a moment of inspiration
I thought long and hard about improving the sound of this Control 5. I was very tempted to discard the original woofer and tweeter and replace them with better ones but that would mean having to cut and drill the front panel. Not that I have an issue with that but I think it’s a bit scary for most people. After pondering for a few days, it dawned on me that the boombox sound may be eliminated if I convert the Bass Reflex to a Closed Box. For a quick test, I cut a piece of cardboard and taped it over the port from the inside. And true enough, the horrible bass disappeared, leaving behind a crystal clear midrange.
Fig 1 – Impedance plot of Control 5 in a Closed Box
Fig 1 is the impedance plot of the Control 5 with the port sealed off. Notice the resonance peak is at about 110Hz. This is typical of a Closed Box impedance. There is a small bump at 35Hz which won’t interfere with the sound. Once I do a proper sealing, it will disappear. It is caused by some air leakage from the hole in the cardboard that the drivers wires pass through.
Reworking the Crossover
Fig 2 – Control 5 Woofer Frequency Response
The Black plot in Fig 2 is the RAW response of the 6-1/2 woofer. In the original crossover, it is a 1st order electrical, only one inductor. I decided on a 12dB instead (Blue plot) because I want to suppress the cone breakup peak further. With a second order filter, the 6kHz peak is now lowered by -10dB. This will go a long way in removing the nastiness in the treble. I can install a LCR to kill off the peak completely but that would mean added cost. If I can’t hear it in the final re-design, then I’ll just leave it.
Fig 3 – Control 5 tweeter Frequency Response
Fig 3 is the response of the tweeter. The Blue plot is the RAW response of the 1″ titanium tweeter. In the original crossover, it is a 12dB/oct at 3,000Hz. I wasn’t too happy with that. 3kHz is a bit too high for my liking. I want to cross lower, at 2,500Hz. To do that, I resorted to an 18dB/oct high pass (Red plot).
Fig 4 – [email protected]/oct | [email protected]/oct | fc=2,500Hz
The Blue plot in Fig 4 is the woofer with the new 12dB/oct Low Pass network whereas the Red plot is of the tweeter with a revised 18dB/oct High Pass.
Upgraded Control 5 Frequency Response
Fig 5 – Control 5 Frequency Response with new crossover
Fig 6 – Original Control 5 Frequency Response
Fig 5 is the new frequency response of the Control 5. Compare it with the original JBL response in Fig 6, the difference is obvious. See how much upper bass and lower mids are boosted in the original JBL response (100Hz~500Hz). This is the cause of the muddy vocals. Further up, the deep 6kHz notch also vanished in the new crossover. The upgraded Control 5 response is now flat from 100Hz~20kHz (+/- 3dB).
Fig 7 – Upgraded Control 5 Step Response
The Step Response (Fig 7) with the new crossover is cleaner and sharper than the original JBL crossover. The transient in the woofer is almost a straight line and it doesn’t exhibit a double peak at the top.
Fig 8 – New Control 5 Waterfall
The waterfall plot (Fig 8) with the new crossover shows less artifacts at 6kHz.
Toneburst Energy Storage
Fig 9 – Upgraded Control 5 Toneburst Energy Storage
Fig 9 is the Toneburst Energy Storage plot with the new crossover. With this measurement, it’s easier to see the new crossover results in less ringing than the original in Fig 10.
Fig 10 – Original Control 5 Toneburst Energy Storage
Raw Drivers Sensitivity
In redesigning the crossover, I came across some unexpected surprises. Usually, the tweeter has a higher sensitivity than a woofer. But this is not the case with the Control 5. In fact, it’s the opposite. The woofer has about +5dB more than the tweeter (Fig 11). Perhaps it’s to cater for the insertion loss in using the SK3 lightbulb in the woofer network. Whatever it is, in it’s raw state, the woofer is louder than the tweeter.
Fig 11 – Woofer and Tweeter sensitivity
To resolve this, I converted the Re of the woofer from 3.5Ω to 6.9Ω. This effectively changes the woofer’s nominal impedance to 8Ω. Sensitivity of the upgraded Control 5 still remains at a high of 91dB.
Fig 12 – Upgraded Control 5 Impedance
Fig 12 is the impedance of the upgraded Control 5. The lowest impedance is 5Ω at 3kHz and 8kHz. The electrical phase doesn’t deviate by more than +/- 30°. Overall, it is a friendly load. It will not stress modern day power amplifiers.
Upgraded Control 5 Sound
I am more than pleased with the sound from the new crossover. It is everything the orginal Control 5 is not. Gone is the muffled, dull sound. Now, the vocals are befitting of a nearfield monitor. Crystal clear. And there’s sparkle too. I can actually hear the titanium tweeter.
An aspect of the Control 5 that is not often highlighted is the high sensitivity. More so with this upgraded crossover. It doesn’t take a lot of power for it to be loud. Great for chip and tube power amplifiers.
Part 1 – Thiele & Small Parameters
Note: Unless otherwise stated, all measurements were made with the mic at 36 ins, tweeter axis. Impulse Window=5ms. No smoothing applied.
February 19, 2020Projects