Post by Shimei on Jul 19, 2019 13:11:47 GMT -5
You're quite right - the impedance of most speakers varies significantly over their operating range (in terms of frequency).
The so-called "nominal impedance" is just a sort of general approximation.
There are, however, a few "guidelines" that are usually followed.
For example, it is lower impedance that requires more current from an amplifier, and so makes it work harder.
Therefore, it's not considered to be important if a certain speaker has an impedance that goes very high at a certain frequency, but having an impedance that dips especially low is considered to be "bad form".
The general guideline is that a "well behaved" speaker shouldn't dip much below about 75% of nominal impedance...
So, for example, a 4 Ohm speaker shouldn't dip much below 2.75 Ohms or perhaps 3 Ohms.
It's also less of an issue if a speaker has an impedance that dips very low at very high frequencies - because there is less energy at those frequencies (so the amplifier isn't being strained as much).
Now... as for impedance and driver excursion.
In theory a driver should have a consistent impedance over its full range of excursion.
(If not, that would suggest that the force driving it will vary as it moves, while reproducing a waveform, which would cause distortion.)
In practice, this is never perfect, and some drivers are designed this way as a way to protect themselves.
(Instead of whacking on the end stops if you overdrive them they just sort of stall instead... which doesn't sound quite as bad and doesn't actually break anything.)
It also relates to driver efficiency.
(As a designer, you're trading low distortion over a long excursion, against better efficiency during short excursions, and a bigger and more expensive magnet.)
I should also point out two flaws in what you said about measuring a speaker with an Ohmmeter.
First, a normal Ohmmeter uses DC current to measure resistance, so it will be measuring DC resistance and not the impedance of the voice coil.
In order to measure impedance you need to use an AC test signal.
Second, there's a lot more going on when you push on a woofer.
Yes, by moving the position of the voice coil relative to the magnetic field, you may alter the impedance (by pushing it to a less linear region of its travel).
(This really shouldn't be the case and would normally be considered to be a design flaw... unless it's being done deliberately as a sort of protection.)
However, the impedance of a speaker is largely determined by its mechanical characteristics... like the moving mass of the cone and the spring force provided by the suspension.
By pressing directly on it, you are restricting its movement, which is going to have a major effect on those other parameters.
It would be the equivalent of measuring the characteristics of an electric motor - while holding the shaft so that it can't spin...
The results will be quite far from those you would see "under normal operation".
(It's also rather dangerous to push on a woofer cone because some speakers can be damaged rather easily that way.)
Awesome and thorough response!
While I agree lemme point out that even a single speaker rated at 4 ohms will have different impedance at various frequencies.
Correct me if wrong! Doesn't speaker impedance change? If I take an ohm meter and measure a woofers impedance and push in or out the woofer at various points of excursion does it not change?
Not meaning to overthink this but I was referring to nominal impedance.
Thank you Keith! You're a treasure of knowledge!
Thank you Keith. And just for a point of clarification I would never "push on a woofer" which is working. I was referring to nominal impedance and as you comprehensively stated the different resistance which can be measured at different positions of the speaker's travel.
Also to clarify AudioHTIT I do understand various passive x-overs utilize caps which can hold a charge. Testing can be difficult which is even made obviously observed when simply testing polarity. Taking an "AA" battery and putting it to a speaker with a passive x-over will work one time because of the caps, then the leads need be reversed in order to accurately test phase and whether a speaker may be out of polarity. In car audio this is one of the things installers do. In order to make sure all speakers are in phase pushing out or in at the same time they apply a small voltage. If a speaker especially a woofer is out of phase it becomes quite apparent even by "ear". This is not to suggest that environment etc doesn't have a play in phase shifting but as a basic guideline in wiring up a correct installation.
And yes AudioHTIT I am assuming that nominal impedance of a bi amp connector on the back of the speaker cabinet which is vertically bridged is using 2 8ohm (nominal) channels to come to 4ohms. I could be wrong and acknowledge that if x-overs are taken into account that that might not be correct. And also there's a chance that each channel is 2 ohm but less probable at the bi amp connector in the back as it would not be vertical or parallel at the terminals unless as you suggested the frequencies are taken into account by various x-overs.
To note: I was really intrigued by how testing phase of a speakers some installers use a 9v battery. Can't imagine how many watts that's equal to! Just guessing in the neighborhood of 400watts.
I was never active on this forum before but I'm enjoying the conversation. Can't wait to test various set up combinations and configurations!
I really love it that you fellas are being analytical etc. I plan to review my final setup and think that your input will help make any review more accurate!