|
Home » Sponsored » Pi Speakers » three π upgrade
| three π upgrade [message #60130] |
Wed, 08 July 2009 16:46  |
 |
Wayne Parham
Messages: 18787
Registered: January 2001
|
Illuminati (33rd Degree) |
|
|
I've been working with the TD12S woofer for a week now and I have some preliminary test data to report.
In short, the woofer looks good. Not only is it well constructed, it also measures and sounds good. Looks very nice too, a high quality part.
The bass response is nice and smooth in the three π cabinet. They're made for each other, couldn't ask for better electro-mechanical parameters. I haven't done high power testing yet so I don't know how much drift there is at higher temperatures, but my expectation is that drift will be lower than normal due to the attention AE has paid to thermal performance. They use aluminum to wick heat from the pole piece similar to the mechanism used in the 12π hornsub. One thing I noticed though that impressed me right away is the T/S specs are pretty close to what's quoted, right out of the box. Most woofers aren't.
The impedance curve is just like AE shows, smooth and flat up to the stratosphere. It doesn't rise until tweeter territory. From that, I knew to expect the Zobel would be reduced if even needed. Turned out, you just throw the Zobel away. That was weird for me, all woofers need conjugate filters, so I sort of wanted to leave it in place with a smaller capacitor. But you just don't need it, and it really shouldn't be there. The impedance is flat through the crossover band and well above it. So ditch the Zobel. It's a waste of money and response is improved without it when using this woofer.
The tweeter high-pass filter is the same as other models, third-order with mass-rolloff compensation but the woofer low-pass filter needs to be second-order. In fact, the crossover needed is almost exactly like what we were using in the 1990's and early 2000's. So if you have an old three π with the original crossover, you can just drop in the woofer and you're done. It will work right out of the box. Just toss the Zobel.
Naah, if you're going to splurge and get the new good woofer, swap out the crossover too. Get the new PCB and populate it with 15 guage coils and other good parts.
Anyway, here are the components you'll want:
B&C DE250 tweeter (note reverse tweeter leads, black to positive)
AE TD12S woofer (red to positive)
* tweeter crossover
C2 6.8uF
L1 1.0mH
C3 20uF
* tweeter compensation
R1 25Ω (20Ω R1a and R1c, 25Ω R1b and R1d)
R2 16Ω (16Ω R2a, R2b, R2c and R2d)
C1 0.47uF
* woofer crossover
L2 1.5mH
C4 20.0uF
L3 (jumper wire)
No Zobel, so nothing in positions R3 or C5
Here's a little video I did that shows some of my tests, and where the vertical nulls are on this model with the crossover described above:
Watch the response curve on the laptop computer, lower right of the video. The S&L measurement system is sending a series of bursts to the speakers, and the microphone captures the signal. When the microphone is positioned anywhere between the vertical nulls, the response curve is basically flat. (Unless I'm talking, of course  )
When I move the microphone to the bottom edge of the speaker, you'll see a notch form in the response curve. That is the lower null. Later, I move the microphone to a position beyond the top edge of the speaker. Watch the response curve as I move the microphone, you'll see it remains flat until I reach the upper null, where a dip again forms.
You might also notice that when I moved the microphone to show the lower null, I moved it just a smidge too far - just past the deepest part of the null. But you can still clearly see the upper and lower nulls, and the smoothness of response in between. Pay attention to the response curve as I move the microphone, you'll see it stays nice and smooth all the way between nulls, over a wide vertical range.
I also want to do some distortion measurements, just to compare the TD woofer with JBL 22xx woofers. Both are really nice, and I think we have a winner either way. But I'd like to see how these two family of woofers stack up in terms of harmonic distortion, because that is an indirect measurement of motor linearity.
More to come...
|
|
|
|
|
|
|
|
|
|
| AE TD12S [message #60388 is a reply to message #60364] |
Sun, 12 July 2009 13:33  |
spkrman57
Messages: 522
Registered: May 2009
|
Illuminati (1st Degree) |
|
|
Wayne,
Found this thread and wondering if you have a link to the AE TD12S speaker specs and pricing/availability.
I did a quick google search that netted me very little info.
And I agree that for 15" duty (4-Pi pro) that the JBL 2226 will reign as the best driver for the price.
Regards, Ron
|
|
|
|
|
|
| Re: AE TD12S [message #60415 is a reply to message #60389] |
Fri, 17 July 2009 12:35 |
spkrman57
Messages: 522
Registered: May 2009
|
Illuminati (1st Degree) |
|
|
Thanks Wayne for the link!
That's a nice driver, my only problem with it is the efficiency of 93.2 db/watt is kind of low for my tube amps.
It has a hefty price tag too!
I'm sure there are some speaker builders for which it will be a great driver to work with.
Regards, Ron
|
|
|
|
| Loudspeaker sensitivity and boundary loading [message #60416 is a reply to message #60415] |
Fri, 17 July 2009 13:35 |
|
Wayne Parham
Messages: 18787
Registered: January 2001
|
Illuminati (33rd Degree) |
|
|
I wondered about the sensitivity too, and how much padding I'd need to match the tweeter. What I found is you don't pad down the tweeter any more than you do with the Delta 12LF. It's acting like its more like 95dB-96dB when baffle mounted. I am anxious to measure the system outdoors with LMS to get an accurate SPL number.
I'm betting the 93dB figure from AE is a freespace value (which makes sense for a raw driver). But you'll never realize this level in actual use. Eminence and JBL list half-space sensitivity mounted on a baffle (which is OK too, it's clearly stated).
Speaker cabinets the size of these transition from freespace to half-space (baffle step) approximately at the same frequency as the room's Schroeder frequency. The baffle sets radiation to half-space through most of the passband. So everything above the Schroeder frequency is radiated half-space, and below that you have boundary conditions and room modes, lots of stuff going on. So basically, the speaker in the box really can't radiate freespace. It would have to be in a tiny box with no baffle front area.
I haven't had a chance to measure outdoors, anechoically. My work so far has been crossover design, setting the frequency and slope to get good response on and off-axis through the crossover region. That means I really don't have any bass measurements yet. But I did use the Smith & Larson WTPro system to measure T/S and in-box impedance, which can then be used to make a very accurate model, estimating system response at low frequency.
The WTPro measures the T/S specs of the woofer and then can measure the impedance of the woofer in the box too. I have to do that to get the ZMA file for the ICD. One of the other cool things it does is to do a BoxPlot style bass response simulation, but with the added accuracy that it is using actual measurements of the woofer in the box. This is what it shows:
I expect anechoic will show a little more rolloff below 200Hz because of collapsing directivity as the baffle starts to transition to half-space. I don't ever use BSC because in room, the power response tends to dominate. Actually, room modes dominate but you know what I mean. In room there's a lot more going on than just collapsing directivity from the baffle transition. I see BSC as more of a way to voice a speaker than anything else. Still, anechoically, I expect some extra rolloff from collapsing directivity due to baffle dimensions.
Boundary conditions almost always make the speaker act like its operating in half-space to quarter-space, even if not on the ground and directly back up against the wall. Only if stand mounted in the middle of a very large room does the speaker act like its in freespace, and even then the baffle reduces directivity to half-space over most of the audio band.
Bottom line is I'll bet the TD12S is more like 96dB in the box. It's matching the tweeter in the crossover region, and that's around 96dB/M. Notice how well the woofer and tweeter are matched in the "Vertical Nulls" video in the first post of this thread. Since we know the padded tweeter is 96dB/M, that tells me the system is probably around 95dB-96dB. I'll know for sure when I get the LMS results.
|
|
|
|
|
|
| Re: AE TD12S [message #60420 is a reply to message #60417] |
Sat, 18 July 2009 10:21 |
|
Wayne Parham
Messages: 18787
Registered: January 2001
|
Illuminati (33rd Degree) |
|
|
I haven't had a chance to live with them, and to me, that's as important as the first hour or two of listening. But to be honest, I probably won't post subjective opinions comparing the JBL to the AE models. I like and respect both too much to set them against each other in a subjective comparitive review.
I will tell you this: Both are top-notch, doesn't-get-any-better-than-this speakers. One is from a large company that has the resources to model, test and design to the Nth degree. The other is from a small niche company that makes quality machining, fit and finish - and sound - a highly personal labor of love. Both hit the target, dead-on bulls-eye, in my opinion. They make my job easy, designing a good loudspeaker with these parts, because they act as expected and there are no anomalous "warts" to design around.
Having both, one in the three π and six π and the other in the four π and seven π models, allows the discerning audiophile choose their favorite "flavor". You can't go wrong with either one.
|
|
|
|
Goto Forum:
Current Time: Sat Nov 23 03:57:10 CST 2024
|
Posting Rules
Members
Search
Help
Register
Login
Get a T-Shirt!
