About a series capacitor with closed-box loudspeaker

[presscot]

I’ve found an introduction of capacitor tuning for closed box speakers technique published by SB Acoustics. It suggests to build a closed enclosure with Qtc = 1.0 and apply the series capacitor directly to the terminal of the speaker driver. I understand that, by doing this, the principle might be to lowering the Qtc to 0.707.

Firstly, is my understanding correct?

Then, if my understanding is correct, secondly, will it work with the system whose Qtc is less than 1.0? I mean if the cabinet’s Qtc is 0.5, would the final Qtc after applying the series capacitor be lowered than 0.5?

Finally, could anybody explain the principle or the operation of this series capacitor technique? I still have some basic knowledge about Physics and Electronics.

 

[Galu]

I believe you are referring to a 3rd order sealed system:

EDIT: You can see the reduction in the system Q that you mentioned.

 

[Freedom666]

it works, putting driver into smaller volume and a fitting C linearizes and filters at the same time

 

[johnnyx]

The Wharfedale Diamond, the original one, had a series capacitor.

I thought they were broken, because they measured open-circuit.

They were ported, though.

 

[Galu]

could anybody explain the principle or the operation of this series capacitor technique? I still have some basic knowledge about Physics and Electronics.

How comfortable are you with complex numbers and the j notation?

The physics is in the "Subwoofer tuning with capacitor in series" section of this article:

https://www.lautsprechershop.de/hifi/aka_tief_c_en.htm

The outcome is that the capacitor increases the voltage applied to the driver over a range of audible frequencies below its frequency of resonance.

 

[Freedom666]

also ported works, there was an article in a german diy magazine on this, Hobby Hifi

you see it bass reflex with high pass filter


http://www.hobbyhifi.de/Archiv/PDF-Downloads/hh111_grundlagen.pdf

 

[Galu]

The Wharfedale Diamond, the original one, had a series capacitor. They were ported, though.

That's correct. In the original, small, rear vented Diamond the large series capacitor resulted in 5th order bass loading.

 

[Salas]

Some classic KEF speakers used closed box & big series capacitance third order alignment

 

[Galu]

Regarding Presscot's 3rd order sealed system:

• The woofer must have a low resonance frequency fs (below 30 Hz) and a high Qts (above 0.40).
• The cabinet volume for the woofer should be calculated based on a sealed box of Qtc around 1.0.

Calculating a rough value of the series capacitor involves an empirically determined constant K.

C = (K x Qts) / (Re x fs)

The constant K depends on the voice coil inductance and, for the 'standard type of woofer', a value of K = 265 is suggested.

If nothing else, the above formula provides a good starting point for experimentation!

There's information about which K values to use and more here: https://www.humblehomemadehifi.com/download/Humble Homemade Hifi_Black Box.pdf

P.S. Neville Thiele has a paper on adding a series capacitor to a closed-box loudspeaker if you are willing to fork out $33 to buy it:

https://aes2.org/publications/elibrary-page/?id=15513

 

[Jeffh01]

More info from Rod Elliot here : https://sound-au.com/articles/cscaling.htm

 

[cracked case]

I guess this would be something easy to experiment with if you used a single rail, capacitor coupled amp.
The above explanations are perhaps better than the previous one I'd heard, that the increasing impedance caused by the cap reduced the damping effect of the amp - a but more complicated to understand though.

 

[Galu]

I guess this would be something easy to experiment with if you used a single rail, capacitor coupled amp.

That could be tried, but perhaps it's not too practical as the typical amplifier output capacitor is of a higher value than that required to provide the desired 3rd order function.

For example, the output capacitor in my single rail Rogers amplifier is 1,500 uF, whereas the series capacitor in question is in the order of 500 uF.

EDIT: I suppose if you want to dedicate the amp to one particular speaker then that is OK.

 

[bwaslo]

If you have an impedance curve file (.ZMA) of a sealed woofer it's pretty easy to fool around with in a simulator like XSIM or VituixCAD and see what's going on. Essentially, the reactance of the capacitor cancels out some of the reactance of the woofer, which lowers the impedance in that range. Draws more current there to give a small bass boost.

 

[eriksquires]

Thanks for that, @Galu - I think I've seen YG Acoustics do this in one of their smaller speakers. I have a couple of questions:

1. Is this what a Linkwitz Transform does?
2. Isn't there significant electrical phase difference which can be some what challenging for an amp especially since it happens in the bass?

 

[eriksquires]

@bwaslo I was thinking of suggesting that. The Component Voltage (V) graph is something I often use in examining crossover behavior. I wish the OP would rely on it more.

 

[Galu]

@eriksquires

1. A Linkwitz Transform circuit can similarly lower the system Q as well as extend the response to lower frequencies.
2. Pass! However there is a phase plot in my earlier link: https://www.humblehomemadehifi.com/download/Humble Homemade Hifi_Black Box.pdf

 

[Moondog55]

The downside can be the cost of a very large capacitor bank, depending on the woofer it could get as large as 2000uF, one of my passive subwoofers used 1550uF and I could have gone a couple of hundred more

 

[eriksquires]

@Galu That looks much better than I expected in terms of altering the electrical phase, not bad at all.

 

[Audio>X]

To make a high value series capacitor quite low price & affordable >
start with some 160V 1000uF pairs wired negative to negative and then 'tune' with bipolar electrolytics + a good 10uF 'bypass'.
https://www.aliexpress.com/item/32538632943.html
PS.
You also end-up with a DC protected speaker 🙂

 

[chris661]

Just a passing note that the reason there's gain in some of the bandwidth is because the cap+speaker impedance is lower in that range. No free lunch etc.

Chris