Disadvantages of Capacitor in series with main driver.

I have just got a pair of Mk1 Wharfedale Diamonds to play around with. Hugely successful cheap speaker from the 1980s. One aspect of the design is a 450uF capacitor in series with the main drive unit (which runs full range) (The speaker is rated at 6ohm)

My understanding is that the reason for this is so that a smaller cabinet can be used, and the cap brings down the resonant peak. I believe KEF used this in some of their designs.

This seems a neat trick, but what are the disadvantages?
 
I think I used to have a pair of those cute little reflex speakers. 4" plastic bass and Audax type "balanced drive" tweeter. Bedroom speaker really. :D

I just ran a sim for a 450uF capacitor. It does next to nothing. The dotted line. :confused:

If it was a 47uF, the effect would be a dramatic loss of bass below 200Hz. Turning it into a centre (voice) speaker for home cinema.
 

Attachments

  • Wharfedale Diamond Mk1 450 Capacitor effect.PNG
    Wharfedale Diamond Mk1 450 Capacitor effect.PNG
    17.1 KB · Views: 1,474
Whatever were Wharfedale thinking of? All I can think is it might have protected the speaker from subsonic rumbles from a vinyl turntable. This is a weakness of reflex loading below what we call the tuning frequency, where the speaker has nothing to restrain it. :confused:

For all that, it WAS an ace little speaker. So the capacitor couldn't have done much harm.
 
Last edited:
If the capacitor puts capacitive reactance of the right amount to cancel the inductive reactance of the driver (which is just below the resonant peak) of the right Q, it can bump up the response there and drop off below there. It also gives a degree of LF (and DC) protection.

Here is the voltage bump given by a 1000uF series cap into the impedance of a pair of paralleled woofers in a sealed cabinet I have here (resonant peak at 91Hz). Without the cap, it's just 4V across the woofers.
attachment.php

Q of this cabinet is pretty high, but I think with lower Q it would work a little better.

in dB:
attachment.php
 

Attachments

  • C boost of woofer.png
    C boost of woofer.png
    14.3 KB · Views: 2,090
  • changeFromCap.png
    changeFromCap.png
    10.5 KB · Views: 2,060
Last edited:
This is the impedance matching the situation for the plots I showed above. You can see how it works -- the impedance just below resonance with the cap (blue trace) is reduced compared to the "without cap" (red trace), which puts more current through the woofer(s) there. Then, a little lower the impedance goes up as the capacitive reactance increases again.

attachment.php

Kind of like a simple passive Linkwitz transform circuit (though lots less versatile).
 

Attachments

  • woofZwithCap.png
    woofZwithCap.png
    21.8 KB · Views: 1,281
Last edited:
I have just got a pair of Mk1 Wharfedale Diamonds to play around with. Hugely successful cheap speaker from the 1980s. One aspect of the design is a 450uF capacitor in series with the main drive unit (which runs full range) (The speaker is rated at 6ohm)

My understanding is that the reason for this is so that a smaller cabinet can be used, and the cap brings down the resonant peak. I believe KEF used this in some of their designs.

This seems a neat trick, but what are the disadvantages?
A main disadvantage is added non linearity. The cap works with the woofer impedance near resonance to create a 3rd order high pass with lower f3 than without the cap. Unfortunately the woofer impedance around resonance varies with drive level so the combined response varies with drive level. This by definition is increased non linearity. The distortion added will depend on the driver. Ive never seen this quantified but it would be interesting
 
I have seen series caps used in Kef, Wharfdales and Infinity's bass crossover ccts and in a certain canton speaker range (reflex) it was incorporated with a large value parallel inductor to make a high pass bass cct and the implementation was even given a trade marked name.
cheers Arthur.

just found the cct for the canton, the cap was 820uf and the inductor was 23mH
 
Another way of extending/boosting bass output from a sealed enclosure using passive components is to use a high value cap and high value inductor in series with the bass driver (crossover< 200~300Hz, as used in 3 or 4 ways ) though the resultant impedance drops below nominal value of bass driver, Infinity used this type of circuit to increase the bottom end extension of some of their speaker designs.
 
It certainly seems to work. I tried bypassing the large capacitor, (since I thought it was there to protect the driver, and I only play at fairly low volumes.) I was expecting an improvement, but all I got was an appreciable loss of bass, and it all sounded "smaller".

This technique seems to have been used in the past at least by KEF and I wondered why it is not more generally used.
 
Not generally used because it is a waste of time and money, I would think.

The sim does not lie. The dotted line is the effect of an ideal capacitor:

694756d1533160168-disadvantages-capacitor-series-main-driver-wharfedale-diamond-mk1-450-capacitor-effect-png


Hardly worth doing, I'd say. Any ESR (resistance) in the capacitor would reduce level too right across the basses output range. Any perceived boosting effects on bass just show how difficult subjective evaluations by memory are.

Wharfedale did this in the 8" plus tweeter Wharfedale Laser 90B too:

680152d1526122842-wharfedale-shelton-xp2-minor-classic-imo-wharfedale-laser-90b-2-jpg


But 90% of the time they didn't. And nobody here does, AFAIK. So I guess someone at Wharfedale was infatuated with the idea. It's possibly worth mentioning that there must be a reduction in the effective damping factor from the amplifier, but the general notion is that good damping factor is essential with reflex.
 
Last edited: