Increase total Speaker Impedance -- Potential Solution

Hi All, 

Years ago I asked about using a resistor to increase the total impedance of a speaker -- I know that's a bad idea now because Impedance is NOT the same as resistance. I did get a pair of Paul Speltz's Zeroformers and they are great!

However, reading more about this, I thought another, more cost effective solution. Why not just buy a low cost, 4 ohm speaker, remove the diaphragm inside, and series-wire it? I'm reading some of the Zeroformer's technical documents and it sounds like it's rather similar to a speaker coil that doesn't produce any sound. Therefore, if I just series-wire a 4 ohm speaker to my existing 2 ohm speaker, wouldn't the total ohm seen by the amp be 6 ohm? 
Yes, sort of; clearly, a definite maybe.  :-)   What you need to do is buy two more speakers identical to what you currently have, and wire them in series.   That would do it.  

The reason one cannot put another 4 ohm driver in series with it is the speaker impedance is the net result of the cabinet, diaphragm, air movement, etc. -  all those nasty speaker variables defined by Thiele and Small back in the late 70's. 

You are on the right track with a transformer though.   Another option is simply buying a amplifier that can drive a 2 Ohm load sufficiently,   Some vacuum tube amps have 2 Ohm taps, which really solves the problem - assuming it provides the power you need. 
I hope Almarg or Atmasphere chime in, they know how this stuff works.
What exactly are you trying to do?

If you add anything in series you can alter the impedance, but you also lower the output.

The question really is WHY.

Thanks Bob (Gdnrbob).

To add to the previous comments, an autoformer such as the Zero inserted between an amp and a speaker will provide very different (and most likely much better) results than if a speaker driver or a speaker is inserted in series between them.

First, an autoformer will not consume a significant amount of the power that is being put out by the amp. While a 4 ohm load inserted in series with a 2 ohm load will consume two-thirds of that power. And a 2 ohm load inserted in series with a 2 ohm load will consume half of that power. (Note that I used the term “load,” rather than “speaker” or “speaker driver.” The impedance of a 2 ohm or 4 ohm speaker will certainly not be 2 or 4 ohms at many frequencies, and will not be purely resistive at most frequencies either. Which leads in to my next point).

Second, note Spatialking’s reference to using speakers that are identical to the ones you have. If they are not identical, most likely the frequency response provided by both speakers will be significantly altered by differences between how the impedance of each speaker varies as a function of frequency. That is because how the power provided by the amp at a given frequency divides up between the two speakers depends on their impedance at that frequency. If the two speakers do not have identical variations of their impedance over the frequency range the power provided by the amp will divide up differently at different frequencies.

Third, if your present speaker and a speaker that is added in series and is also producing sound are not equidistant from the listening position "comb filtering" effects could very well result, which would degrade the sound.

And on top of all that, of course, there would be the cost of adding a second pair of speakers.

The bottom line: Stay with the autoformers :-)


-- Al

To further explore the amp-speaker interface, recently purchased an Audiopax Maggiore100 to pair with Zu Defintions. Noticed the output impedance is a bit high at 2.5 Ohms. The speakers are nominally 8 Ohm but dip to 6 ohms in the bass region. It's been said the speakers are an easy load and 100 watts to power a 101db efficient speaker should result in considerable headroom. Question is will the somewhat high output impedance of the amp still affect the bass negatively?
@dentdog, that sounds like a very nice pairing to me.

It appears that the amp would be driving a pair of "full range" drivers in your Zu Def IVs that are specified to cover 30 Hz to 12 kHz, plus a tweeter. While the speaker also incorporates a powered sub specified to cover 12 Hz to 80 Hz. Given that the speaker is an easy load the phase angles of its impedance probably don’t reach any extreme values, i.e., the impedance is probably mainly resistive. In which case it can be calculated that the dip from 8 ohms to 6 ohms, driven by 2.5 ohms, would result in a dip in frequency response of only about 0.7 db, relative to the response a very low output impedance would result in. And if that proves to be noticeable, which I doubt, chances are that a slight tweak of the controls of the speaker’s subwoofer section would compensate.

Enjoy the new amp! Regards,
-- Al

Thanks Al. Nice to have expertise on the board.
Hello! Thanks all for your input and thanks  @almarg  for your expertise again!

The reason I needed to figure out a way to increase the impedance of my DIY unit is that the Zero Autoformers will soon have another use. I purchased a tube amp to be used with my Martin Logans and I heard the electrostatics' impedance drops really low and a pair of Zeros will help big time. Hence, the Zeros will now be connected to the electrostatics instead.

I'm not ready to chip up another $500 for another pair of zeros since if that's the case, I'd rather just sell all my drives and just buy another pair of speaker and kiss the DIY goodbye. Luckily, I found a creative way to re-wire the speaker so the impedance is now 4 ohm instead of 1.6 ohm.

How I did it:

Originally, almost all of them have separate crossovers and wired in parallel. Now I have three crossovers:

1x 2 way crossover at 2000hz
1x 1 band-pass 800-5Khz
1x low pass 800hz


1x Dynaudio D76AF
1x Dynaudio 15W75 Esotec
1x Dynaudio 17W75
1X Morel Supreme Tweeter (forgot model number, but ideal crossover I recall was 1800hz)
1X Dynaudio MW172 (8 inc)
With the exception of MW172, the rest are 8 ohms

2 way crossover high-pass connection goes into Morel Tweeter. The same crossover, low-pass connection goes into 15W75 and 17W75 wired in series. I understand they are not identical, but I see they have similar frequency response plots. Then, still using the same low-pass connection, wire the MW172 parallel to the 15w75 and 17w75 series, along with a series 1.5 ohm resistor. So in the end, I have all the woofers on the same low-pass connection of the 2 way crossover. However, the MW172 is crossed over a second time with the 800 low-pass filter. The D76AF stands along with the band-pass filter. Overall impedance showing in my multi-meter is 4.1 ohm.
Question is will the somewhat high output impedance of the amp still affect the bass negatively?
Probably not, unless slightly enhanced bass is seen as a negative. It won't be by much though.
Overall impedance showing in my multi-meter is 4.1 ohm.
That is the DC resistance and is only the drivers that don't have a cap in series with them. To really know the **impedance** (as opposed to resistance), you need a different test, since impedance often relates to frequency and can be quite different from resistance!

You'll need a variable resistor that might cover the range of the drivers. A 50 or 100 ohm potentiometer should do the trick. You'll also need a sine wave generator which will drive the speakers. We're not talking about any power here. You might be able to find an app for your phone that will do the job. You'll also need your DVM.

Set the frequency at which you want to measure. Put the potentiometer in series with the speaker. You'll only be using the center connection and one side.  Apply the signal. set the control so that the AC voltage drop across the speaker is the same as the control. Switch to Ohms and measure the control (out of circuit is best). That is the impedance of the speaker at that frequency. Continue to do this for all frequencies of interest.

You'll see that there is a low frequency impedance peak, followed by a lower impedance which then rises as frequency is increased. There might be dips at crossover frequencies. Check the areas of interest (crossover frequencies and then 500Hz above and below each crossover frequency). Then you can draw this all out and see what you got.

As a general rule of thumb when you put drivers in series they really should be the same kind of driver for the reasons Al points out above!