Spice Model for Loudspeaker

Moving a mirror ;-)

Thanks!

...Jim Thompson

I actually designed boom boxes in my youth (mid-40s :), bought several ZXs, and did the whole middle-age-crazy bit, grew a beard (which I still have, but it's salt-and-pepper now).

Thanks for the link!

...Jim Thompson

I read in sci.electronics.design that Jim Thompson wrote (in ) about 'Spice Model for Loudspeaker', on Thu, 23 Sep 2004:

I suppose there is no point in asking why. Go to:

My esteemed colleague Wolfgang Klippel will either already have a solution for you or will, I feel sure, help you towards one. What he doesn't know about loudspeakers isn't known by anyone.

Hi Jim,

And here I thought you wanted to install a huge boom box in your car, tint the glass, put on mag wheels....

This lab might be able to help, at least when it comes to the calculations:

Regards, Joerg

There *will* be a control loop. Model is needed to evaluate best approach.

...Jim Thompson

-- | James E.Thompson, P.E. | mens | | Analog Innovations, Inc. | et | | Analog/Mixed-Signal ASIC's and Discrete Systems | manus | | Phoenix, Arizona Voice:(480)460-2350 | | | E-mail Address at Website Fax:(480)460-2142 | Brass Rat | |

Interesting. What for? If you need precision, you'd be better off designing a control loop to set the precision exactly. Of course, a spice model would help the design... :-)

-Ed

On Friday 24 September 2004 07:41 am, Jim Thompson did deign to grace us with the following:

I wonder how big this mirror is intended to be? The idea of an electrostatic speaker springs to my mind, or maybe capacitive feedback from the voice coil/mirror arrangement - it will need some kind of feedback, right?

Thanks, Rich

Hi Jim,

Just one other idea: There used to be "cone-less" speakers. Basically these were just the coils with a counter mass. This was screwed to whatever surface needed to be insonicated. I don't know whether these are still marketed but I remember they worked ok and were designed for use with full size stereos. I think some vendors called them invisible speakers.

That solution would at least take part of the components out of the equation.

Regards, Joerg

I read in sci.electronics.design that Joerg wrote (in ) about 'Spice Model for Loudspeaker', on Fri, 24 Sep 2004:

They have more or less been reinvented, with a bit more science applied, and embodied in 'flat-panel loudspeakers'.

Jim,

Could you let me know how you get on with this, if that's OK. I've just come across something where a bass loudspeaker may be the best kind (ie cheap, avaliable, right kind of frequencies avaliable) of actuator.

One problem is that I can't figure out the best way to atach the actuee (if that is a real word) to the speaker. Gluing something to the center of the cone may work, but one has to be careful since you mustn't cause the whole thing to move off axis (lots of friction if that happens). Also, loudspeakers are designed for moving air, which suggests that the whole assembly is less efficient that it could be since it's moving large amounts of air around. I can not, however think of a good way around this since the cone also performs the job of keeping the coil in the correct place, rather well.

One could alternatively place both sides in an air tight box to improve the springiness that returns the speaker to the original position.

Anyway, one could presumable measure the displacement of the cone by looking at the inductance of the coil, since the further in it is, the more inductance it will have. However, I can't think of a good, high precision way of measuring the displacement required to calibrate that.

Any thoughts so far?

-Ed

I read in sci.electronics.design that E. Rosten wrote (in ) about 'Spice Model for Loudspeaker', on Mon,

You can glue a tube to the rim of the voice-coil former (under the dust cap). You keep it centred by putting a support across the open end of the bucket, having a carefully-placed hole through which the tube passes.

You can (carefully!) cut holes in it. But if you want more than a few of these, a loudspeaker manufacturer (no, not one in China!) will make you cone-less units with a heavy-duty rear suspension or a double suspension. For $$, of course.

If you are using the driver unbaffled, the effect of the cone is minimal at frequencies below that at which the circumference of the driver rim is less than half a wavelength.

You could. The boxes might have to be quite big, and they can't be quite air-tight because they will 'blow' the cone as the air pressure changes.

The inductance changes in a complex way with position. The magnet structure is close to saturation, and there is sometimes a conducting element in the pole-pieces to create electrical damping.

I posted a link to Wolfgang Klippel's site. One of the topics he has researched in depth is the measurement of cone displacement - to high precision. It's *difficult* - he uses a laser.