or Connect
AVS › AVS Forum › Audio › DIY Speakers and Subs › DIY 3 way main speaker with 15" drivers??
New Posts  All Forums:Forum Nav:

DIY 3 way main speaker with 15" drivers?? - Page 13

post #361 of 862
Thread Starter 
Quote:


Pennyray ought to edit the first post subject field to read "Lambda TD drivers for main speaker".

I can do that still ? I should!

But I can not change the title of the thread, correct?
post #362 of 862
...... nuke
post #363 of 862
Thread Starter 
Quote:


Sealed, why? BTW that's a DSL TH-112 if you are wondering. LOL.

Yes, I know thats your TH-112
post #364 of 862
Thread Starter 
Quote:


The thread loses continuity

Continuity? what continuity ?
post #365 of 862
Thread Starter 
Quote:


Just got my TD15X's but haven't started make a cabinet yet. Patience, patience....................


hehe, just bored on vacation
post #366 of 862
I've done quite a bit of experimenting with the 6ND410 over the past few days here. I'll have some curves to post up soon. I believe we'll be able to make this work fairly well, at least until we can make our own 6.5".

Initially we have about 3dB or so we can lose on the 6ND410 to match well to the tweeter and mids. This will make the passive Xover much easier. We coated the cone, dustcap, and surround to help control the breakup that is pretty significant around 4KHz. There is a clear resonance seen in the impedance curve that appears to be due to the surround resonance. I fully filled the valley between the 2 surround rolls. This is a trick that EV has been doing for years. Currently waiting for this coating to dry, then I can move forward.

Mark also pointed out that a dip I had in the response of the mid around 600hz corresponds quite well with the distance from the backwave of the driver to the back of the cabinet and back out. While fiberglass insulation works well to absorb the higher frequencies, it needs to be more solid to absorb the low frequencies. A piece of blackhole5, some soft foam, and tighter packed insulation cleared up this issue quite well.

Should have more to report by the end of the day.

John
post #367 of 862
Quote:
Originally Posted by penngray View Post

I can do that still ? I should!

But I can not change the title of the thread, correct?

I've done it several times
post #368 of 862
Hey guys,

I thought about what Thy has written about doing ported reducing cone excursion (less IMD, better mids blah). I want sealed (various advantages for my case solely), but of course one can sock the vent and have flexibility here.

I have been playing WinISD and I am wondering if a 80L vented with vent tune of 60Hz + PEQ 90Hz Q=1 -3dB, 2 x 3" vent would work. Any audible problems with an arbitary tune since its EQed? Anyway its no a biggie if its sounds screwy , since I could stuff the vent. But I guess to maintain aesthetics I would back port it.

With venting, I maintain excursion of just 1mm down to 80Hz, 25W, 112SPL and with sealed i hit 2mm at 80Hz.

Again i need to mention i don't wanna dig low.
post #369 of 862
Wow, here's a possible midrange depending on your taste.
Eminence LA6-CBMR
http://www.partsexpress.com/pe/showd...number=290-542

I perfer a BBC dip and looking at this speaker's FR, a notch filter on the top end at 3.5k, right at the XO point, will bring down the 1.5-5k (depending on tweeter XO). It currently ships for free too.
post #370 of 862
After some playing around today, I've pinpointed the issue with the 6ND410. Or at least the big issue. If you look at impedance curves for almost all drivers, the LA6-CBMR that Looneybomber just posted the link to, the B&C 6MD38, etc, they all have a bump at somewhere around 4500hz. The 6ND410 published curve doesn't have it, but it sure is there when measured. In the case of the 6ND410, and I'd guess probably the others as well, this is a dustcap issue. The resonance of the dustcap corresponds to the spike in the impedance curve and a breakup of as much as 10dB in the response curve. You can cut it out with a notch filter, but the resonance is still there so it's going to sound harsh. Adding 6 grams of rope caulk to the dustcap totally removes the resonance and it sounds MUCH better. I didn't measure the response again as I was only trying to remove the resonance which I could view in the impedance curve. The problem is that 6 grams of rope caulk doesn't exactly look good and it knocks the efficiency way down to about 93dB.

So, one of 2 things will solve the problem. One, remove the dustcap and put on a slightly larger soft cloth dustcap that doesn't have the resonance issue. I have a feeling their published response curve was simply done of the driver prior to putting on a dustcap. I've seen many cases where this is done, as drivers are often tested prior to a dustcap going on in case there is something stuck in the gap that can be blown out. We'd simply buy the 6ND410 and cut the dustcap off. We'd put the new soft cloth dustcap on, coat the cone, and fill the valley between the rolls in the cloth surround with the damping compound to remove the rest of the breakup. In the end it would be a very smooth driver that is now 95dB(vs 97.9dB stock) which would be prefect sensitivity to mate to the tweeter passively.

The other option, which I'm going to explore more tomorrow would be to actually have a customized version of the 6ND410 made without a dustcap. Then we can actually fit it with a phase plug which totally eliminates the dustcap issue. This would be a great solution for us to offer until we are in a position to build a 6" driver from the ground up.

John
post #371 of 862
Here are some measurements now of the impedance of the 6ND410. This first is the driver stock with no modifications.


A few things are of interest. the big spike at just over 4KHz is the breakup of the dustcap. It is the reason for the nasty peak you can see in the response curve at the same location below. Notice that although the specified Fs is 120hz, this one is actually a little over 200hz. In this case it's a non-issue.



So we started working to get rid of the dustcap issue. A few experiments confirmed that the cause was the dustcap. In the followling image you will see 3 curves. The red curve is the driver with just the damping compound on the dustcap. you can see that the peak does change, but doesn't really go away. The blue graph is adding 6grams of mass to the dustcap. This smooths the peak, but pushes it lower in frequency, right into the range we want to use it. Not good. But both of these tests confirm the dustcap is the issue.The yellow curve is as we remove the dustcap totally. We have a little bumpiness in there, but no more big peaks or spikes. We also take off a little mass, meaning we can add more damping to the cone or surround where it is beneficial. I'll glue on a soft dustcap shortly and take some new measurements of the response curve. The overall results should be much better than the stock driver.



This little thing is now sounding quite good now compared to the stock driver, and is level matched to where we need it.

John
post #372 of 862
John, what about applying less damping material to the cone and then using a heavier/more dense dust cap? mms should be close to the same?
post #373 of 862
Quote:
Originally Posted by Looneybomber View Post

John, what about applying less damping material to the cone and then using a heavier/more dense dust cap? mms should be close to the same?

The problem is that the dustcap is going to resonate one way or another. The key is to control it as much as possible. Heavier/more dense dustcaps will lower the magnitude of the resonance, but push it into the intended range we're wanting to use the driver. The key is to get it as light as possible and as soft as possible. This puts the resonance higher and damps it more.

Another driver I want to look at is the Beyma 6MI90. It looks to be a good option. It has the flat foam surround like the PHL. It also has a phase plug so there would be no dustcap breakup which is the common problem in all the other drivers.

John
post #374 of 862
Quote:
Originally Posted by John_E_Janowitz View Post

The other option, which I'm going to explore more tomorrow would be to actually have a customized version of the 6ND410 made without a dustcap. Then we can actually fit it with a phase plug which totally eliminates the dustcap issue. This would be a great solution for us to offer until we are in a position to build a 6" driver from the ground up.

John

The B&C 6PEV13 has a phase plug. I can't get the PDF file to load for me right now, but IIRC, the sensitivity isn't quite where it should be at 500hz and rises upwards of 100db at 4-5k? (I could be totally wrong since I can't view the PDF)
Has this driver, and the Audax been ruled out?

How about the B&C 8PE21 (along with a lower XO point for the tweeter)?
post #375 of 862
The B&C 6PEV13 was disqualified as it has apparently been discontinued. It's now only in the archived products section on their website. If we were to move to an 8" and move the Xover point lower, we'd be in a range to just use the 10" TDM as the off axis response wouldn't be much different.

In reality, the MTM with pair of TD10M's and the ribbon tweeter really does sound substantially better than the 3way as Jason had mentioned. I'm having a very hard time justifying the addition of a midrange to the system here as we just haven't found one that works as well as one of the TD15's, TD12's or TD10's. The PHL was better response wise compared the unmodified 6ND410, but in reality it's very poor yet compared to any of the TD drivers. In the following curve, the yellow is the 18sound unmodified and the red is the PHL:



In the 6nd410 you can clearly see the nasty breakup from 4-5KHz. There is a large dip from 6-7KHz and another spike at 8KHz. Although they can be EQ'd out in the response, they are caused by a resonance that still doesn't go away. A resonance causing a dip is much easier to tolerate than a peak as it will simply lower the output in that region. The PHL is better, but there is a rise from 3k-8k that also is associated with cone/dustcap resonance. It's a much better damped resonance, but a resonance regardless.

Now compare that to a curve of the TD15M.

No peaks, very flat to 4KHz with a narrow dip above that and then quite smooth rolloff. Impedance curve has some tiny bumps but nothing as significant as either of the smaller drivers:


At this point I need to keep looking for a little bit. In the case of the 3way system, running the pair of TD12S's up to 3KHz sounded significantly better than having the 6ND410 on. The off axis response may not be quite as good, but on axis is much better. I'll see what it looks like with the modified 6nd410 tomorrow. It's hard to put in a midrange driver that has 2-3x the inductance of the woofer, no shorting ring, lower Xmax, etc. The TD10M's with NeoPro5i has proven to be extremely good sounding. I'm considering doing some kind of traveling speaker audition where I could ship a pair to a few people to try out in their own homes. Need to work out the details first as there would need to be some kind of deposit that is refunded or applied to driver purchase.

In the meantime I'll get one of the Beyma drivers on the way also.

John
post #376 of 862
John,

I will ask a few questions since I do not know this:

1) Why are the graphs with the mid at one level compared to your speakers, based off of the scale on the left hand side
2) Wouldn't you be crossing over to the tweeter way before 3-4K HZ anyway (since you are mentioning break up in that region)?

On a side note I have been informed the ones I have coming sound very sweet being crossed to the PHL.
post #377 of 862
...... nuke
post #378 of 862
Thy, what's the name of that proto Salk spk using TD15? Can't find on the website.
post #379 of 862
...... nuke
post #380 of 862
Wonder what's the pricing of that speaker. Think its a TD12H.
post #381 of 862
Yes, it is a TD12H. The mid was a PHL 1070, I believe. It was a PHL Jeff had on the shelf, but he indicated that the 1120 would be an equivalent. It also uses the Morel MDT33 tweeter. I heard this speaker at the Indiana DIY meet in April. It was perhaps the best I've heard.

I was so impressed, I had to build my own version, which I will be starting on soon. It is designed to have a less massive appearance and save a little on driver costs without sacrificing quality. It will use the TD12H, the B&C 6MD38(pending testing) and the Peerless HDS tweeter. I have recently picked up a pair of 8" PHL 2520's(for $100ea.) that could be an option for the mid also, if the B&C does not work out. There is a thread for it started here, for anyone interested:
http://htguide.com/forum/showthread.php4?t=29988

I will be calling them the BaSSlines, in honor of the creators of the speaker that inpired it from Jeff Bagby and Salk Sound. Here is the latest plans for them:

 

WMT_v3.pdf 75.19140625k . file
post #382 of 862
Quote:
Originally Posted by thylantyr View Post

Salk's new prototype speaker designed by Jeff Bagby using Morel tweeters, PHL midranges,
Lambda TD15's are getting great reviews. Other designs using 'ribbon + PHL + Lambda'
I have seen also received great reviews.
Bah, it's just easier to promote the ribbon + Lambda 2 way if you
have a business to run

Sorry but this is Hall of Shame material I read here.

I'm not real sure what makes this hall of shame material, it seems to me he just wants to get it right before selling it, nothing wrong with that, everyone needs to experiment and it looks like thats where John is right now. The treated driver from PHL may work better, but there are always trade-offs he may be better off making one in house if he really wants to make money and achieve his goals.
post #383 of 862
dlneubec,

I read through the htguide thread by you and get your idea on the cost issue. What were the levels you guys were doing at the diy meet? Peaks at 110 for highs at LP?

Anyway, more options (future?)
http://www.usspeaker.com/beyma%20CP21F-1.htm
Or buy used JBLs 077/2405.

Lots of mentioning of best speaker with best sound at the show in the threads (note that there is none so in this thread of pennyray's). So what's the best speaker? Audiokinesis, Gedlee, Salk? Or DSL, SA, MS?
post #384 of 862
Quote:
Originally Posted by exojam View Post

John,

I will ask a few questions since I do not know this:

1) Why are the graphs with the mid at one level compared to your speakers, based off of the scale on the left hand side
2) Wouldn't you be crossing over to the tweeter way before 3-4K HZ anyway (since you are mentioning break up in that region)?

On a side note I have been informed the ones I have coming sound very sweet being crossed to the PHL.

The measurements were taken at different times and praxis will auto scale the measurements. The one is showing 6dB per octave vs 10dB on the other. I could measure both drivers at the same time with levels matched for a better comparison. The point though was that even these "good" 6.5" drivers really aren't up to par with the TD drivers.

In the case of the 3way system, the goal was to push the Xover to the tweeter up to 3KHz. If there is going to be any kind of breakup/resonance, it really needs to be 48dB down to be completely inaudible. This would mean with a breakup at 4KHz in the 6.5" driver, we'd need to cross 4th order at 1KHz or 8th order at 2KHz. Neither of which is practical. Crossing 4th order at 2KHz would be the steepest slope practical, but in that case, the dual TD10's performs significantly better.


John
post #385 of 862
Army,

Just a question here but I do not recall John mentioning the need to treat the PHL. From my understanding (very limited admittedly), is that the PHL's are some of the best mid on the market and folks do not seem to have problems integrating them into a three way (experienced folks that is).

Is the thought of leaving this out based off of construction costs incurred in making a three way?

Just thinking out loud here.

James
post #386 of 862
Quote:
Originally Posted by thylantyr View Post

I don't know why John can't get his 3 way design working well ... using 'ribbon + PHL + Lambda' considering it's one of the easiest designs to
integrate well, a no-brainer if you do biamp or triamp

Bah, it's just easier to promote the ribbon + Lambda 2 way if you
have a business to run

Sorry but this is Hall of Shame material I read here.

Thy,

I can get it working just fine with the 6ND410, or any other midrange driver with the required sensitivity. This issue is simply that I don't see a point in selling an extra $130 driver to someone to get worse performance. The PHL while it lasted just long enough to take one test seemed promising but they just aren't available anyway. After trying for months I have received no response from PHL, nobody seems to be able to ever get calls back from E-speakers, and Zalytron had 5 of the PHL drivers left before Nate ordered one to have it sent to me.

All of the possible midrange drivers have at least double the inductance of the TD drivers. They just don't play as high as cleanly. I have the whole system +/- 2dB from 100hz up to 20KHz with good phase coherency at both crossover points. It sounds fine and plays extremely loud. I'm sure people would listen to the 3way and think it sounds great, even with the 6nd410 integrated. The interesting thing though is that when muting the 6nd410 and taking the pair of TD12S's up to 3KHz it sounds much better. I'd welcome anyone to come up and hear for themselves. Auditioning something for yourself is always going to be the best way to decide what you need/want.

John
post #387 of 862
Quote:
Originally Posted by John_E_Janowitz View Post

The interesting thing though is that when muting the 6nd410 and taking the pair of TD12S's up to 3KHz it sounds much better.

Perhaps if you're within 2 degrees of the center axis...
post #388 of 862
Quote:
Originally Posted by exojam View Post

Army,

Just a question here but I do not recall John mentioning the need to treat the PHL. From my understanding (very limited admittedly), is that the PHL's are some of the best mid on the market and folks do not seem to have problems integrating them into a three way (experienced folks that is).

Is the thought of leaving this out based off of construction costs incurred in making a three way?

Just thinking out loud here.

James

Hi James,

The PHL is no doubt a good driver, but as I mentioned you just can't get them readily. Last I heard, Zalytron had 4 left with no idea when they'd get more. If you can get them, they still have over 2x the inductance than the TD drivers, much higher distortion at any given spl than with the TD drivers, etc. In the case of the MTM's with the TD10's there isn't a lot of difference in the off axis power response at 2KHz vs what there would be with a 6.5" driver. The main benefit to the 3way with pair of 12's and the midrange included is the extra output capability and a little better horizontal off axis. The other benefit is being able to rotate the mid/tweeter section 90 degrees to use as a center channel above or below a screen.

I'm not saying at all that the PHL's can't work well. I just have a hard time recommending someone to spend an extra $180 on a midrange that doesn't really benefit the system. If the MTM with dual 10's isn't what people are looking for, a single TD15M or TD12M run to 2KHz crossed to the NeoPro5i ribbon could be a very good option as well. This would be able to be rotated as well for a center channel.

John
post #389 of 862
Thread Starter 
Quote:


This issue is simply that I don't see a point in selling an extra $130 driver to someone to get worse performance.

For me its simply a size restriction for my one room...I can put a set of MTMs in my family room though (TD15, tweeter, TD15)

But I did create this thread to build speakers for my HT room and sadly I have very specific needs. Im still going to build the boxes for the TD12S since it models the best in a 1.7 cuft sealed box and with 250W I get I think 117 dB without room gain.

I will still have to find a high end tweeter/mid-range driver design that will match my TD12 boxes Im going to be building mid-august.

Quote:


single TD15M or TD12M run to 2KHz crossed to the NeoPro5i ribbon could be a very good option as well. This would be able to be rotated as well for a center channel.

Hmm....Im interested in not need the mid-range if this is true! Im all for not needing a mid-range driver but there is obviously mixed opinions about that. I guess the only answer here is for me to get something working and just listen to it.

btw, I do have the Center channel box rebuilt and it will fit the TD12 now.
post #390 of 862
Thread Starter 
Oh, I did model the TD12s and I think winISD showed a drop off above 3K so can I really run them high enough to work well with just the ribbon tweeter?
New Posts  All Forums:Forum Nav:
  Return Home
  Back to Forum: DIY Speakers and Subs
AVS › AVS Forum › Audio › DIY Speakers and Subs › DIY 3 way main speaker with 15" drivers??