Seas Tweeter tbcd/gb-dxt

[Mark K]

I'll be getting a pair of these shortly and will take a look. I'll try them with the TDC, which is also a nonferro, but admirably performing motor, in terms of distortion.

[jkrutke]

Correct, I bought a H1212 pair to swap motors with the DXT. I should have said a 27 series motor w/ ferrofluid (edited post above to reflect that fact).

Well, looks like the same dome mounting method, but the DXT has the T25's rear cup and I think an extra magnet inside. The interesting thing to note is that the distortion profile follows the motor much more than the dome. Put anything on a TBFC motor and that's how it behaves, on the low end at least.

The DXT's motor isn't bad, but I'd prefer the smoother low end rolloff of the ferrofluid version. I might even like the DXT with the 29 series motors, no ferrofluid and no rear chamber, but smooth and steeper low end rolloff. I wouldn't be surprised if the DXT could hit an LR4 target at some low frequency with a cap and impedance comp across the terminals.

Jed, I'll be paying attention to your results with the H1212 motor. Then I don't have to do it.

[jkrutke]

I'll be getting a pair of these shortly and will take a look. I'll try them with the TDC, which is also a nonferro, but admirably performing motor, in terms of distortion.

That should effectively point out the differences from the rear chamber and the different pole opening stuffing. (foam vs cotton plug) I'll watch for those results too. Lets us know in this thread.

I guess I'll volunteer to try the 29 series motor and that should round out all possibilities.

[ch83575]

Making a little bit of progress. Here is the DXT's distortion expressed as a percentage. I tested it in a 20L Madisound box.



Looks pretty much like John K.'s results in terms of non-linear distortion. I haven't tested the impedance or IB FR.

It's not bad for a non-ferrofluid tweeter but definitely not state of the art below 2k in terms of 3rd order harmonic distortion. I will take John's advise and try this tweeter with a 27 series ferrofluid motor. That should really improve the low end and allow a nice 1.5kish 4th order slope. I think Seas recommendation of a 2k Fc is spot on and this tweeter will perform really well in that regard.

Jed

It seems to me that this is one of the few times that showing the distortion as a percentage makes the driver look worse than it really is. Once the low frequency hump has been corrected flat (ie. in an actual design) that 3rd should drop dramatically right?

-Chad

[Mark K]

It seems to me that this is one of the few times that showing the distortion as a percentage makes the driver look worse than it really is. Once the low frequency hump has been corrected flat (ie. in an actual design) that 3rd should drop dramatically right?

-Chad

It's still 40dB down at 1k, if I'm reading correctly. Depending on the level, that could be very good, or not. I wonder what the spl level is?

[Jed]

It's still 40dB down at 1k, if I'm reading correctly. Depending on the level, that could be very good, or not. I wonder what the spl level is?

About 95db at 1 meter.

[fjhuerta]

The seas DXT tweeter is speced out to 2k. Should be no problem at all. You could probably go much lower.

I'll let you know how it sounds this next week but I don't think anyone has actually listened to it with music yet.

Jed

Thanks Jed. The whole idea of changing the motor for the 27TBFC sounds very intriguing. Hopefully, the 27TBFC will like the non-ferrofluid motor... or else, you end up with some expensive parts left behind!

[Jed]

Stock TBCD/GB-DXT at about 90DB 1meter. All tweeters were tested under same conditions.



Stock 27TBFC/G


27TBFC/G Dome on TBCD/GB-DXT motor


And what we are all waiting for,
TBCD/GB-DXT dome on 27TBFC/G motor


I took the 2 best profiles from above and merged them, so I can easily compare the differences. Below you can see the stock 27tbfc-g tweeter and the DXT dome unit on a 27tbfc-g motor. I labeled each set of harmonic distortion for the DXT dome unit on a 27tbfc-g motor as "DXT". The DXT hybrid outperforms the stock 27tbfc-g tweeter in all non-linear distortion tests except D2. With the benefits of a waveguide it will be even better.

[augerpro]

Hybrid looks great! The price not as much...

[Jed]

Hybrid looks great! The price not as much...

Well, it's all relative. When you consider it probably blows away most boutique tweeters it's a bargain. I'm going to see if I can sell the hybrid 27tbcd/gb non wave guide tweeters on diymobileaudio to make up for some of the cost. Even that tweeter isn't that bad and is useable for a nice 2 way.

[Silversmoky]

That's pretty cool that worked so effectively! Looks like a great tweeter. Is it surprising that Seas itself doesn't market this combination?

[peter_m]

Pretty cool...

Still wonder if Zaph's 6.5" waveguide over a 27TDFC is not more economical then all this tweeter surgery?

[augerpro]

Well, it's all relative. When you consider it probably blows away most boutique tweeters it's a bargain. I'm going to see if I can sell the hybrid 27tbcd/gb non wave guide tweeters on diymobileaudio to make up for some of the cost. Even that tweeter isn't that bad and is useable for a nice 2 way.

So you have a 2 way in mind using the DXT and the new metal cone Peerless Exclusive?

[Jed]

Pretty cool...

Still wonder if Zaph's 6.5" waveguide over a 27TDFC is not more economical then all this tweeter surgery?

Economical, yes. I haven't compared the performance and plus the DXT faceplate looks cool. As if that matters.

That's pretty cool that worked so effectively! Looks like a great tweeter. Is it surprising that Seas itself doesn't market this combination?

Well at least it gives us something to play with. :B Honestly, I was a bit suprised they didn't offer it. This was really easy to do. Remove 4 screws, pull the faceplate off, and put it on the other motor. Done.

So you have a 2 way in mind using the DXT and the new metal cone Peerless Exclusive?

Remember those Seas W15CYs I was testing a few weeks back? I'm starting to pile up drivers again, so I need to try out some new combinations. That said I'll probably have to try the new Peerless Exclusive when it comes out. The cones remind me of the Seas L16.

[Mazeroth]

Looks great, Jed!

Any chance you have frequency response graphs for the before and after and comparing the stock DXT to the DXT with new motor?

Thanks for doing this. :T

[fjhuerta]

Thanks Jed! I'm now seriously thinking of using the hybrid Seas in my next design.

I have a question regarding ferrofluid... did you just swap the domes? Do you have to refill the magnet with ferrofluid (because there's always some left on the tweeter's coil)?

Do you think the same results would be achieved by filling with ferrofluid the DXT's magnet? I think I remember I saw once ferrofluid being sold separately.

[Jed]

I have a question regarding ferrofluid... did you just swap the domes?

I just swapped the domes and those are the results. I don't plan on adding more ferrofluid in the gap, but the idea of adding ferrofluid to the stock DXT design sounds interesting.

[peter_m]

Adding ferro-fluid sounds pretty straight forward. Wonder why Seas hasn't produced such a model?

[jkrutke]

I just swapped the domes and those are the results. I don't plan on adding more ferrofluid in the gap, but the idea of adding ferrofluid to the stock DXT design sounds interesting.

The only real problem: you can't buy the proper ferrofluid. All that's available is the educational kits, which is the wrong viscosity and composition. (it is far too thick and dries out) My understanding is that FerroTec is very picky about who they sell to, and will not touch end users or sell in small quantities. I'm looking into it however.

[peter_m]

The only real problem: you can't buy the proper ferrofluid.

So if one would sneeze while changing a blown tweeter diaphragm... one would be screwed?

[jkrutke]

So if one would sneeze while changing a blown tweeter diaphragm... one would be screwed?

That stuff doesn't come out of the gap that easy. In fact if the voice coil is pulled out slow, almost all of the ferrofluid stays in the gap. If pulled out fast, a little more seems to go with the coil.

The bigger thing to worry about when changing coils is keeping garbage out of the gap. Dust, magnet chips, screw shavings, and more can wreak havoc on the tweeter's performance.

[fjhuerta]

The only real problem: you can't buy the proper ferrofluid. All that's available is the educational kits, which is the wrong viscosity and composition. (it is far too thick and dries out) My understanding is that FerroTec is very picky about who they sell to, and will not touch end users or sell in small quantities. I'm looking into it however.

I was under the (mistaken) impression that ferrofluid was available... I now see that, indeed, the ferrofluid that is being sold in most places is the "educational" stuff...

[dlr]

Stock TBCD/GB-DXT at about 90DB 1meter. All tweeters were tested under same conditions.



Stock 27TBFC/G


27TBFC/G Dome on TBCD/GB-DXT motor


And what we are all waiting for,
TBCD/GB-DXT dome on 27TBFC/G motor


I took the 2 best profiles from above and merged them, so I can easily compare the differences. Below you can see the stock 27tbfc-g tweeter and the DXT dome unit on a 27tbfc-g motor. I labeled each set of harmonic distortion for the DXT dome unit on a 27tbfc-g motor as "DXT". The DXT hybrid outperforms the stock 27tbfc-g tweeter in all non-linear distortion tests except D2. With the benefits of a waveguide it will be even better.

Filters here prevent me from seeing the curves but I've got a question separate from them anyway. I've made a number of my own hybrids, a couple that I've not posted at my site. I've wondered at bit about the different motors and voice coils when swapping. The issue is in the position in the gap.

At low levels that I use for SPL and impedance tests, the coil likely stays in the area where Bl is linear. However, at the extremes of displacement, near a tweeter's Xmax, I've thought that it would take a Klippel system to characterize the response to show if there is any asymmetry involved in a hybrid. Considering that Seas has done so well with distortion control without a copper cap in some cases, I'd be interested in seeing some results of very high level distortion testing that significantly displaces the coil. Anyone know just how much displacement is occurring in their distortion tests?

[Mark K]

Hi all,

I finally posted some test data on the DXT. See my measurements section.

[Jim Holtz]

Good stuff! Thanks Mark!

Jim

[jkrutke]

Hi all,

I finally posted some test data on the DXT. See my measurements section.

Looks good Mark. I see you had a little bit of the impedance peak variation that I had in my pair too. Those with impedance measuring ability can easily get the impedances to look the same with a little pole piece stuffing adjustment.

Both our results looked good enough for a 1.5Khz crossover. I have to wonder if Seas' conservative 2K rating is set from what they thought might be some sort of thermal limitation.

[rc white]

The Peerless India wave guide is specified as being driven from a 100 Ohm source, this would account for the published frequency response.

[jdybnis]

That's really interesting. It looks like they are running all their tweeters that way. Maybe a good idea. I'm guessing that their new tweeters all have a rising top end and the series resistance is bringing up the bottom end to match. Is this correct?

Does adding series resistance lower the sensitivity of a driver? They report their T26S-X is 94.5dB at 1w/1m, http://www.peerlessaudio.com/products_rd_list.htm which is pretty damn good for a dome tweeter.

[rc white]

What it does is to give psudo constant current drive. In this the frequency response follows the impedance curve. If the drivers frequency response with constant voltage drive resembles the reciprocal of the impedance curve this tends to flatten out the frequency responce, i.e giving more output near resonance and reducing output in the impedance suck out before voice coil inductance comes to the fore and increase high frequency output.
You can do this electronically by means of current feedback around a power amplifier, this is common in solid state guitar amplifiers.
rcw

[Dennis H]

Does adding series resistance lower the sensitivity of a driver? They report their T26S-X is 94.5dB at 1w/1m,

Yes, just putting a 100 ohm resistor in series would make the driver barely audible. It's a really bogus way to spec drivers as they must be measuring the voltage downstream of the resistor when they come up with their sensitivity spec. The only way to make it work is to fake it with a special amp circuit as rc white mentioned.