JVC CS-HX512 5-1/4″ Dual-Cone Review


JVC CS-HX512 Review

This HX512 Dual Cone woofer is meant to be mounted onto car doors. However, in this review, I decided to use my Oriole box out of convenience. The cutout fits the HX512 perfectly. When I played some music, I was astounded by the bass. It was then that I modelled the HX512 and to my surprise, she can be loaded in a bass reflex.

Fig 1 – HX512 in 13 liters Ported Box

Fig 1 is the simulated bass in my 13 liters Oriole bass reflex box tuned to 60Hz. The System Q is rather high at 0.987 but she didn’t sound boomy. What is interesting is the F3 is at 53Hz. At this frequency, I wouldn’t need to add another woofer for the bass.

Fig 2 – HX512 RAW Frequency Response

This is the first time I’m working on a Dual Cone woofer so I was quite anxious when I measured the response. That was when I discovered the beauty of a whizzer cone. It actually works. There’s treble. More than that, no crossover is needed like with a dome tweeter. The response is not flat but the peaks can be easily tamed with a 31-band EQ. 

Fig 3 – HX512 Waterfall

The Waterfall plot shows the peaks at 1.5kHz and 5kHz. These are clearly recorded in the frequency response in Fig 2. What I didn’t expect is there’s not much artifacts in the treble. The HX512 is actually quite clean for such a simple design.

Fig 4 – HX512 Spectrogram

The Spectrogram (Fig 4) shows all the artifacts from 2kHz onwards fully dissipated after 1 msec. Those between 1kHz~2kHz don’t last beyond 2 msec. 

Fig 5 – HX512 Toneburst Energy Storage

The Toneburst plot (Fig 5) recorded some unwanted energy (light blue slices) at about 2kHz and a bunch in the treble from 5kHz onwards. 

Fig 6 – HX512 Distortion

The HX512 Distortion (Fig 6) is impressive. THD (H2-H9) is only 0.601%. 2nd Harmonic is 0.299% while 3rd is slightly higher at 0.329%. What I do find troubling is the peak at about 1.8kHz and 5.5kHz. Perhaps they are the cause of the artifacts in the Waterfall.

Fig 7 – HX512 Step Response

The Step response in Fig 7 is expected from a Full Range. Without a tweeter, the HX512 is naturally Time and Phase Coherent. However, her step response is not as impressive as the Swallow-BC.

Sound of JVC CS-HX512

The HX512 sounded “Rough” during auditioning. I could clearly detect the peak at 1.5kHz. Because of that, she came across as slightly “Shouty”. Not terribly shouty but it’s there. Her treble, on the other hand, did not exhibit any brittleness like in woofers with metal cone breakups. It goes to show the attributes of a simple paper whizzer cone.

Fig 8 – HX512 sweep at 45°, 70° and 90° angles

After listening to the HX512 for a few days, it occurred to me that when used in a car, the driver will be listening to her off-axis. That’s when I decided to sweep her at 45°, 70° and 90° (Fig 8). As I had hoped, the 70° and 90° plots are encouraging.

Fig 9 – HX512 On-Axis and 90°

The Brown plot in Fig 9 is of the HX512 at 90°. This angle a a bit severe but it does flatten the peak at 1.5kHz. The response looks like I applied a Baffle Step Compensation at 350Hz. Come to think of it, instead of using a graphic eq, I may be able to do it with a passive BSC network. And if I want more treble, I can always add a tweeter. This is of course with the HX512 mounted on the front panel facing the listener.

Omni Speakers

Now that I have a better idea of car speakers, I’m having visions of designing Omni speakers. I’m seeing a speaker with this HX512 mounted at the top of the box with her facing the ceiling. For the treble, a small tweeter or a planar mounted on the front panel as high as possible.

If I’m going to work on a Omni design, I’ll have to test out other full ranges because this JVC CS-HX512 seems to be discontinued. Such a shame. I can’t remember how much I paid for it but I’m sure they are very affordable. The sound quality is excellent and the manufacture is exceptional. Can you imagine it is equipped with a Strontium Ferrite magnet.

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.