AUDIOCLINIC (Q and A) Oct. 1985

Cable Length and Signal Degradation

Q. In connecting components together, what effect does cable length have on losses or gains? Is it true that the longer the cable, the greater the signal loss?

Is it also true that certain components should not be placed in close proximity to each other?

-Allan Dorfman; Morristown, N.J.

A. Much of today's equipment has very low output impedance, 100 ohms or less. Input impedances of the equipment that accepts signals from these devices will be considerably higher, but the effects of these high impedances are shunted by the low impedance of the driving devices.

All cables have some resistance, and therefore cause some signal loss. But to cause even 6 dB of signal loss, the cable's resistance must equal the component's output impedance. A typical cable, with a resistance of less than 0.5 ohm per foot (including the effects of both its inner and outer conductors), would have to be more than 200 feet long to cause such a loss. Even a 6-dB loss is unimportant if the driving device has the necessary signal level or the gain of the driven device is adjustable.

If this were the only kind of signal degradation associated with cable length, we would never have to consider the problem. Unfortunately, however, there is something known as "cable capacitance," usually described in terms of pF per foot. Although capacitance is not measured in ohms, a property known as "capacitive reactance" is so measured.

Capacitive reactance is not static, as is true of d.c. resistance or capacitance; it varies inversely with frequency. Thus, at high frequencies, it is easier for the desired signal to flow between the inner cable conductor and its shield than it is to flow into the circuit which it is supposed to feed. To make all this clearer, as the length of cable increases, the capacitive reactance at any given frequency decreases. When the length of the cable is such that the capacitive reactance equals the output impedance of the driver, losses at that frequency will be 6 dB.

The lower the capacitance per foot, the higher the capacitive reactance per foot. Hence one can run a longer cable if its capacitance per foot is low. To me, frequency losses at 20 kHz of more than 1 dB are unacceptable.

If physical arrangements permit, all equipment should be placed close together. Where compromises must be made, avoid locating the phonograph far from its preamplifier. One reason for this is to reduce hum pickup by the interconnecting cables. Also, cable capacitance again can become a factor. Many cartridges require some specific amount of capacitance for proper frequency response. If a cable is short, more capacitance can be added, but if a cable is too long, capacitance cannot be subtracted.

Distances between the preamp and the power amp, tuner, recorder, and CD player are governed by considerations we have already mentioned.

Turntables and loudspeakers should be separated in order to prevent what is known as "acoustic feedback." If such feedback becomes sufficiently intense, damage to both loudspeakers and output stages is possible.

Turntables should be spaced at least 2 feet from sources of strong magnetic fields, such as amplifier power transformers, fan motors, etc. Similar rules govern the location of cassette recorders and VCRs.

Speakers, of course, should be placed far enough apart for proper stereo separation. The exact distance depends on the speaker design and the room's size and acoustics.

"Audible Sidebands"

Update I am writing this regarding the "Audioclinic" item on "Audible Sidebands" in the March 1985 issue. In response to Russel E. Worthy of Massachusetts, you erred in your explanation of the audible information present in his scenario of an AM receiver which is tuned to 1,500 kHz.

Mr. Worthy describes a "perfect" receiver with a "square-topped" i.f. filter stage, tuned to an unmodulated carrier at 1,500 kHz. The 9-kHz i.f. bandwidth means that the i.f. will pass what it "sees" between 1,495.5 and 1,504.5 kHz. Because we're talking ideal, the i.f. wipes out anything outside this range. Now, there also exists a modulated subcarrier at 1,490 kHz whose sidebands would be at 1,483 and 1,497 kHz because of a 7-kHz sine wave signal modulating this carrier.

With the receiver tuned to 1,500 kHz and the signal mentioned, the i.f. will pass 1,497 and 1,500 kHz to the detector, resulting in a 3-kHz tone being heard. There will be no 7-kHz tone.

If the 1,500-kHz signal were to stop transmitting, then only the 1,497-kHz sideband would be passed by the i.f. and there would be no audible result from detecting this single frequency.

If we were to depart from Worthy's ideal situation and assume our i.f. was wider than 9 kHz, then the same signal inputs would give the 3- and 7-kHz tones, as you stated, but also would result in a 10-kHz audio tone and even a 17-kHz tone. Without the 1,500-kHz signal we would get the 7 kHz, poorly tuned.

-Ross M. Jory, Portland, Ore.

I agree with Mr. Jory. I managed somehow not to take note of the "ideal" response of Mr. Worthy's system.

Guitar-Amp Treble Adjustment

Q. I have installed a treble-cut circuit in my guitar. It consists of a double-throw, center-off switch, with capacitors of different value on each side of the switch, so that one switching direction produces more treble cut than the other. I am happy with the results, except that, when the switch is thrown to either treble-cut position, a loud "click" or "pop" is heard in the loudspeaker. Is there a way to stop these sounds? I have tried polarized capacitors, as well as reversing the setup so that the capacitors are fixed to ground rather than to "hot," with switching made to ground.

-Ron Kalstein; Philadelphia, Pa.

A. Without knowing more about the amplifier you are using, I can't say for sure that you can remove the clicks that occur when the switch is thrown to add treble cut. I have one possible scheme which might help.

If there is a small d.c. voltage at the input terminal of the amplifier, this will tend to charge up the capacitors as they are switched into the circuit. This being so, the cure is to remove the d.c.

from your guitar and treble-cut switching system. Add a coupling capacitor of suitable value between your guitar's output and the amplifier input. This should not be an electrolytic capacitor, because such capacitors can leak.

If the clicks are still present, I suggest redoing the circuit so that it employs a potentiometer, placed in series with a capacitor which gives you the greatest amount of treble cut you expect to use. To make adjustment of the pot easier when playing your guitar, and more like the switch to which you have become accustomed, place a pointer knob on the pot's shaft. You should be able to push the pot with a finger, rather than turning it. It will also give you a range of tone color, rather than just two cuts and one flat.

The value of the pot can be so chosen that its operation is crowded toward one end. Thus, you will be able to push the pointer over a short distance and create the tonal changes you need.

Unusual Equalization Problem

Q. Please tell me how to design a tone control for the middle of the audio spectrum. I would like to add this to my preamplifier, which I am using with an electronic musical instrument. Since it is a music producer and not a reproducer, I am not interested in fidelity or in flat frequency response.

-Name withheld

A. Many reference books and "applications notes" produced by the makers of semiconductors contain circuits of this kind. Rather than breaking into the amplifier, you could design the circuit to work between the input of the preamplifier and the output of your musical instrument, or between the output of the preamplifier and the devices which follow it.

Remembering, however, that "work" is a four-letter word, I suggest you consider using a graphic equalizer. There are many such units designed to be placed between the output of an instrument and the input of its amplifier.

If you need a really sharp boost over a narrow range of frequencies, I suggest you do what I did when faced with a similar problem. I needed an oboe effect, which was unavailable on the small organ I had. I used a "wah-wah" pedal between the organ's output and the input to the rest of the system. I peaked the wah-wah to a range of frequencies which added the right nasality to the organ stop I was using, and the result was a very passable oboe.

Note that in this application I did not use the wah-wah pedal for its intended purpose. I set it to the desired band of frequencies to be boosted and left it that way for the duration of my recording project. If the peaked frequency region is too sharp, and if the wah-wah has an intensity adjustment, use it to adjust for proper sound quality for your applications. If no such control exists, and if the device has a gain of less than unity, bridge its input/output with a variable resistor (try 100 kilohms). At its maximum value, the resistance will not affect the action of the pedal. As its value is reduced, the "wah" will be less effective. If the overall amplification is greater than unity, this bridging resistor may well cause oscillation, producing a severe howl.

I did have one problem with a sharply tuned wah-wah. This device will emphasize one frequency more than all others. If this frequency corresponds to a note being played, this tone will stand out. If possible, set the wah-wah frequency so it does not directly fall on a note which will be played-tune it "in the cracks," so to speak.

Amplifier Power and Loudspeaker Voice-Coil Damage

Q. I note with some concern that many amplifiers are capable of delivering high output current. One such amplifier claims power "high enough to weld with." Won't this current damage the fine wire of a loudspeaker?

-Eugene Bershad, Freehold, N.J.

A. If the load impedance "seen" by an amplifier is low, most of the power developed in the load will be in the form of current rather than voltage (though, of course, power is a combination of voltage and current). A loudspeaker will only draw current dictated by its impedance. This is the same situation that you would find when considering your home electrical wiring. It may be able to deliver 60 amperes of current, but this does not mean that a device plugged into a wall outlet will draw 60 amperes. A 100-watt lamp will draw about 1 ampere, even though the wiring is capable of delivering much more.

(Source: Audio magazine, Oct. 1985, JOSEPH GIOVANELLI)

= = = =