Which components in a speaker give it its impedence? Or, what is actually done to a speaker to increase or decrease its impedence? Just curious.

I am by no means an expert on electronics, but I would have to say that the thing that has the biggest effect on a speaker's impedance is the length and size of the voice coil.

There's a good chance I'm wrong about all of this, but here it goes: The resistive part of a speaker's impedance comes from the resistance of the metal used in the windings, just like any old piece of metal. Specifically R= (length * resistivity) / area. Resistivity is an intrinsic characteristic of the material used. So if the length of wire used to make the coil is increased, so will resistance. If a thicker wire is used (i.e. lower guage) then the resistance will decrease. Since the wire is wrapped into a coil it acts as an inductor, adding an inductive element to the impedance. There's also a capacitive element, but I'm not sure where that comes from. Maybe from the voltage created between each turn of the coil? The inductance and capacitance combine to give you the reactance, and the impedance Z is given by Z = R + jX. j = square root of -1. The reactance is dependent on frequency, which is why the value for speaker impedance that we usually deal with is called "nominal" impedance. In actuality the impedance varies greatly with frequency.

Hey Phat, from reading your post, I've managed to confuse myself. The words impedance and resistance are often used interchangably, but I'm starting to think resistance and impedance are 2 different things??? Correct me if I'm wrong but resistance is only one element of impedance? Impedance is the combined effects of capacitance, inductance, frequency and resistance? How am I doing? Am i on the right track?

Resistance is a DC (Direct Current) function. Take a 1 Ohm resistor, put it across a 1 Volt battery (for instance) and 1 Amp will flow........that assumes the battery will maintain it's 1 V with such a high load applied. The evergreen mathematical formula, Ohms Law, applies: V=IxR. Impedance is more complicated because it involves AC. Colis and capacitors behave differently with the frequency of the applied AC. So Ohms law can't be directly applied until the frequency element has been taken into account. That involves some maths...... You can work out the resistance presented by the capacitor or coil at any frequency. For the capacitor there is no effective DC resistance so that doesn't count. For a coil there is so that has to be taken into account to find the total effective resistance at a given frequency. More maths.... A speaker is a coil so has inductance and resistance: there will be capacitance, too, but not that much. Also, the speaker is an electro-mechanical system and the way it operates modifies the electrical characteristics whilst it's operating. [Bruce Gavin, Joris, etc will give you all the finer points, here]. Anyway, the end result is an impedance curve that looks something like the peaks and troughs of a mountain range. To complicate maters further, on the general subject of impedance, amps are quoted with a certain input impedance. Really, it should be more correctly viewed as input resistance because the circuit values are chosen to give a constant input resistance over the normally used range of frequencies. Does that make sense or have I confused everyone still further? John

hi nothing can be done to a speaker to increase /decrease its impedance thats just how its made...ie 16ohm 8ohm 4ohm and others...at least this is what iv ran into building cabs and such for cars and home use..if you bridge to low of impedance (ohm) you can fry your amp...well i bet somthing could be done a long speaker wire could affect it esspessially wound into a coil..that would make your impedance change...get a ohmmeter and test!

More specifically, I meant when a speaker manufacturer (say, Emminence) offers the exact same model speaker in different impedences, what is actually different between them? I'm not trying to change one myself.

This is more along the lines of what I was wonderin'. Just basic, physical "things" or whatever. Thanks everyone.

that all has to do with the cab your trying to build...see if you lets say want 4 15"s in a cab and your opptions are 16 or 8 ohm.....this all depends on your head your using all 4 of these speakers(8ohm) wired together into one chanell puts a load 2.2ohm load on your head thats just above 2 ohm so in other words youd need a 2 ohm stable amp to support that confifuration.all 4 speakears at 16 ohm would be around 4 ohm a little more good for 4 ohm amp stability doing this config with 4ohm speakers would be impossible(read on had athought)sinse that would be 1.13 ohm you could run both chanels though with 4 ohm speakers 2 per side and that would be a 2 ohm load on each side wich would be fine..if your amp is 2 ohm stable agin..peaveys are hehe but if you blew a speaker and are trying to replace go to radio shack get a ohmmeter 7 bux and test it and get the same ohm as it was...OMG thought there is a way to wire to not decress ohms by taking the possitive of one speaker hooking it to the neg of the other then running the other poss to output on amp and same with other neg so you could make a 4 15 cab with 4 ohm speakers wiring both sets of speakers this waybecause then it would be like using 2 4 hom speakers deffinettly test your ohms with the meter

The higher impedance uses smaller gauge wire, hence more turns and higher resistance = higher impedance. The lower impedance model uses bigger wire, hence less turns and lower resistance = lower impedance. They both work out practically the same because the magnetic force is proportional to number of turns X current. With less turns you get lower impedance and more current, but you've got less turns, and with more turns you get higher impedance and therefore less current, but more turns offsets this. So look at it like this: Big number of turns x small current = small number of turns X big current. Same difference. Chris

Naw man, way off. 8//8=4 16//16 = 8 4//4 = 2 8//8//8//8=2 Raw formula for parallel resistances (or simple impedances) = 1/R1 + 1/R2 + 1/R3 + 1/R4 ...... + 1/Rn = 1/Rtotal example: 1/8 + 1/8 = 2/8 = 1/4, so 4 is Rtotal 1/16 +1/16 + 1/16 + 1/16 = 4/16 = 1/4, so 4 is Rtotal. Learn it. Live it. Love it. Chris

You are correct. When speakers are wired in series the equivalent impedance is the sum of the individual impedances of the speakers.

OK. So I've got a 2 Eminence Kappa 15's. One is a 8 ohm model and the other is a 4 ohm model. The 4 ohm model should draw twice the current as the 8 ohm right? If both were given the exact same amount of power (say 10 watts) would they produce the same db (be of the same efficiency) or would the 8 ohm be louder due to the fact that there are more windings?

Same output dB with same input power. I think the efficiency is the same, except for the slightly lower efficiency due to I^2R losses at the lower impedance, although it is probablly negligible. The 8 ohm speaker would need 8.9V RMS to get 10 Watts and the 4 ohm speaker would need 6.32V RMS to get 10 Watts. Using ideal numbers, 8.9V across 8 ohms gives 1.11Amps, and 6.32V across the 4 ohm speaker gives 1.58Amps. Of course the relationship between all this stuff is our good friend 1.414, otherwise known as the square root of (2). You know, I may have been off when I said field strength is proportional to the number of turns. It may be the number of turns squared! That seems like it would work out better since the current is cut by a factor of 1.414. It's been a few years since I got my hard earned "D" in Emag II. Suffice it to say that with identical model speakers of differing impedances, it's like half-dozen of one, and six of the other. Same output for same input. Chris

Well i decided to pull my notes out and if I'm reading my handwriting correctly Ampere's law implies Hl = NI. Assuming I and l stay constant H (magnetic field intensity) is directly proportional to N (number of turns). But can we assume I (current) and l (length) are constant? Ahh I'm too lazy to think. And it's exam week.