Using two different types of speaker wire

So

Are the built-in amlifiers in powered monitors supposed to have the same impedance as the speakers themselves? for example, 8 ohms?

I also just noticed that my computer monitor is buzzing from the back, I never noticed that before, could that have an effect on my monitors, I made sure they were shielded.
 
SO

Does anyone know exactly (or vaguely) how much more wire (than the lower gauges) is in the higher gauges?
For example, does the 12 gauge wire have twice as much wire in it than the 16 gauge?

Is there a less chance of you blowing your speakers if you use a lower gauge wire?
To quote knightfly, "your cones won't flop around and sound like shit" .....or something to that effect.
 
Re: SO

Me myself & him said:
Does anyone know exactly (or vaguely) how much more wire (than the lower gauges) is in the higher gauges?
For example, does the 12 gauge wire have twice as much wire in it than the 16 gauge?

Is there a less chance of you blowing your speakers if you use a lower gauge wire?
Click to expand...

I don't know the ratios involved from one gauge to another. Knightfly would know - why don't you Private Message him and let us know the result! His work schedule only lets him get on here every once in a while.

As far as blowing the speakers, with lower gauge cable your amplifier will be more efficient in powering your speakers. So if you commonly run your amps so hot that your speakers are close to blowing, you might need to turn them down a bit when you switch to fatter cables.
 
'Bout Damn Time...

Hey guys, sorry I missed this one til you started throwing rocks - Here is a short table of wire info, sticking to practical (and commonly used) sizes - First of all, conductor areas of various wire sizes are usually measured in circular mils rather than inch diameters - for example, both 18 ga. stranded and 16 ga. solid wire are listed as .002 sq inches, but the 18 is listed at 1620 circular mils, where the 16 is shown as 2580 circular mils. This is just because the circular mil dimension is more accurate so suffers from less rounding off error. Gauges are not as simple as double or triple the size. An 18 ga wire is 1620 cm (circular mils, not centimeters), a 16 ga is 2580 cm, a 14 ga is 4110 cm, a 12 ga is 6530 cm, and a 10 ga is 10,380 cm. BTW, there is no 0 gauge - The numbers decrease by 2 down to 2 gauge, then 1 - after that, there are 1/0, 2/0, 3/0 and 4/0 (larger is last) - from there, the sizes are listed in MCM, or thousand circular mils. we have 250, 300,350, 400, 500, 600, 700, 750, up to at least 2000 MCM. General idea of size - a 2000 MCM wire has a diameter of 1.632 inches, a resistance of .00643 ohms per 1000 feet, and is rated for currents of up to 750 amps depending on insulation type, temperature, etc. -

Resistance, assuming the same metallurgy (both copper) is inversely proportional to diameter in circular mils. The resistance per 1000 feet is as follows: 18 ga = 7.95 ohms, 16 ga = 4.99 ohms, 14 ga = 3.14 ohms, 12 ga = 1.98 ohms, and 10 ga = 1.24 ohms. These are figures for 1000 feet of single conductor, so a 10 foot speaker lead would have 20/1000 of that value, considering the loop resistance. This would make a 16 ga lead (10 ft ) have a resistance of .02 x 4.99, or .0998 ohms. a 12 ga lead would have a resistance of .02 x 1.98, or .0396 ohms. If a particular amplifier has a damping factor of 500 when driving an 8 ohm load, then dividing 8 by 500 would give you the source impedance of the amplifier, which in this case would be .016 ohms. This is frequency dependent, since there are reactive components in speakers and amps that won't present the same impedance at different frequencies - so these figures are generalities and in real life aren't this simple. However, this will give a good idea of the effect on damping, or speaker control, with changing wire sizes.

Let's take our hypothetical 8 ohm speaker, with our hypothetical amplifier at a damping factor of 500 at 8 ohms. From the above math we see that the amp has a source impedance of .016 ohms. If we add the resistance of 16 ga/10 foot speaker wires, we are now at .1158 ohms source impedance. If we now divide 8 ohms by .1158, we just reduced our hypothetical damping factor from 500 to a whopping 69.08 ! This means that, compared to directly hooking the speaker to the amp with zero-resistance leads, we've just cut the damping (speaker control) factor by 7.2 times. Now, let's look at the 12 gauge leads - lead resistance is now .0396 ohms, amp source impedance is unchanged at .016, so the total source impedance is now .0556 ohms. Dividing 8 ohms by .0556, we end up with a damping factor of 143.88 - almost exactly TWICE as good. (more is better)

As far as wire size versus blowing speakers/amps, it's not likely to make any difference. I suppose that if you were to run everything on the ragged edge all the time, you might be more likely to cause damage by creating a lower (worse) damping factor, but I've never seen it. Besides, in a studio situation you should be trying for the exact opposite. There are some mastering engineers that use amps capable of 1000 watts/channel, just so they NEVER come anywhere near clipping, even on the lowest most powerful notes from a 5 or 6 string bass - Clipping usually takes out tweeters first, because they are not supposed to see any DC current, only high frequency AC. When you clip the waveform into a speaker, there is a period of time during each half-cycle where there is no change in current. This is the same as applying a battery to the tweeter - the steady current through the voice coil causes it to heat up, and will eventually (or quickly) cause the coil to burn out. This is one of the main reasons for choosing a power amp that is 'way bigger than you ever think you'll need (not the only criteria by any means) - the larger amp has more headroom, bigger power supply, etc, so will not clip during any normal usage, so therefore no DC to your speakers. A larger amp is also more likely to have a larger damping factor, so will sound more punchy at lower levels due to tighter speaker control.

As far as using shielded wire for speakers, don't. Amps are designed for the reactive load of a speaker, but usually NOT for the extra capacitive reactance that a shielded cable presents. Typical shielded cable has an average capacitance per foot of 30 to 90 picofarads. Heavy speaker wire has no shield to act as a third plate of the capacitor, and the insulation is thick enough to lessen the effects of capacitance between the two conductors, so typical capacitance per foot of heavy speaker wire will run from 10 picofarads per foot up to maybe 30. Whether or not your particular amp will dislike the extra capacitance enough to cause serious problems, there will still be a loss of high frequency response with higher capacitance cables. At higher power, shielded speaker leads can cause feedback the same as if a mic was pointed at the speaker. I've seen a couple of times where even un-shielded speaker leads, when too close to mic cables, have caused feedback.

I see from a couple of your posts that my previous post on this was found, so I'll stop before I cover it all again. I hope this anwers your questions about speaker leads - generally, get the heaviest ones you can find that will fit your connections, make both channels the same length, don't run them 40 feet if 6 or 10 feet will work, and don't run them parallel to either power or mic level signals.

Littledog, I'm headed for graveyard shift again, so find whatever booboos I missed (again) , OK? Peace, Bro... Steve
 
Knightfly - thanx for your continued efforts to shine light into the esoteric. I'm making an assumption here and I'd appreciate a correction - from anyone.

Shielded (instrument) cables are not recommended for use between amp and speaker but are fine for use as input to self-powered monitors. Yea or Nay?

I have the M-Audio Omnistudio and a pair of Yamaha MSP-5 near-field monitors. I'm feeding the Yamaha's with six foot guitar cable. I assumed (there's that word again) that 'cause the Yamaha's have 1/4" input jacks guitar cable was the preferred medium.

Thanx again,

lou
 
"Man! if highschool had been more like this ....." :=) While I don't deny that I'm absolutely wonderful (my wife and I have a strange and wonderful relationship - she thinks I'm strange, I think I'm wonderful) Maybe this time around you have more motivation to pay attention in "class" , as in "now I KNOW what's in it for me" - Whatever the reason, glad I could shed some light.

Just a little more - SkyBlueLou - you're correct in your "assumption" - a powered monitor needs shielded inputs just like any other mic or line level device. In fact, using un-shielded speaker wire into a powered monitor will cause at least a little bit of hum, and usually a lot depending on how the wires are routed (what other wires they are next to)

Having only 6 foot cables to your MSP-5's, you probably won't notice much (if any) difference between generic guitar cables and high-dollar replacements. Going with special low capacitance cable may get you some slight degree of increased clarity and a tiny bit more high frequency response, but unless you've done all the other esoteric things acoustically, you'd be hard-pressed to notice, even if your ears are beyond golden, all the way to platinum. Besides, unless you're into soldering you will have a hard time finding reasonably priced improvements to 1/4" connectors. The Monster stuff is 'way overpriced, they gotta get their advertising budget SOMEWHERE :=) the only way I know to get truly low capacitance cable combined with good 1/4" connectors, is to "roll yer own" - The best way to do that (that I'm aware of) is to buy bulk "digital" cable ( the stuff that is usually sold as 110 ohm for AES cables) and buy Neutrik ( I prefer them over Switchcraft, better strain relief and easier to work with) and solder your own custom cables. Anybody needing a push-start on soldering, speak up here and I'll re-do an old post I wrote on another BBS a while back. The bulk cable normally sold as digital cable, usually has a very low capacitance, such as 8-10 picofarads/foot. This is because when you transfer a square wave signal such as digital pulses, the bandwidth needs to be at least 10 TIMES as high for any given frequency as it does for a sine wave (analog) - if not, the cable capacitance and series resistance get together and form an integrator. This basic electronic function causes leading edges of pulses to be rounded off, and if the cable is long enough or high-capacitance enough or both, the integration can even cause time-shifting of the crossover points of the digital signal. This introduces phasing and jitter into the picture, and if bad enough it can be heard by a retired jackhammer operator.

StoneB - "Are the built-in amlifiers in powered monitors supposed to have the same impedance as the speakers themselves? for example, 8 ohms?"

No, the built-in amps themselves would have a super-low source impedance the same as I explained in the previous post. The amp needs to be able to exert precise control over the speaker, and as such needs to be able to "slap it around at will", - one possible analogy would be (assuming you could actually ride one) putting a 600 horsepower motor on a bicycle. The engine is the amplifier, the bike is the speaker. In this case, the bicycle has absolutely NO CHOICE but to do what the engine wants, at least until the engine kills the bike and you. Now, consider a Volkswagen van, weighing about 2800 pounds, with a 36 horsepower engine (think 1964) Here, you may WANT to go 180 MPH up that hill, and you may ASK for that to happen by stepping on the gas pedal hard, but the power isn't there so gravity, wind, inertia, BB's in the road, all get in the way of your desire to go 180 up the hill.

"I also just noticed that my computer monitor is buzzing from the back, I never noticed that before, could that have an effect on my monitors, I made sure they were shielded." - This shouldn't have an effect on speakers, it's usually the other way around. You don't state the approximate frequency here, so there could be several possibilities as to the source. If the freq is low, you may have a resonant core in the transformer of the power supply in the monitor. Most computer monitors don't pay much attention to things that will drive audio people crazy. One of my 21" monitors needed 4 different small pieces of self-adhesive weatherstripping placed inside to stop vibrations. This only occured with sound coming out the speakers, but I wasn't able to localize it until I swept the room (no, not with a broom, anybody needing a clarification please post back) If the sound is higher freq, like almost out of hearing range, it could be the horizontal freq of the monitor; although, any more that's usually too high for dogs to hear very well, much less humans)

StoneB - "Also... would my amp's high wattage be something to consider when dealing with this issue?

What's the highest gauge wire you can get?"

High wattage in an amp is better, all other factors being equal (they rarely are =) Generally, the higher the wattage of an amp, the lower the source impedance in order to pass more current. This in turn will improve the damping factor (load impedance/source impedance) also, larger amps tend to have bigger power supplies, and so will more faithfully follow the input signal and not clip peaks on loud transients. This is also better. The quickest way to fry your speakers, other than plugging them into the power outlet to see what 60 hZ sounds like, is to try to get lots of sound out of a small amp. This causes the amp to clip, and a straight DC level into a speaker nullifies the natural current limiting tendency of an inductor (speaker coil) to resist current change. This straight DC (the flat top of the clipped part) causes the voice coil to heat up abnormally, and if allowed to continue can literally burn out the voice coil. This is common more for tweeters, since they usually have smaller gauge (bigger number) wire and less inductance than a woofer voice coil.

Once more, without being part of another explanation - Wire GAUGE is a number that is INVERSE (but not proportional) to physical wire DIAMETER. Up to sizes we wouldn't care about here, the smaller the gauge number, the larger the wire diameter. This gets confusing when people (including yours truly) refer to "higher gauge wire", really meaning "bigger". There is no fixed ratio of diameter or resistance from one gauge to the next, hence the need for wire tables in order to properly size an electrical system.

In electrician's terms, ampacity tables are used to size wires for acceptable voltage drops (losses) in various conditions. In audiophile terms, things get cloudier and individuals need to know some of the theory - but the main thing to consider here would be sound. If it sounds better, who cares why, other than to maybe take it even further. Conversely, if it sounds WORSE, we need to know some of the possibilities to explore in order to improve it. Since we are dealing with AC signals, there are several more factors than just wire resistance to consider. These would include capacitance, inductance, length, juxtaposition, contact material of connectors, sometimes characteristic impedance of cables (usuallly not with audio, more with digital and video signals) and material. A lot of people seem to think that Gold is a better conductor than copper. In actual fact, gold has only 71.2% of the conductivity of copper, where silver has 106%. Aluminum is 64.9%, nickel is at 25%, platinum is at 17.5%, lead and tin (the two elements of non-silver solder) are at 8.35% and 15%. (This is probably one of the (lesser) reasons it is recommended that a good mechanical connection be made before soldering; if the signal has to go thru solid solder, the resistance would be slightly higher) The main reason gold and silver are used on contacts and connectors is that they are less corrosion prone.

If I missed anything, feel free to remind me. If I got anytning wrong, it's Littledog's fault. :=) Remember, sticking your finger in a light socket may be a cheap perm, but it may be more "perm" than you wanted... Steve
 
Re: 'Bout Damn Time...

Steve,

Good to see another rare person providing technical explanations on this bbs.:D

I disagree with a couple of your points however.
knightfly said:
...Let's take our hypothetical 8 ohm speaker, with our hypothetical amplifier at a damping factor of 500 at 8 ohms. From the above math we see that the amp has a source impedance of .016 ohms. If we add the resistance of 16 ga/10 foot speaker wires, we are now at .1158 ohms source impedance. If we now divide 8 ohms by .1158, we just reduced our hypothetical damping factor from 500 to a whopping 69.08 ! This means that, compared to directly hooking the speaker to the amp with zero-resistance leads, we've just cut the damping (speaker control) factor by 7.2 times. Now, let's look at the 12 gauge leads - lead resistance is now .0396 ohms, amp source impedance is unchanged at .016, so the total source impedance is now .0556 ohms. Dividing 8 ohms by .0556, we end up with a damping factor of 143.88 - almost exactly TWICE as good. (more is better)
Click to expand...
Comparing damping factors is a very amplifier-centric perspective and the importance of speaker wire is vastly overrated when looking at it this way. As a speaker designer I tend to look at it another way:

The electrical Q of a speaker driver is given as

Qes = 2*Pi*fs*Re *Mms / Bl^2 = K*Re

Where Re is the voice coil resistance.

When connected to an amp via a wire this changes to

Qes' = K*Ztotal

where

Ztotal = Re + Zsource (v.c. resistance + wire resistance + amp output impedence)

Let's assume we have a 8 Ohm nominal impedence speaker, so Re is approximately 5.5 Ohms, and let's say it's nascent Qes = 0.3

so K= Qes/Re = 0.3/5.5Ohms = 0.0545 Ohms^-1

Now from this perspective lets compare your two speaker wire scenarios where the total source impedence is either 0.0556 Ohms (') or 0.1158 Ohms ('').


Qes' = 0.0545 Ohms^-1 * (5.5 Ohms + 0.0556 Ohms) = 0.3028

Qes'' = 0.0545 Ohms^-1 * (5.5 Ohms + 0.1158 Ohms) = 0.3061

So this turns out to be a 1% difference in effective electrical Q between the two wires. Now, if I could find a driver manufacturer who could supply me units with a 1% Qes tolerance AND which could maintain that tolerance over any reasonable fraction of the lifespan of the driver, I would be very happy. Drivers like this don't exist. Furthermore, the electrical parameters which govern Qes are only a part of picture. Other parameters which contribute to the Qtotal of the entire dirver/cabinet system depend on variables such as temperature and barometric pressure - variables which we cannot well control and which will tend to wash out any small advantage gained from our tight Qes. Qtotal is the lumped parameter which tells us how "in control" our speakers are. And like I've explained Qes and therefore Qtotal is only marginally influenced by normal source impedences.

In a nutshell, the differences in resistance between normal gauges of normal length speaker wire are effectively inconsequential to a loudspeaker's performance - not to mention that paying big bucks for very low output impedence (high damping factor) amplifiers is also money not well spent. Better to spend your money on a higher quality loudspeaker.
knightfly said:
...Clipping usually takes out tweeters first, because they are not supposed to see any DC current, only high frequency AC. When you clip the waveform into a speaker, there is a period of time during each half-cycle where there is no change in current. This is the same as applying a battery to the tweeter.....
Click to expand...
It's true that clipping is most often the culprit of blown tweeters, but DC current is not the failure mechanism in loudspeakers with passive crossovers. Tweeters are connected via series capacitors, so they never see DC. Tweeters are actually blown because of all the high frequency harmonics generated from the clipping which easily pass through their blocking capacitor. The clipping simply converts low frequency energy into high frequency energy which then fries the tweeters.

I see you just added another post. I'll read on.

barefoot
 
Last edited:
Barefoot - I'm also glad to hear from seriously technical people - it's obvious that you've delved much deeper into this area than I - The closest I've ever come to design (as opposed to using, lusting after, etc) was in 1967 when I was allowed to watch/listen at HH Scott when Larry Fisher and Daniel Von Reklinghousen (terrible speller of people's names, sorry) were working on speaker designs in their anechoic chamber. I would like to go back in time after learning a little, and see what my reactions would be today.

My recent dabblings were limited to two different sets of speakers in two different rooms, one a living room and one a small studio control room of mine (22 x 11.5 x 8.3 ft) - in both cases, I replaced 16 ga wire with 10 ga. of the same length, no other changes made. Playing the same CD's thru each system, I noticed quite a bit of improvement in lows and clarity overall, especially in the living room with older Pioneer home stereo speakers. These are ported model CS-88's, which have a 12", 5", 2 domes and a horn tweeter. They have separate level controls for mid and high, which I normally leave maxed out. The studio setup was with a pair of KRK K-Roks and a Yamaha P2200 power amp. I wasn't sure if I heard much (if any) change in the K-Roks, but the living room setup sounded to me like I'd just bought new speakers. After reading your math explanations on Q and damping, I'm now wondering whether I just WANTED to hear a major difference, or actually did. I think I'm going to re-create the test when I get some time, maybe using a speaker switch setup for speed of change. The downside of that is yet another variable, so I'll have to think about that part some more.

Do you design near-fields, or just speakers I wish I could afford, along with the soffited room to put them in? What is your take on things like edge diffraction due to un-radiused cabinet edges, non-soffited speakers, near-fields placed too far back on the bridge, console early reflections, etc?

But mostly, why would I have noticed so much change in sound when the only thing I touched was the wires? (Same CD, same level, same positions, same EVERYTHING, just 16 ga to 10 ga wire) Granted, It wasn't a blind test in any way, but before I changed the wires I expected no difference whatever, as I didn't see how .05 ohms or so could possibly be noticed compared to 8. Sooo, if I didn't EXPECT any difference, and your math says I shouldn't have HEARD any difference, WTF? (Highly technical term, means WTF?) Hoping you can 'splain this some more... Steve
 
knightfly said:
Do you design near-fields, or just speakers I wish I could afford, along with the soffited room to put them in?
Click to expand...
I've designed a bit of everything.
What is your take on things like edge diffraction due to un-radiused cabinet edges, non-soffited speakers, near-fields placed too far back on the bridge, console early reflections, etc?...
Click to expand...
Bad, bad, bad, bad. ;) Lot's of ground to cover there. I've addressed some of these in previous post, but I'll elaborate when I get time.
... why would I have noticed so much change in sound when the only thing I touched was the wires? (Same CD, same level, same positions, same EVERYTHING, just 16 ga to 10 ga wire) Granted, It wasn't a blind test in any way, but before I changed the wires I expected no difference whatever, as I didn't see how .05 ohms or so could possibly be noticed compared to 8. Sooo, if I didn't EXPECT any difference, and your math says I shouldn't have HEARD any difference, WTF? (Highly technical term, means WTF?) Hoping you can 'splain this some more... Steve
Click to expand...
I wasn't there with test equipment in hand, so I can't say you didn't hear a real difference. All I can say is the placebo effect is very powerful, even if you are consciously saying to yourself "I [don't] EXPECT any difference". I have no doubt that I suffer from it's effect myself. Personally I've spent a lot of time trying to teach myself what various measurable differences sound like, and I always defer to my instruments when in doubt. I'm convinced that there is nothing below the sensitivity of my metrics what has any "dramatic" effect on the sound. And if something like that does arise, I'm pretty sure it will be because I'm just not looking in the right place and need to expand my measurement tools in order to see the effect, rather than it being a case of some mysterious "micro" sonic quality.

barefoot
 
Btw, I don't mean to imply I've got it all figured out. Far from it. My ears and instruments clearly detect flaws in my and other's designs all over the place. But as we know, identifying problems and fixing them are two very different things. And audio engineering isn't easy, like many other human endeavors. Wouldn't we all live in a perfect world they were.:)

barefoot
 
Re: 'Bout Damn Time...

knightfly said:
generally, get the heaviest ones you can find that will fit your connections, make both channels the same length, don't run them 40 feet if 6 or 10 feet will work, and don't run them parallel to either power or mic level signals.
Click to expand...

? Could you please clarify that last part about running them parallel to power or mic level signals?


Oh, I'm also very interested in getting into soldering.
 
You must log in or register to reply here.
Back
Top