Wood Cones - Listening Tests
the academic tests previously performed between a newly purchased wood
cone speaker of diameter 6.5 inches and a Wharfedale Super 8
CS/AL modified with a white oak veneer cone, copper voice coil and cloth
surround. That page is HERE
The new speaker is purposely unnamed and suffice it to say that it is a visual beauty, too damned nice to hide in a box. It has a 1.5 inch voice coil with an inductance of 1mh and an Xmax of several millimeters. In short, it's a woofer but oddly enough, it's high frequency response is flat within 5dB to about 5khz. That is due in part to it's very lightweight wood cone and a phase plug. In my humble opinion, it should be classified as a bass mid-range. It wasn't inexpensive either. Two were purchased.
The Super 8 CS/AL used was NOS, New Old Stock and all original. The cardboard ring was easily removed as was the cloth (felt) annulus from the basket. The spider is a phenolic one held in place with two screws, making disassembly easy and without damage. It has a 1 inch diameter hand wound voice coil of copper wire with an inductance of about 0.3mh. It's high frequency response extends to a little beyond 7khz, albeit a several dB rise between 3khz and 5khz. This was discovered to be caused by the 60 year old cone which, compared to a new paper cone, is brittle and subject to serious cone breakup. This smoothes very well with a new cone. For more detail, see fig.2 in the aforementioned page.
|One of two Furman electronic crossovers. This one was used to limit the test bandwidth to 400hz-4khz. This decade was chosen due to its having all the necessary musical info to identify any musical instrument and also it's being in the most sensitive band of the human ear. It is also the band used for telephone receivers, although some do extend a little lower and higher. Only one channel was used and only the mid section of that channel. The low and high ends were turned off. The filters are 4th order Butterworth.|
is the actual frequency response of the crossover mid section.
Taking 8dB as a reference 0dB level, it can be seen that the knees (-3dB
points) are at 400hz and 4khz.
The input here from the LMS is 0.445v as is the output from here to the pre-amp. The unit is set for unity gain. The output from the power amp to the speakers during the test was about 2 watts. The SPL was about 95dB. See Fig. 2.
The test setup, such as it is; functional but not fancy. The + leads from each speaker were held in my left hand and the hot lead from the amplifier in my right hand. This allowed a simple but fast way to switch between the two speakers, albeit risky for if the amplifier lead was accidentally dropped, Murphy's Law says it would touch the common leads and pop the internal fuse in the power amp, which is a pain in the butt to access.
The tweeter and floor firing woofer are not in the circuit. That array was previously used to set up a 3-way crossover for another project. Anyone reading this familiar with the EMPIRE Grenadier 9000 of the sixties? The next project.
The panel with the holes is from another previous project using four eights and a ten and a papyrus coned 6 incher; a two way system. The cabinets were assembled with about 80 screws to allow disassembly as I'm virtually out of room. Imagine that. I live alone in a house of 2300 sq.ft and I'm out of room. Needless to say, the Grenadiers will have to designed with disassembly in mind. They will be about 21" in diameter and 39" tall.
The response curves. RED: Super 8; GREEN: The new wood cone speaker
| The first test showed the Super 8 as being considerably louder than the other speaker. The purchased speaker has a
1.5" voice coil with a higher inductance. 1mh than the Super 8,
0.3mh. That other speaker is actually a woofer despite its having a phase plug. It does well to just a little over 5khz. So,
a resistor was inserted in series with the new speaker. After a few tries, a
1.5 ohms resistor did the trick. The new speaker is about 7.5 ohms and the Super 8 is
13 ohms. However, at frequencies above 3khz, its impedance rises to
11.5 ohms at 3500hz. Above 2 khz it's close to 10 ohms and at 4khz, it's 11.8 ohms.
The added resistor pretty well matched the impedances of both speakers. The Adcom is a direct coupled amp, no impedance matching transformer. Anyway, after matching the impedances, the volume levels were very close.
Admittedly, the speakers could have been driven by separate channels but
that introduces two more variables, one from the pre-amp and one from
the power amp, despite their probable insignificance.
As for the difference in sound, there was very little. The Super 8 still seemed to be slightly more prominent at the extreme high end of its passband. This is most likely due to the much lower voice coil inductance of the Super 8. The 6 inch wood cone has a 5dB rise between 500hz and 800hz. This too will have an effect on the similarity. What was found surprising is that the difference between the two is very small; one has to carefully listen. If one wasn't listening attentively, one could switch between the two speakers without detection.
Keep in mind also that the cone in the six inch unit is comprised of two halves; there are two seams. The cone in the Super 8 has but one seam and the shape deviates from that of a perfect cone.
Back to the loudspeaker main page