Seas DXT Tweeter, Very Interesting!

[ridikas]

I came upon this little guy from Seas. The 27TBCD/GB-DXT (H1499-06). http://www.seas.no/index.php?option=com_content&task=view&id=184&Itemid=179

I see that Seas has the Idunn kit and there's a DIY design by MarkK, http://www.audioheuristics.org/projects_gallery/ER18DXT/ER18DXT.htm

I have a few questions for the gurus around here:

- Any other DIY examples? If not, how come there's not so much interest in this tweeter?

- I understand that this is not a full waveguide (as we would want it to be) and is only roughly 4" in diameter. What frequency is it effective to? I'm guessing 2kHz-2.5kHz maybe?

- Am I correct in assuming that this tweeter will give a very nice power response?

- What is the correct way of crossing a waveguide tweeter? Let's say the guide is effective to 2.5kHz... Does it make sense to cross the tweeter over at 1.8kHz? Or do you always want to cross it over at where the guide still has some effect?

- And lastly, can it be surface mounted? That is if crossed over at, or higher than the guides effective frequency range. I assume that if it's crossed over below this frequency, diffraction will begin to come into play.

Thanks for the responses :)

[Dirk]

I'd *like* to say that this tweeter could be crossed with nothing but a 1st order electrical filter plus some impedance compensation, but that impedance signature is a strange one. Maybe a parallel LR?

There are good reasons not to surface mount this tweeter. ONE is that the acoustic center is pretty far back (this dovetails nicely with your Dynaudio thread). Also, I don't think the horn loading will be sufficiently low enough to get below the region of baffle diffraction ripples.

[Dirk]

- What is the correct way of crossing a waveguide tweeter? Let's say the guide is effective to 2.5kHz... Does it make sense to cross the tweeter over at 1.8kHz? Or do you always want to cross it over at where the guide still has some effect?

From what I've been able to gather (and I've not read NEARLY as much as some around here), where the guide still has affect. You may find this read useful:

http://zaphaudio.com/hornconversion.html

It provides a great picture of exactly what the waveguide does to an otherwise normal tweeter.

[Paul Ebert]

There's a fr curve with off axis curves here:

http://www.zaphaudio.com/offaxis.html

The off axis behavior is excellent - one of the reasons I like it. I don't know why it's not more popular.

- Paul

[ryanbouma]

This is a nice tweeter, but there are better tweeter/waveguide combos out there, lots that cost less too.

ND28 and a 150mm x 60mm waveguide:

http://i874.photobucket.com/albums/ab306/tuxedocivic/TN%20Allen%20waveguides/ND28insmalloval.png

Or an SB/Dayton combo that would cost around $60:

http://i874.photobucket.com/albums/ab306/tuxedocivic/Challenger/tweeteronaxisandwithcap.jpg

And off axis looks good on those too :D

[CUtiger]

I'm currently using this tweeter paired with an usher 8945A, and honestly I'm not in love with it. Don't get me wrong it measures well and does nothing 'wrong' it just has never wowed me. I'm going to give this crossover one more redesign but this tweeter has never captivated me in any of the half dozen iterations of this pairing I've tried.

[CUtiger]

I'd *like* to say that this tweeter could be crossed with nothing but a 1st order electrical filter plus some impedance compensation, but that impedance signature is a strange one. Maybe a parallel LR?

There are good reasons not to surface mount this tweeter. ONE is that the acoustic center is pretty far back (this dovetails nicely with your Dynaudio thread). Also, I don't think the horn loading will be sufficiently low enough to get below the region of baffle diffraction ripples.

I wanted to think this too but unless you're crossing high, this tweeter has been a royal pain to hammer into a target rolloff, I'm not an x-over guru but it has easily been the trickiest driver to work with for me. Partly due to the strange resonance but simple filters tend to leave some tiny peaks in the low treble upper mids which I didn't think sounded pleasant.

[Pete Schumacher ®]

I wanted to think this too but unless you're crossing high, this tweeter has been a royal pain to hammer into a target rolloff, I'm not an x-over guru but it has easily been the trickiest driver to work with for me. Partly due to the strange resonance but simple filters tend to leave some tiny peaks in the low treble upper mids which I didn't think sounded pleasant.

Have you looked at what MarkK did to shape the driver's response to fit his target? Believe it or not, that tweeter network is a first order crossover with an LR2 target response.
http://www.audioheuristics.org/projects_gallery/ER18DXT/images/ER18DXT-final-crossover.gif

[Dirk]

Have you looked at what MarkK did to shape the driver's response to fit his target? Believe it or not, that tweeter network is a first order crossover with an LR2 target response.


Twin LCR shunts, and neither where I would have expected them. :confused:

This is why I'm barely more than the proletariat.

[RINNAV]

I'm currently using this tweeter paired with an usher 8945A, and honestly I'm not in love with it. Don't get me wrong it measures well and does nothing 'wrong' it just has never wowed me. I'm going to give this crossover one more redesign but this tweeter has never captivated me in any of the half dozen iterations of this pairing I've tried.

I agree… I was excited about this tweeter at one time until I heard it in a couple of applications and it never really impressed me—subjectively, it just seemed so-so and lacked any hint of vibrant qualities.


So the question is, what do you hear by looking at the plots ridikas?

[GranteedEV]

FWIW, these commercial speakers:

http://www.grimmaudio.com/pro_loudspeakers_ls1.htm

use that tweeter.

But I think overall there's better choices. An RS28 or SB 29 in a SEOS-6 probably outperforms it, or a lot of the custom waveguides here at tech talk.

[markk]

A couple of things worth commenting on.

Are Seas overpriced? A bit I suppose. But the QC is second to none. As good as scan. Virtually everyone else, Usher, Dayton, TB, vifa, peerless, you name it has a bit more unit to unit variance. Is that worth the extra money?

Second, the waveguide is quite small. Really, too small for my tastes. But the design works because the dip in power response at the xover frequency is carefully matched, effectively extending the uniform power response lower. '

This is not an instantly easy driver to design around. It takes a certain amount of perseverance and thought to get the most out of a design using this driver.

If there was a readily available 6-8 inch WG that bolted on to a nice dome without milling, a custom cnc or lathe involved, well, I'd like that too...

As far as the SEOS drivers-yes the graphs look awesome. Well, where are they? I can't order them on the web site. I pm'd the person who was doing these-never heard back. I'm sure he's a nice guy. But my point is that a bunch of the WG's you all post about are really not really available. It's nice posting about custom and semi custom WG's and how nice they are. But until I can order them online easily in a fashion that's drop in on a baffle, well...

[Dirk]

It's nice posting about custom and semi custom WG's and how nice they are. But until I can order them online easily in a fashion that's drop in on a baffle, well...

^-- This!

I am really interested in giving one a shot, but I'm fairly certain that I'm too hamfisted to modify an existing WG accurately--OR safely. The best I can think of is a jig under a router table and way too much trial-and-error.

[GranteedEV]

As far as the SEOS drivers-yes the graphs look awesome. Well, where are they? I can't order them on the web site. I pm'd the person who was doing these-never heard back. I'm sure he's a nice guy. But my point is that a bunch of the WG's you all post about are really not really available.

They are available, though. It's a group buy site, so there's some patience you'll need as the pallets stack up. Did you try emailing him?

http://www.diysoundgroup.com/contacts/

[ridikas]

To the people that don't care too much for this tweeter: Is it because it sounds dull, or lacks that top end air and sparkle? Which gives 3 dimensionality to sound?

I think that can be very easily taken care of in the crossover. A simple L-Pad can shape the top end to whatever one desires. The Seas Idunn kit does just that: http://seas.no/index.php?option=com_content&view=article&id=274&Itemid=248

No droop after 10kHz whatsoever. In fact, it has a slightly rising top end. I've modeled this tweeter in PCD and by changing the values of the L-Pad I can contour the top end to my liking. This is actually the same issue that some people have with the drooping Vifa DX25TG tweeter. Which can similarly be fixed with the right L-Pad.

If this is the reason why so many people dislike this tweeter, then it can easily be fixed :) Or are there other reasons, which are not so easy to fix? The brain might also be playing tricks after seeing that drooping SPL plot at Seas.

It should sound very similar (because it practically is) to the Seas 27TBC/G (H1147) and it's brother 27TBFC/G (H1212). And I've never heard anyone not like these tweeters, nor complain that they sound dull, or lack excitement.

[Dirk]

It should sound very similar (because it practically is) to the Seas 27TBC/G (H1147) and it's brother 27TBFC/G (H1212). And I've never heard anyone not like these tweeters, nor complain that they sound dull, or lack excitement.

I have the H1212 in my main pair. Honestly...I don't find them that exciting. Also, they have a naturally DROOPING top octave. There's a little plastic lens affixed to the inside of the mesh cover that limits beaming and, consequently, the top octave, too.

Regardless, the top octave is over-hyped. 5k-10k has a lot more to do with the sound of a driver. We humans CAN hear stuff up there, but we do a horrible job of localizing the sound. Our ears are too big.

[ryanbouma]

It's nice posting about custom and semi custom WG's and how nice they are. But until I can order them online easily in a fashion that's drop in on a baffle, well...

Well, I'd like to see that change.

http://techtalk.parts-express.com/showthread.php?p=1810105#post1810105

[CUtiger]

To the people that don't care too much for this tweeter: Is it because it sounds dull, or lacks that top end air and sparkle? Which gives 3 dimensionality to sound?

I think that can be very easily taken care of in the crossover. A simple L-Pad can shape the top end to whatever one desires. The Seas Idunn kit does just that: http://seas.no/index.php?option=com_content&view=article&id=274&Itemid=248

No droop after 10kHz whatsoever. In fact, it has a slightly rising top end. I've modeled this tweeter in PCD and by changing the values of the L-Pad I can contour the top end to my liking. This is actually the same issue that some people have with the drooping Vifa DX25TG tweeter. Which can similarly be fixed with the right L-Pad.

If this is the reason why so many people dislike this tweeter, then it can easily be fixed :) Or are there other reasons, which are not so easy to fix? The brain might also be playing tricks after seeing that drooping SPL plot at Seas.

It should sound very similar (because it practically is) to the Seas 27TBC/G (H1147) and it's brother 27TBFC/G (H1212). And I've never heard anyone not like these tweeters, nor complain that they sound dull, or lack excitement.

No I don't think so, I'll have to remeasure since my system crash lost most my files but in my latest version I tried a similar approach and the top octave and a half is actually lifted a few DB from reference. Don't get me wrong this is my reference system but the top end is just so so in a subjective sense compared to other areas.

My original plan was to try and get LR2 acoustic similar to Mark but have moved it to LR4 closer to 1700. I've since invested in a legitimate measurement rig and will likely give the original approach one more shot after I finish up my next project. Can't disassembled it till then or else I'd have to listen to something else, so I guess it's not all that bad ;)just not stellar.

[dlr]

- Am I correct in assuming that this tweeter will give a very nice power response?
As Mark mentioned, it's in the application of it. What is probably more important than the power response of the tweeter is the power response of the system in which it is used. That's a design issue, one Mark addressed.


- And lastly, can it be surface mounted? That is if crossed over at, or higher than the guides effective frequency range. I assume that if it's crossed over below this frequency, diffraction will begin to come into play.
Diffraction can still come into play higher than one might think.

I'm using the DXT in my current 3-way, but it is crossed with the Ultimate Equalizer, PC-based, so the difficulty was not in creating the crossover, it was determining which was optimal. I ended up with an LR8 @ 1200Hz because it provided by the best horizontal polar (hence power) response. This is coupled with a SS 15W/4531G as a dipole midrange. The other benefit is that the polar response in the front hemisphere approximates the same shape as would a true dipole tweeter.

I'm very pleased with this setup. I haven't encountered any distortion issues running it this low due the LR8. I haven't checked what it would take to do the LR8 passively.

dlr

[fjhuerta]

As Mark mentioned, it's in the application of it. What is probably more important than the power response of the tweeter is the power response of the system in which it is used. That's a design issue, one Mark addressed.


Diffraction can still come into play higher than one might think.

I'm using the DXT in my current 3-way, but it is crossed with the Ultimate Equalizer, PC-based, so the difficulty was not in creating the crossover, it was determining which was optimal. I ended up with an LR8 @ 1200Hz because it provided by the best horizontal polar (hence power) response. This is coupled with a SS 15W/4531G as a dipole midrange. The other benefit is that the polar response in the front hemisphere approximates the same shape as would a true dipole tweeter.

I'm very pleased with this setup. I haven't encountered any distortion issues running it this low due the LR8. I haven't checked what it would take to do the LR8 passively.

dlr

Thanks! I was wondering how low I could cross these units. I'll be using a MiniDSP with them, I'll aim for that frequency (the lower, the better!)