If I remember correctly :
the papers/articles probably refer to Ottala's claims in the eighties... with asymetrical square waves representing signals in real life. Douglas Self and Keith Howard did not find facts allowing to confirm this.
Last edited:
Pretty universal, and not a sign of good or bad design practice, more a judicious interface solution. Re Andrews comment, I was thinking you might like to go with an active speaker design
Last edited:
I'm kind of in between, to be honest. While I would never spend $500 for 1m of wiring, there is no doubt that difference in cables will measure differently.
The fact that it is not detectable is opinable and I hope we won't get into yet another cable thread diatribe.
I read somewhere that a guy noticed a definite clearing up of the sound going from 14 AWG to 2 (two!).
While it would be fun to go into this rabbit hole (once more), I think the amperage spike question has been addressed and figured out. Lower AWG than voice coil is enough.
I would be interested to see if somebody ever tried to fry speaker wires and if they succeeded before frying the driver.
You would be well served, instead of asking questions tying people to your way of thinking, to instead ask the question of "how does one choose speaker cable gauge? what are the considerations> what is the minimum practical size and how would one calculate it?"
The "peak music current" path you are going down is a foolish concept. Power dissipation in the wire is a safety concern, not a sonic one, and it isn't based on peak current. Insertion loss and peak-peak frequency response variations due to load impedance should be considered well before any current capability unless you are using 30AWG wires. You will probably find that satisfying insertion loss and frequency response goals will provide a wire gauge that will easily handle any amount of power a typical 100W amp will provide.
A 24AWG wire (going smaller would be crazy) is 25 milliohms per foot a 10' run is 1/2 ohm, plus contact resistances at connections. If you used this with a 100W amp into a 2.5 ohm load, you would dissipate ~80% of the power in the speaker and ~20% in the cable/connections. The peak/peak frequency response variations would depend in the impedance curve, but would most likely be objectionable (but perhaps not as audible as you might think) at this point. 20W dissipation along a 10' piece of wire will get warm, but probably isn't going to start a fire, but I wouldn't recommend it.
Look up Fred Davis' cable measurements. You will find that 12AWG is a good compromise for resistance and inductance. A 2AWG cable would most likely have very high inductance due to greater separation from the conductors.
The "peak music current" path you are going down is a foolish concept. Power dissipation in the wire is a safety concern, not a sonic one, and it isn't based on peak current. Insertion loss and peak-peak frequency response variations due to load impedance should be considered well before any current capability unless you are using 30AWG wires. You will probably find that satisfying insertion loss and frequency response goals will provide a wire gauge that will easily handle any amount of power a typical 100W amp will provide.
A 24AWG wire (going smaller would be crazy) is 25 milliohms per foot a 10' run is 1/2 ohm, plus contact resistances at connections. If you used this with a 100W amp into a 2.5 ohm load, you would dissipate ~80% of the power in the speaker and ~20% in the cable/connections. The peak/peak frequency response variations would depend in the impedance curve, but would most likely be objectionable (but perhaps not as audible as you might think) at this point. 20W dissipation along a 10' piece of wire will get warm, but probably isn't going to start a fire, but I wouldn't recommend it.
Look up Fred Davis' cable measurements. You will find that 12AWG is a good compromise for resistance and inductance. A 2AWG cable would most likely have very high inductance due to greater separation from the conductors.
Last edited:
Thanks for the suggestion, but I was actually trying to get some answers specifically regarding peak current handling.
And no, power dissipation wasn't my main concern, but rather investigation of some non linear behavior when a cable is pushed close to its max current handling capability.
I started this thread because I couldn't find another one addressing the peak current issue specifically.
It looks like it is a non issue with standard AWG anyways.
A peak voltmeter across a low value series resistor should allow good evaluations of the current peaks.
Good enough for ball park evaluation, but resistor would need to be small enough to be negligible compared to driver impedance, with lower related voltage drop.
Calculation of I = V/R would therefore have a low accuracy due to both measured V and R real value.
I ended up putting my Fluke meter in series with the cable and measured peak amps that way.
You can just as easily calculate the measurement error using an external series resistor, and you have control over the value.
Also note the bandwidth of the multimeter. Most are not fast enough to measure fast transient voltages. If you want to determine the actual transient current draw you will need to use an oscilloscope to measure the voltage across a series resistor.
Also note the bandwidth of the multimeter. Most are not fast enough to measure fast transient voltages. If you want to determine the actual transient current draw you will need to use an oscilloscope to measure the voltage across a series resistor.
It is not just as easily, because you have to go out and get a resistor. But that's probably not an issue for many.
The note on the multimeter is partof what I was asking in post #2.
Somebody mentioned other people did this test with 100W speakers and measured 25 A peaks.
I don't have an oscilloscope, so I'm inclined to buy that.
In any case, I needed that value to figure out wire AWG required.
As I said before, my conclusion is that as long as internal wire is heavier than coil wire, it should satisfy both peak amperage and low resistance requirements.
Litz wire is probably a nice touch to guarantee same wire resistance at all frequencies. I wouldn't mind spending a little extra for that.
I think you can buy spools of it relatively cheap on ebay.
Any material, close to its maximum handling, is non linear.
Be it tensile strength, vibration dampening and yes, amperage.
And please stop calling me foolish. I don't know you, and you don't know me.
Peace.
Or, since you originally opened this discussion relating specifically to the Eikonas, you could save a lot of wasted time debating perhaps irrelevant factoids, and simply ask members experienced with them in particular, and medium sized full range drivers in general:
Colin, Frank, how loud do play your systems?
What type and power level of amps, lengths and gauge of wire do you use?
How long have you been playing this game, and do you have war stories to relate of catastrophes that can be attributed solely to using the "wrong" wire?
In the over 15 yrs that I've been playing around almost exclusively with full range drivers, I've only ever seen drivers damaged twice - both happened to be Fostex FE127s, but the circumstances had nothing to do with even the higher end of normal playing conditions. One was a shorted output of a direct coupled class A amp that fed the full DC rail supply through the voice coil - do do know that under "optimal" conditions they can glow a delightful orangish/amber while heating up the glue enough for the coil layers to delaminate and seize in the gap, before finally fusing to protect the amplifier output stage?
The other was definitely "transient" related - when at a DIY get together, someone who definitely should have known better repeated hot-jacked the input signal on a mid level powered chip amp without disconnecting the speakers first.
Several of those "pops" were pretty loud, then suddenly, nothing - again, seized coils, just not in the gap.
Colin, Frank, how loud do play your systems?
What type and power level of amps, lengths and gauge of wire do you use?
How long have you been playing this game, and do you have war stories to relate of catastrophes that can be attributed solely to using the "wrong" wire?
In the over 15 yrs that I've been playing around almost exclusively with full range drivers, I've only ever seen drivers damaged twice - both happened to be Fostex FE127s, but the circumstances had nothing to do with even the higher end of normal playing conditions. One was a shorted output of a direct coupled class A amp that fed the full DC rail supply through the voice coil - do do know that under "optimal" conditions they can glow a delightful orangish/amber while heating up the glue enough for the coil layers to delaminate and seize in the gap, before finally fusing to protect the amplifier output stage?
The other was definitely "transient" related - when at a DIY get together, someone who definitely should have known better repeated hot-jacked the input signal on a mid level powered chip amp without disconnecting the speakers first.
Several of those "pops" were pretty loud, then suddenly, nothing - again, seized coils, just not in the gap.
Chris,
My question was about internal speaker wiring and what gauge is required to prevent peak current saturation phenomena (or wire melting, as an extreme).
By looking at the coil I quickly determined that I should focus on lowering R-L-C, which will take care of the amperage potential issue given its more stringent AWG requirements.
There really is no debating going on. Just me asking a question and finding the answer by having a second look at the physical characteristics of the driver.
I am not forcing anybody to waste their time, but if one wants to add to the conversation, all I ask is not to be told what and how I should have asked my questions, and to not be called foolish.
It's all good on my end
My question was about internal speaker wiring and what gauge is required to prevent peak current saturation phenomena (or wire melting, as an extreme).
By looking at the coil I quickly determined that I should focus on lowering R-L-C, which will take care of the amperage potential issue given its more stringent AWG requirements.
There really is no debating going on. Just me asking a question and finding the answer by having a second look at the physical characteristics of the driver.
I am not forcing anybody to waste their time, but if one wants to add to the conversation, all I ask is not to be told what and how I should have asked my questions, and to not be called foolish.
It's all good on my end
That's what I have hooked up to the Eikonas right now. One cable, one leg colors, and stripes the other.
However I think I saw the calculations in some article that, to get the full benefit of litz wiring throughout the whole audio bandwidth, one should have individual strands of 30 AWG or more.
Ethernet is 24 AWG, but it does its job for now.
However I think I saw the calculations in some article that, to get the full benefit of litz wiring throughout the whole audio bandwidth, one should have individual strands of 30 AWG or more.
Ethernet is 24 AWG, but it does its job for now.
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.