Testing Woofer Impedence

[fini]

I spent a good part of yesterday yanking components out of some Barzilay cabinets (and then smashing to bits said cabinets), trying to carve out some room in the garage. None of the boxes suited me, but the innards are most certainly salvageable, probably headed for eBay.

Anyway, I pulled out my DMM, to do a quick check of the woofers. This meter is really the only piece of test equipment I have. So, these Altec 415A Biflex woofers have a noted impedence of 8 ohms. When I tested them, it was only 4.8 ohms (give or take). Has this test shown me anything useful, or does the speaker need to be measured in another way? My guess is that 8 ohms is an "average" measurement, while the speaker is "working."

These woofers were paired with Altec 2000B tweeters. Nothing special, really. The other set of Barzilays had some pretty nice stuff in them, though. JBL LE-14's, and JBL LE175's (with 1217-1290 horns). Talk about beefy components!!

[scriven]

You are measuring DC resistance not

impedance. The impedance will vary with

signal frequency and can't be measured with a DMM. 4.8 DCR sounds a little low for an 8 ohm woofer but not totally

out of line. Your right about the rated

impedance being a "nominal" figure.

[DrWho]

I dunno, a 3-5ohm DC resistance sounds about right for a woofer...at least all the woofers I've been modelling as of lately are in that range (some do go up to about 6ohms). If nothing else, this measurement indicates that there are no shorts in the voice coil. You could always set your meter to AC and play test tones through the speaker while you measure. You generally get a huge impedance peak around the Fs of the speaker (free air resonate frequency)...or at the tuning point in a vented enclosure. Then it will typically dip to its lowest impedance about on octave above this point and then gradually rise as frequency increases.

[scriven]

I will certainly yield to your experience. Thanks for the correction. It seemed to me that most of them that I have

measured were closer to 6 ohm but I don't go around measuring the DCR on

woofers very often.

[BEC]

Some examples:

A 4 ohm K-33E measures about 3.2 ohms DCR.

An 8 ohm K-22 measures about 6.8 ohms DCR.

A 16 ohm K-55V measures about 10 to 11 ohms DCR.

An 8 ohm K-77 measures about 6 ohms DCR.

Bob Crites

[dragonfyr]

Here is an article that will explain measuring with an impedance meter - essentially everything you need to know up to, but not including, Nyquist plots.

MeasuringImpedanceX.pdf

[Al Klappenberger]

Fini,

I did a lot of going around in curcles about woofer impedance and found

there is a huge amount of bad information going around! Most everybody

forgets about the inductance of the voice coil! Even if you measure the

AC impedance (not just DC resistance) it is expressed in simple Ohoms,

like 6 or 8. The fact is, a woofer looks like a resistance in sereis

with an inductance. A K33 woofer is 6 Ohms (roughly) in sereis with a 1

mHy inductor! I have found the only way to measure it properly is with

a noise generator, power amp to kick up the level above the energy it

generats itslef as a microphone from ambient noise, a resistor bridge

and a FFT analyser. This is not stuff most audiophiles have

sitting around. The plots come out as a plot of complex impedance ( R

+-jX or Zo and Phase) across the woofer frequency range. You then need

to calculate the incuctance from the complex impedance.

Al K.

[fini]

Thanks for the info! I decided to refer to a book ("Electronics

Demystified") to try and understand impedence, and started reading

about "imaginary numbers." I didn't get too far, as I started

having that dang Atlanta Rhythm Section song, Imaginary Lover, going

through my head.

[Al Klappenberger]

Fini,

You are on the right track. Don't get overwhelmed by the imaginary

numner thing. All that is is the effect of the inductance. If you

remember the relationship of the sides of a right triangle (h^2 = a^2 +

b^2) then you already know it. The resistive componet is "a" and the

reactance of the inductance is "b". That will give you the R component

of the impedance. The phase angle of the impedance is the trigonomery

function just as you would expect. The analogy of complex numbers is

used becasue of the fixed 90 degree relatioship of resistance to

reactance just as the 90 degree side of a right triangle.

Al K.

[Al Klappenberger]

Fini,

Here's an example. This is the plot of the woofer in one of my Belles.

Notice the marker data at the bottom. It says 7,41 Ohms at a phase anle

of 34.3 Deg. The 7.41 is the simple impedance that most people would

round off and say the K33 is 8 Ohms, BUT: if you change that to the a

and b sides of a right triangle (convert polar to rectangular

impedance) you have 6.12 ohms resistive in series with 4.18 Ohms if

inductance (inductive reactance, actually). Since inductive reactance =

2 X Pi X frequency X henrys you can calculate that you have .949 Mhy of

inductance at 700 Hz. That is the actual impedance of the K33 woofer

driver.

Al K.

[WMcD]

I used to do a lot of fooling with my LMS system. It measures impedance magnitude and then calculates phase. This then can be recaluclated to the R and jX.

I believe that what Al K. is showing us is just the impedance of the woofer in free air. The woofer input impedance in a bass horn is different. You see some additional peaks caused by the rather rough throat impedance of the horn, particularly at about 120 Hz.

Also. the overall impedance is elevated by the horn. It is not quite up to 2x DCR as theory predicts.

The big concern, for me, was the value of impedance at 400 or so Hz where we place the crossover. There the phase angle is fairly low, lower than Al's plots.

You have to draw the triangles, as Al says. The point is that the resistance is relatively higher than the driver in free space. Therefore the phase angle, and the effect of the inductance is relatively lower.

The bottom line is that for our horn loaded designs, we can can rely on impedance magnitude for crossover design.

This is somewhat in contrast to what you see about designs of crossovers for direct radiator systems. One reason is that there is no R increase overall (for lack of horn loading). Also, the bass is crossed over at 1000 Hz or more. At that frequency, the inductor has a higher impedance, and therefore has more affect.

Gil

[Al Klappenberger]

Gil,

That plot is of the woofer installed in my Belle. It is the response in the horn. I did the example at 700 Hz becasue that's where the marker data was written out directly below the plot. 700 Hz is where I did the crossover in this case because I was using Altec 811b squawker horns at the time. For the Khorn you would want to know the impedance at 400 Hz. I did some data at other frequency point too. I skiped 400 Hz though. At 300 Hz it came out 5.07 +j1.63 Ohms. At that frequency the inductance is .864 mHy. I also did it at 500 Hz where the inductance seems to peak at 1.13 mHy. I haven't hand a chance to actually measure a Khorn myself but a friend did. It's similar.

Al K.

[WMcD]

Hmm, looking more closely, I can see that I jumped to a conclusion which I should not have.

I did misread the vertical axis and mistook a Zo in one place for a 20.

Smile,

Gil

[Al Klappenberger]

Gil,

1000 lashes with a wet noodle at sunrise!

Actually, the analyzer is realy measuring V/V. That's voltage A .01 / Voltage B .02 scales. The vertical scale then becomes 0 - 20 Ohms. The bridge has a 10 Ohm resistor in the B channel as a reference.

[Nklipsch]

www.allaboutcircuits.com or http://www.bcae1.com

Using a pc with moderately linear sound card,and sine wave generator software,a plot can be created. Use a 1k in series,and you can also use a large resistor for calibration. Cant find the value right now!(lent a book to a mate,who moved house!)

Then you hope your DMM is accurate from in the range your requiring-for subs probably their ok.