Need good drivers for 'copper' speaker cabinet

[bluewhale]

I would go with a couple of SLA 5-10 Ah batteries for your project. You could get a pair of SLAs and recharge one while the other is running.
..

No way to charge a Lead Acid at the same time you are running off of it?

Those panels look very nice. Thank you! I haven't found a durable looking one yet, until these.

The 5 watt model is probably the right choice since it may get cloudy and would still provide electricity at dusk.

 

[infinia]

Stumbling around on the Econo-wave threads, I found the Selenium compression driver. It looks like a decent step up from the APT-50, for not much more money.

Would going to the Selenium d220Ti improve things? Or is the lousy HF response a result of using the 10CX cone as the waveguide?

The Econo-wave thread(s) are informative, not to mention huge! I even got to thinking mounting the tweet and a waveguide on top of......

I think youre pretty much stuck with using an APT-50 unless you
1) Spend alot of time and money investigating the throat to horn transition with a different compression driver. I think that would be using a bolt on driver with custom threaded adapters.
2) Buy the custom adaire unit ie two Adire ACD1
1�� compression drivers - see http://www.audioxpress.com/reviews/media/203dell2119-1.pdf
3) do the external waveguide thingy on top

They don't publish the FR curves of the APT50 on the Eminence website, but some of their other tweeters have awful frequency linearity. I would think that the response has to do with the particular tweeter at the volume levels they test at.

Can we mount other tweeters in the Eminence coaxials? I would think so but one of us should call them.

Here is the graph that I found 10cx and apt-50 (post 38)... looks like it's on a baffle with maybe a standard text book crossover ie the low bass droop. No attenuation on the APT-50. But what worries me is from 550-1.5K Hz looks very problematic.

 

[ZilchLab]

The required adapter for the APT-50 converts it to standard thread mount. Get the 10CX and bring one and one of your cabs over here. I have APT-50, Selenium D220Ti, and a bunch of other suitable compression drivers, E'Wave crossovers, and CLIO to measure the works.

Yeah, I'm the EconoWave guy, I'm local, and I'll eMail you the number....

 

[infinia]

Originally posted by ZilchLab
Yeah, I'm the EconoWave guy, I'm local, and I'll eMail you the number....

Hi Zilch welcome
I'm sure you can help, and I hope you will come back and share lots of data for others here.
FWIW I suspect there will be issues with the 2 transitions between the (phase plug and adapter/horn entry) and (horn to woofer cone) that will limit what can be done. ie mostly limiting the lower crossover frequencies.

 

[kstlfido]

Thanks infinia for the link- that Adire kit review is pretty good insight on how the APT50/10CX combo works together. It's almost the same box size as mine, too.

Zilch- wow- thanks for the offer! Didn't realize we're so close.. Ok, I'll need to get the 10CX's. Parts Express site is down right now.

Have a good weekend, all!

-Kent

 

[bluewhale]

Looking at the specs, both the 8cx and 10cx have a frequency response rolloff around 1.5 kHz. Using one of the Eminence high frequency drivers that go down lower with a 1.6 kHz crossover might give better coverage

The Eminence people say that their other tweeters work.

Other drivers will work on the 8CX, but you need the screw-on versions, PSD:2002S and NSD:2005S. The ASD:1001 will also work.

>To: info Subject: Beta 8cx Is the Apt-50 the only tweeter recommended to work with the 8cx coaxial, or can we use psd:2002 or nsd:2005 tweeters with it?

...and the PSD:2002S as a frequency response down to 1.2 kHz for far less than 100 bucks.

Does anybody have experience with speakers crossed around this region? With the PSD:2002?

 

[ZilchLab]

http://www.pispeakers.com/Products.html

You're not going to get it to play that low worth a whit on the little horn in the coaxes, tho....

 

[kstlfido]

Tweets for 10cx

I had a good chat with Zilch; once the 10CX's arrive and I make a new front baffle, we will try different compression tweets and see how they do. A very generous offer on Zilch's part to help- thanks Zilch!!

I heard back from a guy on an old thread trying to use the 10CX with the psd2002, he didn't like it and went different direction- I do not know what.

Alas, the 'proof is in the pudding'; it will be interesting to see what different drivers do.

 

[bluewhale]

Re: Tweets for 10cx

I had a good chat with Zilch; once the 10CX's arrive and I make a new front baffle, we will try different compression tweets and see how they do. A very generous offer on Zilch's part to help- thanks Zilch!!

I heard back from a guy on an old thread trying to use the 10CX with the psd2002, he didn't like it and went different direction- I do not know what.

Alas, the 'proof is in the pudding'; it will be interesting to see what different drivers do.

I agree with everything, and thanks Zilch! The proof will be in the pudding.

In the meantime, Eminence will get back to me with their advice on the PSD2002 coaxial combo. Their frequency response graphs for the PSD are taken without a horn. A search on Google Groups for 'PSD2002 coaxial' gave me some hits suggesting that using it in a coax is not a disaster (see below). The Pi documentation didn't discuss the differences between PSD2002 coax and horn loading. Still, the above post makes me worry that the frequency response will not be even enough at low frequencies at higher volumes. Any more info on this dynamic FR distortion?

psd 2002 info:
http://www.fractalaudio.com/forum/viewtopic.php?f=13&t=6926
http://srforums.prosoundweb.com/index.php/t/6990/0/
http://audiokarma.org/forums/showthread.php?t=147170
http://www.hardforum.com/showthread.php?t=730506&page=114

 

[bluewhale]

Re: Re: Tweets for 10cx

Their frequency response graphs for the PSD are taken without a horn.

Another poster here reveals Eminence does measure the FR in a horn (although their spec sheet does not mention that). An eminence lab technician measured the PSD in coax configuration and it turned out suitable for PA use but not high end listening.

The post is here: http://www.diyaudio.com/forums/showthread.php?postid=1188609

 

[bluewhale]

Dust could be a major factor in shortening the lifespan of the 10CX and 8CX coaxial speakers. Very fine dust can get into the compression driver and voice coil through the holes in the coaxial dustcap, especially when the wind blows hard. The ports also can cause a vulnerability

The dustcap helps a little, but are there any reasonable suggestions for a sonically transparent dust shield? A guy on the pro-audio boards suggested womens' stockings. I vageuly remember some speakers using decorative grills made of some cloth with custom designs.

 

[Pano]

Staying tuned to hear what you and Zilch come up with. Will be interesting to know what other CD works with the 10CX.

 

[kstlfido]

Dust-

I'm not worrying too much about the dust. Sure, some will get in there. But (I hope) there is no airflow through the compression driver to the inside of the box, so little dust will make it there. Come September, I will know for sure!

The 10CX drivers just arrived. Now I need to make a front baffle for one to be able to do the testing with Zilch. As I'm swamped with work again this will need to wait a week or so.

I'll keep you all posted!

-Kent

 

[ZilchLab]

Device Under Test

With Eminence Beta 10CX:

 

[bluewhale]

Used the Selenium compression driver mentioned in this thread with the Beta 8cx. It's crossed at 2.5 kHz using a Dayton audio crossover. An l-pad is turned down almost all the way. I used an iphone RTA to adjust it.

20-quart commercial grade stock pot with 1/4 inch aluminum walls. Crossbraced with a dowel held in place with plumbers epoxy. Walls sprayed with damping spray. Stuffed with some fiberglass insulation. The back of the pot uses some fiberglass sandwiched between mdf and the aluminum.

Amplifier is the Dayton rebrand of the t-amp.

All in all, it sounds quite nice. It has enough power so that the dynamics are realistic and surprisingly loud for ~10 watts. My first PA speaker, and a rugged one at that!

I think the front will have something decorative made out of cloth to keep the dust out and make it look nice. For storage the stock pod lid can be put on. Feeling sleepy...

 

[ZilchLab]

All drivers were tested using EconoWave HF compensation, a minimalist implementation of Wayne Parham's Pi-Speakers crossover. We were surprised to find that the low-sensitivity variant, designed for use with woofers <95 dB sensitivity provided too much attenuation to balance with the Beta 10CX which is 93.3 dB. Switching to the >95 dB version with ~5 dB more HF output was required. Subject to verification, the coaxial "horn" appears to be less efficient than a typical waveguide.

First up, the Eminence standard APT-50, Red, which would require different HF compensation to play flat. Seeing the two HF notches and generally uneven response, we moved on immediately; more on this below.

 

[ZilchLab]

Next up, the ring radiators, BMS 4540 and JBL 2408H, Green and Grey, respectively, which would also require different HF comp, and likely, a notch filter at the very high frequencies to roll off the spike. Ut, oh, the notches are there with these drivers also. What's up with Beta 10CX?

Theory: the coax horn is axisymmetric, and on-axis "holes" develop where edge diffraction at the sharp transition between the central throat and the cone, which comprises the flare, at a substantially wider angle, generates interference. Yet more to come; we test this theory below.

 

[ZilchLab]

Finally, Selenium D220Ti, Blue, and the diminutive JBL 2414H, Violet. Notch City also, but the HF comp was designed for the former, as is apparent:

 

[ZilchLab]

20° off-axis measurements with D220Ti and 2414H, Cyan and Orange, respectively, substantiate our axisymmetric interference theory. We listened to music with both to determine if it is audible. It is, but acceptable, considering the application.

Is there a clear winner here yet?

 

[infinia]

Winners?

Zilch thanks for lots of data
Man thats plainly ugly (plugly ) as in plugme . I'm sure it looks worse because of 1/6 smoothing, but tis good data here ie doesn't hide the 2 (or 3) high Q notches that becomes a familar pattern. Not sure how to identify the cause of the 2 notches or that they might combine to cause a general suckout in the band between them. My guess as posted before is 2 poor transitions between the HF driver entry adapter at 10.1KHz and the horn to cone at 6.8 KHz . Hmm... I wonder why the horn output is non centric to the cone? (The horn to cone transition may actually cause 2 dips looking closer)
Data looks very bass shy... because not much of a baffle and locating the port in the rear along with mike position?

important>Please post the complete Xover as tested (w/ lpad position if you didnt adjust too much)

I think the woofer is closer to 95dB/2.8 V/m