Increasing Efficiency of Amplifier

[qtipextra 5,000+ posts]

Has anyone done an expiriment adding a capacitor (non electrolytic) with the subs coil, and having a input freq. to have the whole circuit at resonance, and have a high gain at the woofer? I'm curious to know if it would work.

In theory, it would work. You would have to know the Le or you subs coils, as well as the resonce freq, and find a capacitor that would provide the best gain.

Click to read more...

 

[qtipextra 5,000+ posts]

Well I have about a week off. Maybe ill do some experimenting of my own.

But I need to know (and someone already asked) how many Henry's an SE's coil is. (The inductance, in Le for those who don't know).

 

[thch 10+ year member]

Has anyone done an expiriment adding a capacitor (non electrolytic) with the subs coil, and having a input freq. to have the whole circuit at resonance, and have a high gain at the woofer? I'm curious to know if it would work.
In theory, it would work. You would have to know the Le or you subs coils, as well as the resonce freq, and find a capacitor that would provide the best gain.

if placed in series with the coil, there would be an impedance minimum @ resonance, for lows, its probably going to be a large cap. and it may end up acting too much like a highpass.

in parrallel. good luck with that. best not to have a device trying to force a voltage to a device that doesn't want to change voltage. its a good way to trip overcurrent protection or cause an amp to be generally unhappy.

even if it worked, in a sealed box, all sound comes from the cone. to make it louder you'd need to be able to move the cone more. a ported box @ resonance is already a low impedance point, with rises in impedance due to temperature, mechanical losses (reflected back through the speaker), and distributed effects.

However, you might try some of this out. keep in mind that parallel connection of the cap might fry your amp. Le won't matter as Mms and Cms are energy storage devices connected to the moving motor and would appear to be an inductor and capacitor attached to the amp as well (see winISD's help file for the T/S models). Z(Mms) and Z(Cms) will both be more dominant then Z(Le).

IIRC, Le was frequency dependent. in practical woofers it would be between 0.2 and 1 mH per Ohm (@ DC).

maybe you could use this to move the resonance to a more favorable point. it may be worth trying out.

 

[PESteele 10+ year member]

wow im gunna go take a shit and ponder what the hell that means in english. **** i wish i was smart.

 

[thch 10+ year member]

wow im gunna go take a shit and ponder what the hell that means in english. **** i wish i was smart.

basically, the speaker is a transducer. if converts electrical energy into mechanical energy. but it also converts mechanical energy into electrical energy.

so when the amp trys to make the cone wobble back and forth, it is possible for the amp to send energy to the speaker and make the heavy cone start to move. then later in the cycle the energy from the moving cone gets sent back to the amp.

the dual happens with the springiness of air -- it stores energy too. the mass tries to move the coil forward, and the spring tries to move it back.

at resonance, the mass and spring forces are equal -- the heavy cone wants to move forward just as much as the springiness wants to move things back.

(this is the basis for the T/S specs).

Similar things happen in the electrical world. coils try to keep current flowing, and capacitors try to keep voltage constant. coils (inductance) are similar to mass, and caps are similar to a spring. the voice _coil_ clearly has some inductance as it is a coil. the OP is asking if there is any benefit to placing a capacitor in some connection to the speaker, in order to resonante with the coils natural inductance.

In this case, the coil's inductance can store very little energy compared to the mechanical storage capabilities of the mass/spring of the cone (which are attached to an electrical-mechanical converter and thus are in the electrical circuit despite being mechanical elements!)

 

[ngsm13 5,000+ posts]

I'm glad I took 2.5years of EE classes... heh.

But honestly... I wouldn't do it qtip... your thinking doesn't work well with speakers.

nG