Questions on loudspeaker coil design

[grinninghanis]

Hello Everyone,

Any help would be greatly appreciated.

I need to build a speaker for a project at school. I know the theory behind how speakers work, but I'm having a lot of trouble figuring out how many turns my voice coil has to have to achieve the best results.

I thought that in order to minimize the power lost in the coil resistance and to maximize average power transferred I needed to match the impedance of the speaker with the thevenin resistance of the amplifier. (Which I've been told is usually 8ohms)

My lecturer has hinted at the fact that the number of turns in the coil will also affect the frequency response of the speaker, and that I should choose the frequency range I want the speaker to operate in and work backwards from there.

I've been scouring the internet to find a equation that ties these things together, but have come up empty handed.

So my question is, how do I determine the size and number of turns the voice coil of a speaker should have in order to optimize its performance?

Thanks for your time!

[infinia]

Originally posted by grinninghanis

I thought that in order to minimize the power lost in the coil resistance and to maximize average power transferred I needed to match the impedance of the speaker with the thevenin resistance of the amplifier. (Which I've been told is usually 8ohms)



So my question is, how do I determine the size and number of turns the voice coil of a speaker should have in order to optimize its performance?

IMO the voice coil is about 10% of a speaker design.
The concept of matching Zout and Zin is for higher frequencies and concerns mostly VSWR and has a 3dB power loss penalty built in. For most audio amplifiers using negative feedback, it's Zout (thevinin impedance) is very low which also acts in operation as an electric brake on the speaker motor/cone. In speakers a 3 dB loss is undesirable.
An 8 ohm speaker is a standard used for most average to lower cost home amps, this relates to an average value (complex impedance) over the bandwith of operation. The DC resitance of an average voicecoil designed for 8 ohms is roughly about 6.5 ohms at DC.
a short google query yeilded a couple of links below to start your journey.
http://en.wikipedia.org/wiki/Voice_coil
http://gboers.xs4all.nl/daisy/home/g...voicecoil.html

[Shaun]

To add to some of what infinia said:

Matching the load to RThevenin is only required if you are matching according to the maximum power transfer theory. But this is inefficient, and the amplifier then dissipates (wastes) a lot of power due to I^2R heating. It is not done in audio (that I am aware of). Also, loudspeakers have a non-flat (not constant with frequency) impedance curve; although it might be rated as an 8ohm speaker, its impedance varies with frequency. Having significant series resistence (RTh) will cause the voltage seen at the speaker terminals to change according to the loading as it changes with frequency. Power amplifiers are generally designed to have negligible output impedance; they attempt to be constant voltage sources.

[sreten]

Hi,

TBH you may understand the "principles" of how a loudspeaker
works, but i doubt you know the "theory behind designing them".

The seminal stuff is by Theile and Small - there is a lot in the Internet.

http://www.speakerplans.com/page89.html is a starting point.

The serious stuff e.g. is here :

http://www.readresearch.co.uk/html/articles.htm

Download WinISDpro alpha and play with the driver editor with
autocalculate unknowns turned on to get a feeling of the various
related parameters. Also play with the simulations it allows.
(Note though easy use it is not guaranteed to be accurate.)

/sreten.

[infinia]

The "seminal or serious stuff " by Theile and Small is published long after most of the innovations in speaker technology is discovered. It is useful mostly for low frequency system analysis by way of electical models of the behavior of drivers in bass reflex and acoustic suspension enclosures. It rarely is used as a start for devolopment of transducer designs.
see http://history.sandiego.edu/gen/reco...bell-labs.html
see http://history.sandiego.edu/GEN/reco...udspeaker.html
see http://www.audioheritage.org/html/hi...ef-history.htm

[sreten]

Originally posted by infinia

The "seminal or serious stuff " by Theile and Small is published long
after most of the innovations in speaker technology is discovered.
It is useful mostly for low frequency system analysis by way of
electical models of the behavior of drivers in bass reflex and
acoustic suspension enclosures.
It rarely is used as a start for devolopment of transducer designs.

Hi, how you work that out is beyond me, /sreten.

[infinia]

Originally posted by sreten


Hi, how you work that out is beyond me, /sreten.

Just trying keep the OP on the straight.