More resistance is less load??

Yes- you get lower power into higher impedances with solid state amps. But you also get lower distortion- essentially the amp will sound smoother and more detailed as a result.

With common 60V transistors (like 2N3055) highest supply would be +/-30V making maximum RMS power with 16ohm speaker only 28W.

It's easier for a speaker manufacturer to get a steady impedance (which amongst other things increases the potential for frequency linearity) with lower impedances than higher numerical impedances.

This statement is false in several ways. Impedance has nothing to do with the linearity of the impedance curve- that has entirely to do with the design of the speaker. Secondly, a linear impedance curve likely benefits a tube design rather than a solid state design- the real issue is whether the amplifier can operate as a true voltage source- if so the linearity of the impedance may well be moot.

A long time ago amplifiers had a great deal of trouble dealing with low impedances and high power. With the advent of reliable ss amplification that can double down, speaker designers were free from those constraints to advance other aspects of speaker design.

This statement is also false. What really happened in history is that the idea of an amplifier being able to operate as a voltage source was proposed by EV and MacIntosh in the 1950s. The idea was to eliminate the guesswork of setting up a loudspeaker. At the time, a speaker was usually set up with a midrange control and a tweeter level control. These were not there to adjust the speaker to the room, they were allow one to adjust the speaker to the voltage response of the amplifier used (as such, both the speakers and the amplifiers in use would be what I call Power Paradigm technology).

With the introduction of the Voltage Paradigm, level controls on the speaker were no longer needed.

At the same time, the industry was transitioning from tube power (which is expensive, so much so that most loudspeakers were fairly high efficiency) to solid state. It was a lot cheaper to build a solid state amp (no filament circuit and no output transformers) and yet the industry was able to charge nearly the same money for the amps. Solid state amps, requiring a lot of feedback for linearity, had a side benefit from that feedback (and the otherwise lower output impedance of the output devices) which was that they could much more easily operate as a voltage source.

The significance of this (especially the increased profit) was not lost on the loudspeaker industry. It takes a lot of precision (costs more) to build high efficiency loudspeakers, and so with the new higher powered solid state amps, it was possible to build lower efficiency speakers and yet **charge nearly as much for them** (and if you are following the dollars here then you see what this transition was really all about). To get back some of the loss in efficiency, loudspeakers began to appear that were 4 ohms rather than 8 or 16 (or even 32 ohms...). The lower impedance asked more power of the solid state amps, and so a speaker might be 10x less expensive to build and not really seem to be all that much harder to drive (while costing the customer nearly the same retail dollars).  Of course the amplifier ran hotter...

In time solid state amps also became reliable driving these impedances, but to say that there was a constraint removed to advance speaker design is not at all accurate, unless you look at it from a perspective of profit.

A long time ago amplifiers had a great deal of trouble dealing with low impedances and high power.

I’m not sure why. Output transformer can be designed for any load. Perhaps higher impedance speakers had less distortion? Perhaps it was difficult for high power SS amps to deliver high output voltages?

A long time ago amplifiers had a great deal of trouble dealing with low impedances and high power. With the advent of reliable ss amplification that can double down, speaker designers were free from those constraints to advance other aspects of speaker design.

AFAIK long time ago all speakers had higher impedance, mostly 16 ohm but some even 32 ohm.  

Linearity would be much better with underhung motors (coil within a gap) instead of commonly used overhang motor design (gap within a coil). Only few manufacturers use more expensive underhung (larger magnet) design.  One of them is Acoustic Zen.

there might be measurements that suggest that there were other benefits that made foregoing better measurements at the amplifier worth it.

That measurement is power, which equates to sound pressure. Otherwise there is no real advantage to a low impedance load.

One thing I forgot to mention is how much more important the speaker cable becomes when driving a lower impedance. Into 4 ohms the speaker cable is critical and differences are easily heard between them as a result. Conversely, at 16 ohms the speaker cable is far less important and longer runs can be made with less deleterious effects.

Into 4 ohms, the speaker cable can easily degrade the damping performance of the amplifier. So if your speaker requires a higher damping factor (+20:1) then it will be important to keep the cable as short as possible to minimize its effects. I've seen this have dramatic effects on the resulting bass impact!

Almarg, thank you for your elegant, articulate and patient responses!

Hi Jim,

As a consequence of the equations that were cited earlier for the relations between power, voltage, current, and resistance, it can be inferred that delivering a given amount of power into 8 ohms requires much more voltage and much less current than delivering the same amount of power into a very low impedance, such as 1 or 2 ohms or less. While conversely delivering a given amount of power into a very low impedance requires much more current and much less voltage than delivering the same amount of power into 8 ohms.

And, hypothetically speaking, if the load impedance were to truly approach zero (i.e., a true short circuit), the amount of current required to provide any voltage and deliver any power would approach infinity.

For an amplifier to be able to deliver amounts of power that are generally desirable into both 8 ohms and very low impedances it therefore has to be able to supply BOTH very large amounts of current and relatively high voltages. To be able to do that it will have to be much larger, heavier, and more costly than would otherwise be the case. And most likely sonic compromises would result as well.

There are some amps that can supply substantial amounts of power into impedances of 1 ohm or thereabouts, and in a few cases perhaps into even lower impedances, but in all of those cases I am familiar with the amps are big heavy monsters, which consume large amounts of electricity, generate a lot of heat, and don’t necessarily sound as good as many other amps that are in the same price range.

Regards,
-- Al

almarg - ok, here's a variation of the question. When building an amp, couldn't the manufacturer's assume the worst - that speaker resistance, while "typically" 4-8 ohms will occasionally get very close to zero and build the amp on that basis?

Jim, yes, any frequency other than zero Hertz (which is DC) is alternating current. And of course a music signal nearly always consists of a mix of a great many AC frequencies that are simultaneously present, at a wide variety of "amplitudes" (i.e., strengths, or magnitudes).

why can’t the resistance from the speaker, that the amp relies on be built into the amp itself?

In order for a speaker to absorb electrical power, some fraction of which it converts into sound, it has to have resistance. And for it to absorb a reasonable amount of power when provided with voltages that are reasonably practical, that resistance has to be relatively low (e.g., in the vicinity of 4 or 8 or 16 ohms or so). If a similar resistance were placed into the amp, the resistor in the amp would absorb power but convert it into heat rather than sound. So that resistor would serve no useful purpose, but would reduce the amount of power the amp would be capable of providing to the speaker.

Best regards,
-- Al

almarg - believe it or not I didn't "realize, the signal provided to a speaker consists of various frequency components each of which is AC (alternating current)".

I'm kind of amazed, now, at how little I know and how little I questioned. This stuff is actually quite fascinating.

Ok, here's another question: why can't the resistance from the speaker, that the amp relies on be built into the amp itself? Why build a product that has a vulnerability even though the manufacturer knows the amp will be paired with an unknown speaker, of unknown quality and impedance?

^^ Yes. Take a look at the specs and you will see.

This is because, regardless of the amplifier technology (tube, traditional solid state or class D), the distortion will be higher driving lower impedances. This is both measurable and audible.

Ralph, is this true for amps that double down to one or two ohms and are stable into those loads?

Since gardening season is starting here, this thread makes me wonder about the impedance of my watering faucets…I don't want to water a 2 ohm garden with a 16 ohm hose.

OK- now we got the basic part- a greater load is less resistance, less of a load is higher.

There is more of a take-away than this however. It has to do with power vs distortion (and also in the case of tubes bandwidth may be affected).

**If** high fidelity is your goal, your amplifier dollar investment will be better served by a loudspeaker of higher impedance (all other things being equal). This is because, regardless of the amplifier technology (tube, traditional solid state or class D), the distortion will be higher driving lower impedances. This is both measurable and audible.

**If** your goal is sound pressure, there is a slight argument for a lower impedance speaker (in the case of an 8 ohm speaker, a 3 db higher sound pressure will be had if you go to 4 ohms) all other things being equal. This however is only true if you have solid state, and still might not be true unless the amplifier can support the additional current required to double its power (a 3db increase in volume requires a doubling of power, and cutting the load impedance in half could win you that 3 db).

The kind of distortion that is increased in all cases by the use of a lower impedance load is audible, despite the increase usually being rather slight. This is because the additional distortion produced is of the kind to which the ear is extremely sensitive: higher ordered harmonics and additional IM distortion.

The presence of distortion obscures low level detail due to the ear’s masking principle. The ear/brain system converts all forms of distortion into tonality; thus the slight additional distortion contributes to brightness and harshness.

As a result, generally speaking, any amplifier will therefore be smoother and more detailed driving a higher impedance load. Its a simple fact that you can make an amplifier work hard by making it drive a difficult load, but that it not the same as having it sound its best!

I know that many people have fallen in love with certain low impedance speakers (many of which have excellent properties) and they are also quite happy with the combination of amp and speaker that they have. All I am pointing out is that if that same speaker were somehow 8 (or better yet 16 ohms) instead of 4 ohms, it would sound smoother and more detailed with no real downside except a slight amount of power in the case of solid state amps (tube amps would make the same or slightly higher power, likely with wider bandwidth as the output transformer is more efficient driving higher impedances).

For speaker designers, a simple way to make their speaker seem smoother and more detailed is to simply make it higher impedance...

Yep open circuit is no load just like an empty truck.

Falconquest,

    I=E/R   Indian sees Eagle over Rabbit
    R=E/I   Rabbit sees Eagle over Indian
    E=IxR  Eagle flying sees Indian and Rabbit on the same level   :)

Yes, RMS = root mean square. As you probably realize, the signal provided to a speaker consists of various frequency components each of which is AC (alternating current). Amplifier power capability is defined based on the simplified assumption that the signal consists of a pure sine wave at a single frequency, with that single frequency being anywhere within some range of frequencies, such as 20 Hz to 20 KHz. The RMS value of a sine wave equals its peak (maximum) instantaneous value divided by the square root of 2, or approximately 0.707 x the peak value.

In audio voltages and currents are usually defined on an RMS basis, in part because the amount of power supplied to a resistive load that can be calculated based on RMS voltage and current numbers (even for waveforms that are not sine waves) equals the amount of power that would result from a DC (direct current) voltage and current having the same values, that amount of power in turn being proportional to the amount of heat that is produced when supplied to ("dissipated in") a resistive load.

By the way, one thing that often causes confusion in this context is that the word "peak" can be used to mean two different things. It can refer to the peak (maximum) value of a sine wave or other signal at any instant of time during each of its cycles (corresponding for a sine wave to the RMS value divided by 0.707), or it can refer to the peak (maximum) RMS value that can be reached by that sine wave or other signal during normal (or other) operating conditions.

Kudos for your interest in these matters. Regards,
-- Al

Thanks this is very helpful. I agree falconquest it is counter-intuitive, but the equations and explanations are clear & helpful. I have some clarity now.

almarg has brought up the notion of "power".

Used to be, when I paid attention to these things, 40 watts per channel was 40 watts "RMS".

How is RMS (root mean square I believe) to be understood?

In other words, think of the flow of electrons like a water hose as someone already mentioned. The less resistance, the more flow and therefore the more draw on the amplifier to push those electrons through the wires. The more draw, the harder the amplifier has to work. It's rather counter intuitive unless you think about it in those terms. 

Good responses by the others above, except that the reference to E/IR should be E = IR (meaning E equals I multiplied by R). (E is commonly used in the context of Ohm’s Law to denote volts, and means the same thing in that context as V).

So when Ghosthouse referred to V/I = R, he was correct. By simple algebra V/I = R is equivalent to V (or E) = I x R = IR.

Regarding the voltages that are typically provided to speakers, for a resistive load:

Power = (I squared) x R = (E squared)/R
where, for example, P is expressed in watts, I in amps, R in ohms, and E in volts.

Therefore E = (Square root (P x R)).

So for example 100 watts into an 8 ohm resistive load corresponds to:

Square root (100 x 8) = 28.28 volts.

Regards,
-- Al

I thought it was E/IR   :-)

Expanding on the relationship Audioman pointed to...
The relationships between resistance (ohms), current and voltage are defined by Ohm’s Law: V/I = R where V is voltage (volts); I is current (amps); R is resistance (ohms).

See link here....
https://www.physics.uoguelph.ca/tutorials/ohm/Q.ohm.intro.html

At constant V, if R increases, I will decrease.
At constant V, if R decreases, I will increase.

As a simple example (I’ll defer to Almarg or Atmasphere for their input on a better voltage value or range to use in discussing an operating amplifier)....

At 120V if R = 8 ohm, I = 15
At 120V if R = 4 ohm, I = 30
At 120 V if R = 16 ohm, I = 7.5

So you can see how greatly current requirements are affected by resistance. It’s important to remember that a speaker’s resistance might not be constant across the frequency range. Some portions of the frequency range might be associated with low resistance and consequently run up against the current limitations of the amp reproducing sound in that frequency range.

Hope this helps.

Easy - The lower the resistance, the more current flow can occur.  Think about resistance as a water hose.  The less resistance the more flow.  When you have more current flow, there is more heat generated because conductors are not perfect, they also have a resistance.  The amp runs hotter and requires bigger cooling heatinks, bigger caps to store all those electrons, wider electrical traces, bigger posts, etc.