Osprey-12 (Eminence 12CX with Selenium D220Ti)

Osprey-12
60 Liters Bass Reflex

The Osprey-12 is the long awaited 12″ coaxial using Eminence Beta 12CX. As in the earlier Osprey-BR, I used the same compression driver, the Selenium D220Ti. This CD is modestly priced yet exhibits exceptional performance. For home use, I can cross at a low of 1.5kHz without stressing it.

Fig 1 – Eminence 12CX with Selenium D220Ti RAW Response

The plot in Fig 1 shows the relative loudness of the 12CX woofer and the Selenium D220Ti. Both measurements are the RAW responses, that is without any crossovers. 

Fig 2 – Selenium D220Ti in 12CX

As in all horns, the difficulty lies in the control of the response. Get it wrong and the speaker falls apart. The Black plot (Fig 2) is the RAW response of the Selenium D220Ti through the 12CX short horn. Fortunately, it is sloping downwards linearly as the frequency increases. If not, I will have to find another CD.

The Red plot is with my High Pass network. The response now is flat at 95dB. This would make it easier to integrate with the woofer.

Fig 3 – Blue plot=12CX with Low Pass • Red plot=D220Ti with High Pass

Fig 3 shows the 12CX woofer and the D220Ti crossing at 1.5kHz. This is about the practical limit for the Selenium D220Ti. Actually, it’s not the D220Ti. She can go lower. It’s the horn that’s cutting off at 1.5kHz.

Fig 4 – Osprey-12 Frequency Response

Fig 4 is the final response of the Osprey-12. She is wonderfully flat from 200Hz to 15kHz. There are some notches along the way but they are to be expected from such a short horn. What is most important is these notches are not noticeable during auditioning. Nonetheless, let’s take a closer look at them.

I’ll start with the first two, the one at 2kHz and the other at 2.5kHz. If you are to take a look at the Raw responses in Fig 1, you’ll find neither the woofer or the compression drivers has these notches. They only appear when the two drivers are summed. Why is that?

These two notches are actually caused by the 2kHz and 2.5khz peaks that are seen in the woofer roll-off (Fig 3). If you don’t want the notches, those peaks must be removed. For a passive crossover, I rather leave them alone because the additional components won’t improve the sound quality.

Farther up at 6kHz and 12kHz, those are cancellations from the horn. There’s nothing that can be done to remove them short of changing the horn or the compression driver. This is just one of those things that one must live with when using a short horn.

Fig 5 – Osprey-12 Excess Phase

Fig 5 is the Excess Phase of the Osprey-12. Here, we see the phase rotation at exactly where the notches are. Some users will be tempted to fill in the notches by applying eq electronically but that is not advisable. It’ll do more harm than good.

Fig 6 – Osprey-12 Spectrogram

The Osprey-12 Spectrogram (Fig 6) reveals the behavior of the frequency response with respect to time. What stands out most is the delayed hot spot at about 2.2kHz. It is not harmful because it is almost gone by 4 msec. From that on to 20kHz, all excess energy are dissipated by 2 msec. Unless your ears are like my microphone, you won’t hear the ringing in the horn.

Fig 7 – Osprey-12 Toneburst Energy Storage

The Toneburst Energy Storage (Fig 7) shows what happens after the fundamental. With reference to the hot spot at 2.2kHz, the excess energy (light blue slices) only came about at -20dB below the fundamental. This delay indicates that the hot spot is from the two peaks in the woofer roll-off.

However at 6kHz, excess energy is very close to the fundamental. This explains why the spectrogram (Fig 6) recorded a burst but the frequency response (Fig 4) is a notch. It cancels then diffracts. And this notch is repeated one octave up at 12kHz (Fig 4). We can safely deduce that these notches are the fundamentals of the horn resonance. Basically, it’s ringing. But fear not, you won’t hear it. They last for only 2 msec.

Fig 8 – Osprey-12 Waterfall

Fig 8 is the Waterfall plot of the Osprey-12. The hot spot at 2.2kHz is clearly visible as are the artifacts in the higher frequencies. Note that these artifacts are largely gone by 2 msec.

Fig 9 – Osprey-12 Harmonic Distortion

The Osprey-12 distortion (Fig 9) is quite normal. Generally, the 2nd and 3rd distortions are about -55dB below the fundamental. What I didn’t expect is the rising distortion from 4kHz onwards. They are no cause for concern as they are 2nd harmonics (Red plot). The 3rd harmonics (Violet plot) are impressively low, about -65dB below the fundamental.

Fig 10 – Osprey-12 Step Response

The Osprey-12 Step Response (Fig 10) is not as pretty as the Condor-III. There are 3 peaks in the D220Ti as she reaches the top. This step is very similar to the 10CX that is used in the Osprey-BR. During auditioning, I couldn’t detect any adverse effects so nothing to worry about.

Fig 11 – Osprey-12 Port

The Violet plot in Fig 11 is the output from the port. I’m relieved that there are no sharp port resonance in the midrange. Presently, the ports are two pieces of standard 2″ PVC tubes used for plumbing. 

Fig 12 – Osprey-12 Impedance

The Osprey-12 is not a difficult load. The Nominal Impedance is 8Ω (Fig 12). Nowhere does the Osprey-12 dip below 6.6Ω. The bass reflex tuning is 45Hz as seen in the saddle at left. There is a slight fluctuation in the electrical phase but it will not stress power amplifiers. 

Summary

It took me a while to muster enough courage to work on this Beta 12CX coaxial. The main problem I had is with the Thiele & Small parameters that I extracted with DATS. They are not the same as Eminence specs. Actually worse. Even with the published specs, I already had difficulty deciding whether I should go for a sealed box or a bass reflex.

After months of procrastination, I finally bit the bullet and mounted the 12CX onto a 60 liters bass reflex. This is much smaller than the optimum size of 100 liters to 150 liters that’s recommended in my box simulation. I don’t really have much of a choice as 60 liters is about the largest box I can lift on my own. Don’t forget, it’s not an empty box. The 12CX and the D220Ti are mounted onto the baffle.

I played some music using the woofer only and was surprised the bass is quite decent. There’s more body and it goes deeper than the 10CX. Having established that, I proceeded to work on the crossover.

One would be tempted to use the Osprey-II or BR crossover but that’s not me. The 12CX and the 10CX are not the same drivers and I was proven right. Strangely, it was the horn response which is different.

When I finished with the crossover, I gave the Osprey-12 a run for a few days. In the end, I would say the Osprey-12 exceeded my expectations. The vocal clarity and treble are outstanding but it is the bass that contributed to her musicality. The Osprey-12 is ideal for hifi and home theater. Tube amp owners will be pleased with her high sensitivity too.

For professional applications, I don’t recommend the Osprey-12 because I seriously doubt the D220Ti is robust enough for the abuse. If I insist on using the D220Ti, I would have to shift the crossover to 2kHz but I cringe at the thought. Swapping the D220Ti with the Eminence PSD2002 is a better option but I have no idea how well they’ll mate. I may explore this in future. Right now, I want to enjoy listening to the Osprey-12. She’s one of those speakers you fall in love with the minute she sings.

Crossover is available on request. Free for DIY. Not for Commercial use.
Unless otherwise stated, all measurements were made in Full Space (4 pi) with the mic at 36 ins, tweeter axis. Impulse Window=5ms. No smoothing applied.