After my two epic failures, first with the Morel CAW634, then the Satori MW16P-4, I was utterly distraught. Never before have I felt so dejected.
In my despair, I turned to my Eminence 10CX coaxial. I already have a 45 liters bass reflex box made. All that’s required is for me to design the crossover. And since I have worked on the 10CX before, I am certain I will not face the same disappointment.
My first design with the 10CX was the Osprey-II back in July, 2019. That project was meant for “live” voice reinforcements, meaning speaking into a microphone. Maximum speech intelligibility was my target. Though she proved very popular, I received numerous requests for a bass reflex. It would have been very easy to use a high powered subwoofer as is normal in the pro industry but I suppose home users find that too complicated or expensive. This is where the Osprey-BR comes in.
Fig 1 – Eminence Beta 10CX Bass Reflex Modeling
In my box simulation (Fig 1), the optimum box volume is 40 liters. I decided to over-size her slightly to 45 liters. Predicted -3dB (F3) is at 57Hz. This should satisfy most music playback.
Eminence Beta 10CX in a 45 liters Bass Reflex
Fig 2 – 10CX Frequency Response • Baffle Width=14-1/2″
Fig 2 is the 10CX RAW response in my new 45 liters box. Baffle width is 14-1/2″. No crossover is used for this plot.
The most striking part of the 10CX is the awful cone breakup. There is a series of spikes, the major one being at 2.5kHz. A response like this would be suitable for a guitar amp but that’s not what I’m designing. For HiFi, we must avoid such spikes as much as possible. Prudence dictates I should cross at least one octave below.
The other notable difference from the Osprey-II sealed box is the bass. With this 45 liters box, the bass extends to about 50Hz.
Fig 3 – 10CX RAW and Low Pass
The Blue plot in Fig 3 is the 10CX with a new low pass network. In this Osprey-BR, I went one step further than her predecessor. I installed an EQ network to suppress the 2.5kHz peak even further. Now, it’s -20dB less than before. That should prevent the spike from interfering with the treble when I integrate in the Selenium D220Ti.
Fig 4 – 10CX Low Pass with Selenium D220Ti High Pass
The Red plot in Fig 4 is the D220Ti with another new High Pass network. I am stretching the envelope with this crossover. Normally, I would cross the D220Ti at 1.5kHz but in this instance, it’s 1.1kHz. I can expect a bit of a sag in the summation.
Fig 5 – Osprey-BR Crossover Passband
The Black plot in Fig 5 is the crossover passband. No cancellations are observed in the summation, indicating the two drivers are integrating properly. As expected, there is a slight dip at 1.2kHz but that’s a small price to pay. What is important is the 2.5kHz peak did not cause any issues in the treble.
Fig 6 – Osprey-BR Null
The Violet plot (Fig 6) is when I flipped the D220Ti wires around. It resulted in a beautiful null centered at 1.15kHz. It is not -40dB deep but very close to being time aligned.
Fig 7 – Osprey-BR Frequency Response
Fig 7 is the final frequency response of the Osprey-BR. There are two notches in the treble, the deeper one at 6.5kHz, followed by a lesser one at 12kHz. They are caused by cancellations in the horn used in the 10CX. This is inevitable as it is a short horn. Fortunately, they are notches. Had they ended up as peaks, that would have been terrible.
Fig 8 – Osprey-BR Nearfield Response
The Blue plot in Fig 8 is the Nearfield response of the 10CX. For this measurement, my microphone is about a 1/4″ away from the surface of the cone. At such a close distance, room reflections are not recorded. Basically, the nearfield response is a good approximation of the lower frequencies when measured in an anechoic chamber.
As can be seen, the horrible notch at 150Hz is absent in the nearfield. This is proof that the notch is not a woofer defect but a measurement anomaly. My microphone was actually picking up a floor bounce.
Fig 9 – Osprey-BR Port
The port output (Fig 9) of the Osprey-BR is made with the microphone inserted about 1/4″ into the mouth of the port. The peak is about 45Hz and fairly sharp, surprising for a 3″ diameter port. What is more astonishing is how low the pipe resonance is. It looks like it’s -30dB below the fundamental.
Fig 10 – Osprey-BR Step Response
The Step response (Fig 10) of the Osprey-BR is unlike any other. We can see the woofer is ahead of the D220Ti. As one would expect from a compression driver, the transient is very fast. However, it is not one continuous line. There’s some hesitation at 50% and another at 75%. I doubt it has any impact on the sound quality.
Fig 11 – Osprey-BR Waterfall
Usually, I use a time span of 1.5 msec for the waterfall plot but with the Osprey-BR, I had to use 3 msec because there was more decay all round. The effects of using a short horn is clearly visible (Fig 11).
Fig 12 – Osprey-BR Toneburst Energy Storage
The Toneburst plot (Fig 12) reveals massive excess energy in the treble. This is what horn ringing looks like. But are they as bad as they look.
Fig 13 – Osprey-BR Spectrogram
The Spectrogram (Fig 13) gives a more realistic picture of the cone ringing. Note that all the ringing is fully dissipated by 2 msec. They are there but I seriously doubt any of us can hear down to 2 msec. Even at 1kHz, the decay is gone by 7 msec. So in practical terms, no need to worry. Our ears are not microphones.
Fig 14 – Osprey-BR Frequency Response – 1/6th Octave Smoothed
For readers who are still uncomfortable with the cancellations in the horn, Fig 14 should offer some consolation. This is the Osprey-BR response smoothed to 1/6th of an octave. This smoothing is normally applied in most published measurements, that is if the manufacturers even bother to publish any.
Fig 15 – Osprey-BR Impedance
The Osprey-BR is a very friendly load for power amplifiers (Fig 15). Her Re is only 6.7Ω. Amplifiers rated for 4Ω will have no issues driving the Osprey-BR. The saddle on the left shows the port tuning at 45Hz, exactly the frequency depicted in the port measurement in Fig 9.
Summary of Osprey-BR
The first thing I noticed when I auditioned the Osprey-BR is her effortless delivery. Music simply flows out of the speaker. This is one of the hallmarks of high sensitivity drivers. They sound natural, something that low sensitivity HiFi woofers cannot match.
The midrange in the Osprey-BR is outstanding. Crossing at 1.15kHz is difficult but the rewards are worth it. The upper midrange is now from the compression driver, not the woofer. Voices jump out of the mix. There’s no shrillness in the vocals. Neither is there any annoying sibilance or boxiness.
The treble is superb too. Even with the less than stellar measurements, the sweetness of the Selenium D220Ti compression driver beats most dome tweeters. There’s none of the brightness normally associated with titanium compression drivers. Furthermore, the depth in the sound-stage is superior to dome tweeters.
When it comes to bass. the Osprey-BR rose to the occasion. There’s just about the right amount of punch and dynamics. Overall, the bass is tight and has a strong attack.
On top of all these, the Osprey-BR is blessed with high sensitivity too. Class A and tube amp owners would love this. So if you want better performance and value for your money, forget the Morel CAW634 and the Satori MW16P-4. Go for the Osprey-BR instead.
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.
April 25, 2021Projects