Raising Qts with partial current drive?

[Circlotron]

I've spent quite a lot of time lately inside the fantasy world of WinISD trying all sorts of drivers and boxes. It seems that really good drivers often have a Qts of 0.3 and less, and if you put these in a vented box they start rolling off very early because the magnet has such a tight grip on the voice coil it doesn't allow the cone to increase it's displacement very much as the frequency goes down. It makes it follow the [not increasing] amplitude of the drive signal as frequency goes down.

One very naughty way to "fix" this would be to put a suitable resistor in series with the speaker. This would reduce the magnet's grip and therefore the damping would be lowered until you get the response you want. But... you're gonna throw away a whole heap of amplifier power too!

Seems to me that a better way would be to include a =small= but adjustable amount of negative *current* feedback in your amp so you can set the amplifier output impedance to something like several ohms. I'm not talking about a pure current drive where the amp loses all control of the speaker, but only just enough to knock the response into shape. Maybe even have it vary the impedance with frequency? Mmmmm... maybe not; dunno.

Seems as if a Qts of about 0.5 works well with a vented box; down at 0.25 a bandpass box works well.

[wintermute]

Hi Circlotron,

I found when modeling my vifa 10" woofers that putting in a series resistance of between .5 and .7 ohms made a big difference. At first I thought that I would have to get a high power low value resistor and put it in, but then I started to suspect that the allowance for series resistance in the modeling programs is there to simulate the effect of the resistance of the speaker leads and crossover components.

Maybe the speaker designers take this into account when designing the speakers. I know that the data sheet for my vifa woofer gives recommended cabinet volumes and tuning freq, and actually mentions .7 ohm series resistance.

Regards,

Tony.

[phase_accurate]

Hi Graham

You can indeed modify your amp's output impedance by sensing the current through the load and feeding it back. I will look for the maths.
Depending on phase you will end up with either positive or negative output resistance. You can even make it complex (as you can see within the paper I once sent you).

Regards

Charles

[mrfeedback]

There was a series of Yamaha amplifiers that had an 'Ro' control that went - /0/+ to several ohms.
I never had a proper to listen to one however.

Eric.

[Circlotron]

Originally posted by phase_accurate
Hi Graham

You can indeed modify your amp's output impedance by sensing the current through the load and feeding it back. I will look for the maths.
Depending on phase you will end up with either positive or negative output resistance. You can even make it complex (as you can see within the paper I once sent you).

Regards

Charles

Hi Charles and everybody. I knew that you can fiddle the output impedance with NCF, but was actually wondering if it would make a suitably "lossless" adjustable resistance.

As for negative output resistance, that should *lower* the Qts. It would be interesting to see what you could achieve doing this to a cheap junk speaker with a pea-sized magnet.

[sreten]

http://sound.westhost.com/z-effects.htm

try this sreten.

[Fuling]

Could this be something for open baffle speakers, maybe?