16 ohm speakers: any amp sounds better with more resolution. speaker cables less critical.

McIntosh Speaker Wire Gauge/Distance/Resistance Chart (pdf pg 6)

http://akdatabase.com/AKview/albums/userpics/10004/MC2250%20Owners.pdf

Indicates: any given gauge: 8 ohm load allows double the length of a 4 ohm load. Presumably 16 ohm load would allow twice the 8 ohm allowable length (4 times allowable 4 ohm length).

I have to believe many if not most are using larger gauge than needed, however, some with 4 ohm speakers that dip to 2 .. some people might not be using a large enough gauge. And people are overly concerned about long speaker cables.

..............................

Text from my MC 2250’s manual:

"Selection of the proper gauge wire to connect the loudspeakers to the amplifier preserves the quality of sound reproduction for which the loudspeakers have been designed. If undersize wire is used, resistance is added to the amplifier/loudspeaker combination which adversely affects the performance. Added resistance causes depreciation of damping characteristics, modification of frequency response and reduction in power to the loudspeaker.

Use lamp cord or wire with similar insulation to connect the speakers to the amplifier. In all cases, the leads to and from the speaker should be twin parallel conductor or twisted pair.

When using 8 ohm speakers and for the normally short distances of under 30 feet between the amplifier and speaker, #18 wire or larger can be used.

For distances over 30 feet a larger diameter wire is required. Select the correct size from the chart below. The DC resistance of the speaker leads should be less than 5% of the speaker impedance. Resistance of the leads must be computed for the length of wire both to and from the speaker or speakers.

Wire lengths above represent the wire resistance equal to 5% of the speaker impedance. For multiple speaker operation, run separate leads from the amplifier to the speakers."

Reactance of the cable is also important. But now you’re starting a different topic.

My amp has 4, 8, and 16 taps. If my speakers are 4, what strain, if any, on any of my amp tubes will that have?

If you put the speaker on the 4 ohm tap, none, but the output transformer will likely be losing some performance- they can often lose as much as an octave of bandwidth in the bass, and they will run warmer. That warmth is also causing the amp to make slightly less power, as the heat is coming from the power tube output.

And yet, the guy who designs some of the most fantastically successful speakers made anywhere in the world today says nevermind 16 ohms, even 8 ohms is obsolete.

As far as solid state is concerned, this may well be true from a marketing perspective. But this is high end audio, where minimizing colorations is important, and distortion causes most of the colorations we hear- for example the brightness/harshness of solid state. The 3dB more power thing is a really weak argument- put another way, if you want to get the **best sound quality**, your amplifier investment dollar is best served by a higher impedance speaker. OTOH, if **sound pressure** is your goal, then you have a 3dB argument for 4 ohms (as opposed to 8) **if** you have a solid state amp that doubles power into 4 ohms. But you really pay a price doing 4 ohms- not only do you have more distortion, but most amps will run warmer (even class D) and the speaker cable becomes critical. On that basis it would be a stretch to claim state of the art performance when the amplifier is thus compromised.

His demo is about loudness, saying 4 ohm will produce 3db more VOLUME than 8 ohm. IOW, it helps him with efficiency.

To be completely correct, it helps with **sensitivity** which is not the same as efficiency. Efficiency is unaffected. What the higher sensitivity number is saying is that the amp is now being asked to make twice as much power! That's completely different from a speaker that is 3dB more efficient- the latter would mean that an amp with half the power would play just as loud!


Like everything electronic, math is involved. Efficiency is stated as a certain sound pressure with 1 watt. Sensitivity is stated as a certain sound pressure with 2.83 volts.  Into an 8 ohm load 2.83 volts is 1 watt; into 4 ohms its 2 watts. This is a 3dB difference. Since tube amps don't double power as impedance is halved, efficiency is the more useful specification, but since the industry went solid state, the sensitivity spec has taken over. You can make any speaker seem more "efficient" by reducing its impedance though parallel drivers but what is happening is the sensitivity is being increased while the efficiency is unaffected. You do have to pay attention to this nuance when selecting a speaker to work with an amplifier!

1. how importance is bias to sound? (I don't care about heat or tube life).

......................

2. speaker cable inductance ..... gauge ____ awg

mijostyn said

" the wire you are using probably has high inductance. You should use 18 gauge wire that is a jacketed twisted pair like this".

Guage: Cat 5 8 strands is 15 awg. Two Cat 5's, 16 strands, is 12 awg.

https://www.wirebarn.com/Combined-Wire-Gauge-Calculator_ep_42.html

I don't really understand inductance. If speakers are high efficiency (mine are), and I/you have more than enough power from the amp, speaker

3. equiv 15 awg, is inductance an issue?

Inductance (Back EMF) might increase amount of current needed, but

4. enough power: does it have any effect on the music's frequencies?

Bias can be pretty important. Usually it is used to place the output device (tube or transistor) in the most linear portion of its operating curve. So it can have a noticeable and measurable effect on the sound.


The inductance of the speaker cable plays a role, but so does capacitance and DC resistance. I would not place too much importance on any one characteristic. In theory, the cable has what is called 'Characteristic Impedance' which is the character where the cable is properly terminated by the impedance specified. Put another way, if the cable has a characteristic impedance of 8 ohms, if an 8 ohm load terminated the cable, there will be no reflections from the load back to the source. In this case from the speaker back to the amplifier. But in practice, no speaker is a perfect resistive load, and the back emf from the speaker dominates the minor reflections, so what becomes more important is that the speaker cable simply be kept as short as possible so that its errors are minimized. This FWIW is why I developed a balanced line preamp, so the amps could be placed as close to the speaker as practical. When doing this you hear an immediate improvement in bass impact and resolution. In a nutshell- keep your speaker cables short and used cable with plenty of conductor to do the job.


The amount of power you have might affect how much bandwidth the amplifier has. This is particularly true of SET amplifiers, where power past about 7-8 watts falls short of the definition of 'hifi' (the smaller the SET, generally the wider bandwidth it is; this is why the type 45 power tube, which is only good for 0.75 watt, is the 'best' sounding). With push-pull tube amps this limit is more like 60-120 watts, but in both of these examples its assumed that there is an output transformer on the tube amp (some tube amps don't have output transformers so are not bandwidth limited at any power level). Some solid state designs don't scale well as power is increased, but most of those I've encountered are older designs employing coupling capacitors at the output of the amp. 

Dear friends: I respect all gentlemans opinions but this is what defines the amplifier need it. Not what said atmasphere or Dudley or Fremer or Teektonany one else  because one is the manufacturer theoretical impedance ( been 4ohms or 16 ohms. ) at the end it does not matters till we know what these facts shows us:


https://www.stereophile.com/content/acora-srb-loudspeaker-measurements

https://www.stereophile.com/content/g%C3%B6bel-high-end-divin-marquis-loudspeaker-measurements

https://www.stereophile.com/content/von-schweikert-ultra-55-loudspeaker-measurements

https://www.stereophile.com/content/magico-m2-loudspeaker-measurements

https://www.stereophile.com/content/wilson-audio-sasha-wp-loudspeaker-measurements

https://www.stereophile.com/content/monitor-audio-gold-300-loudspeaker-measurements

Regards and enjoy the MUSIC NOT DISTORTIONS,
R.

I am not sure that generalization (16 ohms = lower amplifier distortion) would apply to every variety of solid state amplifier.

@lewm If you know of one that actually has lower distortion into 4 ohms as opposed to higher impedances I'd love to know about it. I've yet to find one, starting from primitive driver transformer designs of the 1960s to self oscillating class D amps of the 2020s. If there is one out there it is a very rare exception!