 7:28pm
Thread Starter 
is the Eminence Deltalite II 2512 Neo 12" Driver the woofer that you are working with?

Mrkazador, there is a difference between the deltalite ii and the deltalite i. i'm not sure which is being used.
Reply
Reply
post #3436 of 7506
3/31/12 at 7:49pm
Yeah, it's the Deltalite II.

Coming up next: using the Designer 12 (that one has bass in sealed box!).
Reply
Reply
post #3437 of 7506
3/31/12 at 8:04pm
I have to say that this one so far sounds the best to me. That is probably because it has bass in the given configuration (sealed box, no sub playing). But it does have a nice richness to it even with stuff that doesn't have bass. It took a long time between changes, so it might be mind tricks. Or it might be the steeper slope on the crossover. Or it might be that it's not as sensitive and the amp gets worked harder. Who knows. Anyway, I like it a lot.

For reasons I don't really follow, this one needed more delay on the woofer to match up. I didn't want to invert the tweeter to get there, since I want it to be able to play with the middle channel speaker Malcolm which has non-inverted tweeter (not sure it matters, but why not keep it the same?). So I added another two filter sections, which has the added benefit of rolling off the break-up region of the woofer even further. There are some slight changes to the tweeter section, too, to help with the match at crossover.

Schematic:


I won't measure the horizontal pattern, since that's mostly just the waveguide and woofer diameter and it's the same. Vertical pattern is less touchy on this (the crossover is steeper). Here's the on-axis response just.... because:


Edit: It is also the prettiest, I think. Cosmetically matches the SEOS12 nicely:
Reply
Reply
post #3438 of 7506
3/31/12 at 9:00pm
That does look nice. I've got to work on the new baffle really soon so you don't have to change the crossovers later. That would stink. Can't believe I somehow messed that up.


Bill, earlier you mentioned that the crossover was a bit more complicated due to your "finicky tweaks on the waveguide" portion.

What part are you talking about? And how many extra components does it add? What happens if it's removed?

The reason I'm asking is because I think you're shooting for a response as flat as possible. But I'm wondering if the crossover can be shrunk down a bit and still work good for home theater versus "hi fi". Not sure if that even makes sense or not. Probably not.

I think once things get rolling good and I can make boxes quicker, we should look into getting some designs solely for home theater that probably wouldn't even need the L-pad.

Then again, I could be nuts too. But maybe a couple versions could be offered up with a choice of the simple crossover or the "BWaslo Finicky Deluxe". But we probably shouldn't call it the BFD crossover network.
Reply
Reply
post #3439 of 7506
3/31/12 at 9:12pm
Hi Erich,

It's those Inductor-Capacitor-Resistor (LCR) strings that go across the CD driver. Can't just take them out, other stuff would need to be adjusted to compensate or it would sound awful. I did run some designs without those LCR, but really think they should be left in. One of the bumps they deal with is at 2kHz, which is a terrible frequency to have a bump at (near where ears are most sensitive). Why go to all the trouble of waveguides and CDs and then cheap out on a few components?

Besides, I'm too lazy to do it all over again! Did my mass speaker designing this past week, got to get back to some other things now.

I don't really think there is a difference between HT and HiFi design goals. HT might need to play louder (and maybe not), but flat response is still good -- overall response shaping can be added at the receiver or equalizer, better than hardwired into the speaker.

The components inside the dashed-line box (in the schematics) are fixed resistors. The "L-Pad" notation is just to say they could be replaced with an L-Pad if someone wanted to for whatever reason. I wouldn't, myself, L-Pads are just another mechanical part to go bad, fixed resistors are lots more reliable.

Crossovers shouldn't have to change if you just are moving the woofer closer to the WG.
Reply
Reply
post #3440 of 7506
3/31/12 at 9:13pm
I think I have now spent more time deliberating woofer chooses than ANY other decision I have ever made.

Today I had the "no more distinctive color choices left" moment when doing the WinISD plots. lol

After weeks and weeks, I am not sure I am any closer either.

1. AE TD12S with Apollo upgrade (I know the M might be "better", but the S is still has very low inductance and a light weight cone all with more than twice the excursion)
2. MTM with deltalite
3. TMM 2.5 way with deltalite ~300hz low-pass on the lower one.
4. side by side Anarchy woofers below. With maybe Another row above
5. SbS Aurum Cantus AC165 (6.5") with a AC300 (12") below it
6. side by side Dayton reference 6.5"s and a 15reference3 DVC below those.


Steve
Reply
Reply
post #3441 of 7506
3/31/12 at 9:24pm
Quote:
Originally Posted by bwaslo View Post

Hi Erich,

It's those Inductor-Capacitor-Resistor (LCR) strings that go across the CD driver. Can't just take them out, other stuff would need to be adjusted to compensate or it would sound awful. I did run some designs without those LCR, but really think they should be left in. One of the bumps they deal with is at 2kHz, which is a terrible frequency to have a bump at (near where ears are most sensitive). Why go to all the trouble of waveguides and CDs and then cheap out on a few components?

Besides, I'm too lazy to do it all over again! Did my mass speaker designing this past week, got to get back to some other things now.

I don't really think there is a difference between HT and HiFi design goals. HT might need to play louder (and maybe not), but flat response is still good -- overall response shaping can be added at the receiver or equalizer, better than hardwired into the speaker.

The components inside the dashed-line box (in the schematics) are fixed resistors. The "L-Pad" notation is just to say they could be replaced with an L-Pad if someone wanted to for whatever reason. I wouldn't, myself, L-Pads are just another mechanical part to go bad, fixed resistors are lots more reliable.

Crossovers shouldn't have to change if you just are moving the woofer closer to the WG.



Don't get me wrong Bill, they look great. I was just curious what that "finicky" portion was that you were talking about and what it was for.

I'm no crossover guru and didn't know if it was 10 components to fix a blip in the top end or something like that which made it finicky to you, but others might not mind. But now I see that's not the case. Certainly no reason to rework anything when it's actually needed. I have no clue on complex crossovers at the moment, but I know it's not easy.
Reply
Reply
post #3442 of 7506
3/31/12 at 9:43pm
Quote:
Originally Posted by Erich H View Post

Don't get me wrong Bill, they look great. I was just curious what that "finicky" portion was that you were talking about and what it was for.

No issue, Erich. It's just that I know Zilch's crossovers were simpler and some folks feel like "simplest crossover is better" (which I think is just superstitious nonsense). After playing with MiniDSP on my home speakers, I've become a believer in getting the waveguide section smooth, it does matter. Small narrow blips would be ok, but broader bumps can be heard.

MoFinWiley - Forget doing MTM with these waveguides. You will have bad vertical problems if you try (I've tried, using 5" mids and a shorter waveguide). CTC problems get twice as bad as with a TM, since when the tweeter stops playing below crossover, the relevant CTC is then the space between the two Ms, twice as far as between M and T. TMM(2.5) can work, or TMW, but design won't be trivial. Just plain TM (plus a few separate subs placed where THEY want to be in the room) works really well, not sure what there is to get from the other configs. (note here, that the only qualification for being an audiophile is to be opinionated, and I'm showing my qualifications )
Reply
Reply
post #3443 of 7506
3/31/12 at 9:54pm
I've been listening to this one mono speaker while writing all this, and don't want to stop. Whatever it is, this setup has something really nice going for it. Everything seems to sound good on it, I've just been flipping around on my Pandora channels. I'm not sure this Designer12 isn't maybe better than the TD15Ms in sounding coherent with the waveguide.
Reply
Reply
post #3444 of 7506
3/31/12 at 10:01pm
Quote:
Originally Posted by bwaslo View Post

No issue, Erich. It's just that I know Zilch's crossovers were simpler and some folks feel like "simplest crossover is better" (which I think is just superstitious nonsense). After playing with MiniDSP on my home speakers, I've become a believer in getting the waveguide section smooth, it does matter. Small narrow blips would be ok, but broader bumps can be heard.

MoFinWiley - Forget doing MTM with these waveguides. You will have bad vertical problems if you try (I've tried, using 5" mids and a shorter waveguide). CTC problems get twice as bad as with a TM, since when the tweeter stops playing below crossover, the relevant CTC is then the space between the two Ms, twice as far as between M and T. TMM(2.5) can work, or TMW, but design won't be trivial. Just plain TM (plus a few separate subs placed where THEY want to be in the room) works really well, not sure what there is to get from the other configs. (note here, that the only qualification for being an audiophile is to be opinionated, and I'm showing my qualifications )

Yeah. I hear you. I have pretty much decided that that i should just get what I actually wanted, the AE speaker. They are claiming 4-6 weeks which works out well for me anyways. I really do like the look of the AE speakers, and if I wanted to compromise, I would keep the 99 db/w cerwin vega re-38s I have now.

LCR= SEOS-12 with "Greater" clone and AE TD12X with apollo upgrade(not S that I put earlier)
SR= SEOS-12 with deltalite

Not sure what to do about amps LOL...Going minidsp route. Been thinking of trying the Connexelectronics Class T setups. "supposedly" similer performance to the hypex UCD stuff, but less money.


Steve
Reply
Reply
post #3445 of 7506
3/31/12 at 10:11pm
Quote:
Originally Posted by bwaslo View Post

I've been listening to this one mono speaker while writing all this, and don't want to stop. Whatever it is, this setup has something really nice going for it. Everything seems to sound good on it, I've just been flipping around on my Pandora channels. I'm not sure this Designer12 isn't maybe better than the TD15Ms in sounding coherent with the waveguide.

Damn now I want to hear them even more
Reply
Reply
post #3446 of 7506
3/31/12 at 11:33pm
Quote:
Originally Posted by LTD02 View Post

Mrkazador, there is a difference between the deltalite ii and the deltalite i. i'm not sure which is being used.

I updated my post, looks even better!
Reply
Reply
post #3447 of 7506
3/31/12 at 11:42pm
Quote:
Originally Posted by bwaslo View Post

I have to say that this one so far sounds the best to me. That is probably because it has bass in the given configuration (sealed box, no sub playing). But it does have a nice richness to it even with stuff that doesn't have bass. It took a long time between changes, so it might be mind tricks. Or it might be the steeper slope on the crossover. Or it might be that it's not as sensitive and the amp gets worked harder. Who knows. Anyway, I like it a lot.

For reasons I don't really follow, this one needed more delay on the woofer to match up. I didn't want to invert the tweeter to get there, since I want it to be able to play with the middle channel speaker Malcolm which has non-inverted tweeter (not sure it matters, but why not keep it the same?). So I added another two filter sections, which has the added benefit of rolling off the break-up region of the woofer even further. There are some slight changes to the tweeter section, too, to help with the match at crossover.

Schematic:

I won't measure the horizontal pattern, since that's mostly just the waveguide and woofer diameter and it's the same. Vertical pattern is less touchy on this (the crossover is steeper). Here's the on-axis response just.... because:

Edit: It is also the prettiest, I think. Cosmetically matches the SEOS12 nicely:

What kind of sensitivity do you estimate with this design? Does the new DS series have shorting rings in the motor? Looks like an interesting design!
Reply
Reply
post #3448 of 7506
4/1/12 at 12:25am
Quote:
Originally Posted by MoFinWiley View Post

4. side by side Anarchy woofers below. With maybe Another row above

While the ae is tempting, probably best to stick with stuff actually available in real life. (if a 12, why not 12tbx100?)

A "Malcolm" style four-abreast shaded array with Anarchies would bring the bass, if that configuration will work with drivers of that size....
Reply
Reply
post #3449 of 7506
4/1/12 at 6:12am
Sensitivity will be the sensitivity off the data sheet, minus maybe a dB from the series resistance of the inductors in the woofer xover section. So, around 89dB.

No mention of shorting rings on the data sheet, so probably not.
Reply
Reply
post #3450 of 7506
4/1/12 at 6:50am
Quote:
Originally Posted by bwaslo View Post

No issue, Erich. It's just that I know Zilch's crossovers were simpler and some folks feel like "simplest crossover is better" (which I think is just superstitious nonsense). After playing with MiniDSP on my home speakers, I've become a believer in getting the waveguide section smooth, it does matter. Small narrow blips would be ok, but broader bumps can be heard.

Do you work for the government?

Just kidding. Government employees, bureaucrats, and politicians think the same thing.
Reply
Reply
New Posts  All Forums:Forum Nav:
  Return Home
  Back to Forum: DIY Speakers and Subs

AVS Top Picks

AVS › AVS Forum › Audio › DIY Speakers and Subs › Hey guys...we need a little rallying here...