Of course, bigger wire is the best choice for all of the reasons cited. But, let's not lose sight of the math behind the question. Assuming that the 20% loss figure is accurate, 20% = 0.9 dB Is that amount of loss audible? No. Whether the loss of damping factor and variations in frequency response are more noticeable is debatable. Would it be a disaster to use 18 ga wire? No. Is doing so good practice? No.
Speaker wires shouldn't be longer than 60' ?
- Thread starterstonehedge99
- Start date
I don't think that's a fair statement for audio.
Absolutely heat will be dissipated when running high currents. But heat behaves linearly as opposed to audio's logarithmic behaviour.
I just can't see there being enough current being pushed for long enough to cause melting, UNLESS the dynamic range is very small AND the volume is really loud (like clipping) AND the music is absolutely non stop.
As a thought for future reference, this may be a situation where running 2 cables is an answer.
Because you are then looking at 8 ohm loads and half the current, your loss will be a much lesser proportion of the power.
And that reason is apparently that the people publishing those ratings don't understand there is no such thing as RMS power. I believe that basing the power on RMS voltage and RMS current measurements results in calculating the average power rather than RMS power.
Cable and voice coil failures are often related to excessive average power levels over some period of time and whenever you discuss RMS or average values you have to consider the time variance of the signal. So addressing topics such as heating in a wire means considering the time varying nature of the signal and not just the RMS or average ratings of the amp and speaker. What Chris was referencing is the fact that unlike some other electrical signals, with audio the signal usually varies significantly over time and thus there is typically a significant difference between the peak and average levels or a higher crest factor.
I'd guess that a cable between an amp and speaker 'burning up' generally relates to some amp or speaker failure, damage to the cable, etc. and that you would be hard pressed to find a situation that would result in 18AWG speaker cable 'burning up' that would not have the same result if the conductors were 12AWG or 10AWG.
Strictly speaking it is possible to calculate RMS power.
No one does and most people indeed are talking average power with a scant level of accuracy.
12/2 is quite possible not so easy to obtain in Sri Lanka where the OP is...
Power is indeed power. But we are talking about dumping 3.33 watts max per lineal metre. On the other hand, the voice coil of the speaker in question is dissipating 311 watts as kinetic and heat energy. In a space of maybe 100mm by 100mm. The heat density is just so much greater which is why voice coils burn up.
I'm willing to be wrong, but I just can't see there being enough heat in this to melt insulation...
No one does and most people indeed are talking average power with a scant level of accuracy.
12/2 is quite possible not so easy to obtain in Sri Lanka where the OP is...
Power is indeed power. But we are talking about dumping 3.33 watts max per lineal metre. On the other hand, the voice coil of the speaker in question is dissipating 311 watts as kinetic and heat energy. In a space of maybe 100mm by 100mm. The heat density is just so much greater which is why voice coils burn up.
I'm willing to be wrong, but I just can't see there being enough heat in this to melt insulation...
This addresses that, http://www.eznec.com/Amateur/RMS_Power.pdf, and yes, you can calculate the RMS Power, but that is that is not what is actually measured, calculated or reported for power amplifiers or speakers and is not relevant to the issue of line losses or heating of cables. http://www.n4lcd.com/RMS.pdf offers an explanation for how "RMS Power" became commonly misapplied, so widely so despite it clearly being errroneous that many people have given up fighting it, however I think it is relevant for this topic.
Similar threads
