Coast to Coast Sound - Part 2
First of all, I rediscovered the remarkable build quality of those JBL drivers. The 12" woofer (JBL 123A) in particular is a gem. Black crackle finish covers a cast aluminum alloy frame. The motor contains a 6 lb. Alnico magnet of 10,000 gauss with a 3" edge-wound copper ribbon voice coil. Immersing the coil in such a strong magnetic field, results in a driver with outstanding sensitivity, range, and dynamic response. The cone, an early version of coated paper, with its characteristic white color, is unusually well-damped, as we shall see. All around close construction tolerances were common, as exemplified by the tiny gap between the voice coil and magnet. They are used in the L-100 with no low-pass filter at all in a vented 1.6 ft³ (45.3 L) cabinet. The JBL owner’s manual claims the vent tuning is at 27 Hz; however I am unsure about that.
The 5" midrange driver (LE5-2), with a potent 2¾ lb. Alnico magnet of 16,500 gauss, is even more sensitive. Used in the L-100A with a 1st order high pass filter at 1.5 kHz without any low pass filter, I could easily hear prominent upper midrange peaks, and feared that it may not be a very smooth driver.
The 1½" cone tweeter (LE25), although clearly not up to today’s dome tweeter standards, does OK below 10 kHz. It is used in the L-100A with a 1st order high pass filter at 6 kHz. Note that on my L-100As the polyurethane foam surrounding the tweeters has long since decayed.
The cabinet is heavy, solid and strong, with a beautiful walnut veneer that has stood up to time without showing its age. An obvious problem is the layout of the drivers.
Photos of these drivers can be see on a website (
http://www.troelsgravesen.dk/vintage.htm) by Troels Gravesen, a DIY speaker builder in Denmark. He also shows frequency response curves of the woofer and tweeter from L-100As he intends to refurbish. Note that his woofer is smooth out to 6 kHz! Try and find a 12" hifi driver today that can do that! The tweeter looked OK, at least from about 3 to 10 kHz. That left the midrange driver’s performance an unknown. Gravesen’s information suggested that the L-100A drivers were good enough to deserve a better crossover, and inspired me to try.
In a similar effort, Troels Gravensen restored a pair of JBL L-26 Decade 2-way speakers of a similar age as the L-100s,
http://www.troelsgravesen.dk/JBL-L26.htm. He designed a new crossover that seems to perform much better than the original design. If you are interested in restoring any of the older JBLs, his website is worth reading.
The original L-100A crossover is a good example that vintage is not always better. It contains only 1st order high-pass filters at 1.5 kHz for the midrange and 6 kHz for the tweeter. The woofer had no filter at all, and the midrange lacked any low-pass filter. It is certainly simple, but as we’ll see, it’s far too simple. Also note that old JBL drivers were made with the opposite absolute polarity compared to most other manufacturers of today. When a positive voltage is applied to the positive terminal of a JBL driver, the cone moves in-wards not outwards.
Figure 2 JBL L-100 Original Crossover.jpeg
At this point, I asked Dennis Murphy, if he was interested in this vintage make over project. His eager response generated the rest of the details below. I am most grateful for his expert help and enthusiastic guidance.
His first look at the speakers provided a frequency response curve that graphically shows just what “The West Coast Sound” means. Several features are prominent:
- A big ugly peak from 6 to 7 kHz. Perhaps caused by unfiltered breakup of the midrange driver, this peak certainly would have to be tamed.
- A general rising response as frequency increases, especially above 2 kHz. This probably can be easily corrected.
- Destructive cancellations were seen resulting in deep troughs at 3.3 kHz and above 9 kHz, producing a prominent comb filter effect. This is probably due to the unfortunate placement of the midrange driver relative to the tweeter and woofer on the front baffle.
The prominent rise and fall of frequency response below 200 Hz is the product of room reflections and their resulting standing waves and cancellations. It is not directly due to the speaker.
Fig 3 Old on-axis response.gif