What's with 4 ohm speakers?

Using solid state amplifiers without output transformers a 4 Ohm nominal impedance allows peaks 3dB louder than 8 Ohm speakers with the same cabinet size and low frequency cut-off which ultimately limit efficiency per Hoffman's Iron Law. This is generally a better engineering choice than doubling cabinet size (thus halving the spousal acceptance factor ) or choosing a low frequency cut-off 1/3 octave higher.

People don't go too over-board with lower impedance in the home market because of

1. How the FTC requires manufacturers to rate stereo and mono home amplifiers - they must be "pre conditioned" at 1/3 of rated output power and the power dissipated into low impedance loads would make the numbers look bad so the capability to run 2 Ohm loads isn't usually advertised and consumers would be leery of buying such speakers to go with their "4 and 8 Ohm compatible" electronics.

2. Some amplifiers are unstable (they start to oscillate) driving low impedances; and starting with a 2Ohm nominal impedance minimums of 1 Ohm aren't unreasonable.

3. Some audiophile amplifiers have silly high output impedances which interact with the speaker's varying impedance to change the frequency response and this is exacerbated with low load impedances. Output Transformer Less Tube amps are especially bad although single ended triodes without global feedback can also have problems.

For instance an Atmasphere M-60 Mk.II.2 has a 4.1 Ohm output impedance.

Driving a 3-way speaker with impedance varying from 16 to 64 Ohms this would cause a 1.4dB output difference between the minimum and maximum impedances.

With 4 to 16 Ohm impedance the difference would be 4dB. This is not atypical for a 3-way - the reactive components for a Zobel network to counter the bass driver's resonant peak would be too big and expensive so the best you can do is bring it down with a resistor in parallel.

At 2 to 8 Ohms it'd be 6dB.

Apart from this edge case the effects on distortion aren't interesting compared to what the speaker is adding to the sound.

If you do want to run such an amplifier you'll do well going out of your way to buy speakers with high (16 Ohm nominal) and intentionally flat impedance.

I wrote:
Apart from this edge case the effects on distortion aren't interesting compared to what the speaker is adding to the sound.

That's not quite right - my power amplifier biases run towards push-pull solid state class AB for practical reasons although I like building with tubes where I can actually see how a negative charge on the control grid surrounding the cathode limits electron flow to the plate on the outside.

There are output stage device, topology, and biasing combinations which won't play nice with low load impedance.

AudioKinesis writes:
>In reply, I'd like to point out that, with the same 4 to 16 ohm impedance difference described here, the power that a transistor amp puts out varies by 6 dB, because it is putting out constant voltage rather than constant wattage. Why does the audio world accept this without a blink, and yet think there's a problem when a tube amp exhibits less variance in power output into the same load??

Those of us with a little technical knowledge accept and expect it because at lower frequencies dynamic loudspeaker driver output is proportional to voltage.

Individual drivers together with their enclosures form spring-mass systems where oscillating movements at frequencies approaching system resonance take less energy for a given excursion and therefore less power since that distance is traversed in a fixed time interval like the five milliseconds for a half wave of a 100Hz tone.

Even order acoustic cross-over networks with their drivers in-phase (or to be pedantic a multiple of 180 degrees out of phase with the polarity of one driver flipped for odd multiples) have an efficiency peak at the cross-over point due to mutual coupling which is 3dB assuming the two drivers have matching efficiency.

To apply audiophile pseudo-science dictating simpler is better you want to go with this flow (implying negligible output impedances like the .01 - .02 Ohms of many transistor amplifiers) instead of adding electrical circuits and their errors to mash things together.

This is completely orthogonal to the maximum output an amplifier may deliver where some technologies (output transformer equipped amplifiers with taps optimized for various load impedances and some digital amplifiers) allow more similar limits regardless of load impedance.

I'm also ignoring that you can design an amplifier with output current proportional to the input signal and combine it with an appropriate cross-over network to yield better performance because audio output is no longer reduced by voice coil heating (which increases resistance and therefore decreases current flow with a constant voltage source) and affected by inductance changes as the voice coil moves through its range (this causes IM and harmonic distortion) since such combinations are uncommon and would be a hard sell to the audiophile market that could no longer make arbitrary amplifier and speaker swaps. DSP cross-overs with current/voltage programs and relays to switch amplifier configuration would make a fun Burning Amp presentation.

> Back to the original question, what's with 4 ohm speakers if they're harder to drive, well in general they can play louder with a solid state amp, and most people have solid state amps, so they get more sound per dollar with 4 ohm speakers (quantity outsells quality).

Quantity becomes quality when it avoids clipping.

When I play a nice jazz recording at a less than live but realistic sounding 85dB average SPL with 20dB peaks that can be pushing 108dB 1 meter from a speaker.

With 86dB / 2.83V / 1 meter speakers having a 4 Ohm bass compliment I'm just going to miss clipping an amplifier built to meet a 100W into 8 Ohm FTC rating (they measure with sine waves that have a 3dB crest factor).

Keeping the same efficiency but increasing impedance to 8 Ohms takes a 200W amplifier.

This fits well with the desire for single box speakers especially where last octave extension is desired (example - the Revel Salon 2). Keep the size, loose the bottom octave or move it to a sub-woofer, and you could have another 9dB of efficiency for a 92dB sensitive/efficient 8 Ohm speaker.

The two approaches work for different market segments and I'm aware that yours is the latter.

Atmasphere writes
> I wrote:
> Those of us with a little technical knowledge accept and expect it because at lower frequencies dynamic loudspeaker driver output is proportional to voltage.

>The statement is ambiguous. We know that doubling power is 3db, and that there is or should be a direct correlation with driver output. Since this is so then driver output is also proportional to power.

Nope. You're confusing voltage and power where power is voltage squared divided by impedance.

Reactive impedance varies with frequency. Inductive impedance magnitude is 2 pi f L with f ferquency in Hz and L inductance in Henries. Capacitive impedance magnitude is 1 / 2 pi f C with capacitance in Farads.

Driver + enclosure combinations are reactive loads with their mechanical parameters reflected in the electrical characteristics at the driver terminals. The driver's in-enclosure compliance Cms shows up as an inductive reactance Lces. The moving mass Mms works as a capacitive inductance Cmes.

Driver voice coils are inductive which also causes impedance to increase with frequency.

You can have a 40 Ohm maximum impedance at driver + enclosure resonance with impedance around a driver's 6 Ohm voice coil resistance as you move through its pass-band before its voice coil inductance becomes significant and increases to 20 Ohms before leaving the audible spectrum.

Over the same range the same voltage can yield the same output. 2.83V might be 90dB SPL although that varies somewhere between 1/5W and 1 1/3W electrical.

As noted complete speakers complicate things more. You get increased output as the speaker transitions from full to half space radiation (baffle step) and can have rising response as driver directivity increases with frequency. When cross-over designers compensate for that with a series load power impedance increases at those frequencies and power dissipated decreases.

Amplifiers which don't accommodate these physical realities with terminal voltage that's a fixed multiple of input voltage regardless of load impedance aren't universally useful in high-fidelity applications for speakers having impedances that are otherwise compatible causing neither instability nor power dissipation issues.

I should probably proof-read better before posting
>The moving mass Mms works as a capacitive inductance Cmes.

capacitive impedance

>When cross-over designers compensate for that with a series load power impedance increases at those frequencies and power dissipated decreases.

series load impedance increases at those frequencies and power dissipated decreases.