Filters: Learn How Inductive Capacitive Work
Filters: Learn How Inductive Capacitive Work
Filters: Learn How Inductive Capacitive Work
RAY MARSTON
a
FREOUEN( -s
series -resonant filter, and Fig. value of R) will be 9 kilohms at
1 -b shows its simplified equiv- that frequency. Many other sim- 100
alent circuit. The R represents ilar examples can be given.
the resistance of the coil. The key point to be made here
is that at resonant frequency,f., V rtII
Series-resonant filter the reactances of C and L will be VN
50
The fundamental response of equal (but 90° out of phase),
the series filter is that capacitive and the filter input impedance
reactance C decreases with in- will equal the value of R, as indi-
creased frequency, while induc- cated by the dotted line at the o
tive reactance decreases. The bottom of the impedance vs. fre- FREQUENCY --b-
n
inverse relationship also holds. quency characteristic curve Fig.
The filter's input impedance is FIG. 2 -LC SERIES -RESONANT FILTER:
2 -a. For example, if this occurs Plot of input impedance vs. frequency at
equal to the difference between when the reactances of C and L resonance, a, plot of voltage output vs.
these two reactances, plus the are both 1000 ohms, and R frequency at resonance taken across L
value of resistor R. equals 10 ohms, the input im- or C, b. 69
www.americanradiohistory.com
www.americanradiohistory.com
voltage generated across C and Notice in Fig. 2 -b that the in- +v
L is 100 times greater than the ductive and capacitive voltages 11
actual input signal voltage, as are 90° out of phase, and the -1 10:1
shown in Fig. 2 -b, the curve of voltage generated across the se-
voltage vs. frequency. This volt- ries LC combination is effec- á
age magnification, indicated by tively zero. The impedence of
the sharp peak, is known as the the filter at f- is known as the OUTPUT
z2/R
a b matic for a parallel- resonant fil-
ter, and Fig. 4 -b shows its
equivalent circuit. The induc-
C7
CX0.9
OUTPUT
-o
tor's resistance is represented c
T
I
c
FREQUENCY - creasing frequency. The re-
ciprocal relationship also holds.
Each component draws a sig-
nal current that is proportional
INPUT
FIG. 4 -LC PARALLEL-RESONANT fil- to its reactance, but the two cur-
FIG. 5 -LC TUNED AMPLIFIERS with
ters: simple schematic, a, equivalent cir- rents are 90° out -of- phase, so low-impedance outputs: transformer
cuit, b. and plot of input impedance vs. the total signal current is equal coupling, a, auto -transformer coupling,
frequency, c. to the difference between the L b, and capacitive-divider coupling, c.
www.americanradiohistory.com
www.americanradiohistory.com
+6 TO + 18V come this drawback are illus- LC oscillators
111
47k
trated in Fig. 5. Figures 6 through 10 illus-
One way to obtain output trate the different schemes for
coupling is to consider the pri- using a parallel- resonant filter
L1
100µH í Cl
40 TO
100pF
mary winding of an RF trans-
former as the filter's inductive
component, and to take the out-
as the tuning element in tran-
sistorized LC oscillators. The
simplest of the LC oscillators is
put from the transformer's sec- the tuned -collector feedback
ondary, as shown in Fig. 5 -a. form shown in Fig. 6.
01
2N3904
This approach provides a fully 'IYansistor Q1 is connected as
OUTPUT floating output. If the trans- a common -emitter amplifier, L1
R2 C3 R3 C2
former has a 10:1 turns ratio, and C 1 form the tuned collector
22K .047 1 O 047 the output signal will have an filter, and L2 provides the collec-
_J
attenuation factor a of 10. tor -to-base feedback. Inductor
In a second method, the coil L2 is inductively coupled to L1,
FIG. 6-TUNED -COLLECTOR feedback can be tapped as shown in Fig. providing transformer action.
LC oscillator. 5 -b, to obtain an output by au- By adjusting the phase of this
totransformer action. In the feedback signal, the circuit will
third method, as shown in Fig. give zero phase shift at the
+9V 5 -c, the required tuning capaci- tuned frequency so that, if the
loop gain (determined by Ti's
turns ratio) is greater than uni-
+9V ty, the circuit oscillates. With
R1 the component values shown,
4.7K oscillation frequency can be var-
r Cl
ied from 1 MHz to 2 MHz by
L1 .01
R2
trimmer capacitor Cl.
2 5mH
C2 470K
1
--t
01 L3
provides high gain at its reso- the larger capacitor by capaci- '2N3904
nant frequency and lower gain tive divider action. C2 _iL
above and below that frequency. In these schematics each cir- 0.01'r
The drawback to these cir- cuit has arbitrarily been given
cuits is the problem of gaining an attenuation factor a of 10. R2
1 oH
access to the circuit's output Each has an output impedance
signals without loading the of Zc/a2. Thus, if Ze equals 100
tuned circuit and lowering its kilohms and a equals 10, the Z FIG. 10- SIMPLE REINARTZ LC os-
effective Q. Three ways to over- output equals 1 kilohm. cillator. 71
www.americanradiohistory.com
www.americanradiohistory.com
o
HIGH -Z
OUTPUT
Gouriet oscillator offers excel-
lent frequency stability. With
the component values shown in
--- zo
oscillator. Another capacitor, C is low, so the circuit offers FIG. 15- HIGH -PASS LC FILTERS: T -sec-
C3, with a value that is small high attenuation. tion schematic, a, and -r- section, b.
relative to Cl and C2, is put in Consequently, the circuit acts
series with L1. This circuit's res- like a low-pass filter. It is called it circuit is actually a series-reso-
onant frequency is determined a "false" filter because the cir- nant filter (like Fig. 1) with its
principally by the values of L1 cuit will only function correctly output taken from across ca-
and C3 and it is almost inde- if it is driven from a source im- pacitor C.
pendent of variations in tran- pedance equal to Z0. (This is If the circuit is driven from a
72 sistor capacitance. The Clapp/ not shown in the diagram.) The continued on page 89
www.americanradiohistory.com
www.americanradiohistory.com
ace a required sound -pressure all play a part in squeezing the max- but informative, power-handling ca-
vel from a given loudspeaker, it imum power rating from any given pability specification, such as that
ips the positive and /or negative speaker. used by Allison Acoustics: "At least
laaks of the musical waveforms. The test signal used to derive a 15 watts continuous at any frequen-
hort- duration overloads may not system's power rating must be cy. Over most of the frequency
3 audible; longer overloads are fre- chosen carefully. The ubiquitous range, at least 350 watts for 0.1
uently perceived as level compres- pink -noise signal used in so many second, 125 watts for second, 60
1
ion, rather than distortion. How - other audio tests is totally inap- watts for 10 seconds."
ver, badly overdriven amplifiers propriate for power testing be- Note the distinctions Allison
roduce a raspy distortion, not un- cause, unlike music, it has equal makes between continuous and
ce that of a mistracking phono car- energy per octave. In contrast with transient wattage levels. The dif-
idge. Low bass passages are likely the midrange energy hump dis- ference between them is what al-
take on a "mushy" quality be- played by most music, pink noise lows you to play very loud music
ause of the spurious harmonics shows up on a real -time analyzer as without problems, even though a
enerated by the overload. And as a virtually straight line, which is a continuous sinewave at the same
iscussed earlier, prolonged opera - poor representation of music. peak level would certainly damage
on with hard clipping is a frequent The single rms ratings used by your audio equipment. In other
ause of driver damage, so clipping some manufacturers imply the use words, your 100 -watt (or even 200 -
hould be avoided. of a sinewave test signal, which, watt) amplifier is certainly safe to
again, is totally unlike a musical use with typical speakers rated at
'roper power ratings waveform in shape or energy con- 50 watts maximum so long as you
Arriving at a speaker system's tent. The most valid and informative don't feed continuous tones or pink
lower rating is no easy task, even way for a manufacturer to specify a noise to them, drive the amplifier
Dr its designer. Ideally, a manufac- speaker's power-handling capability into hard clipping, drop a tone arm,
urer designs for the highest power is tostate, however loosely, the or lose a cable ground at high vol-
;apability that can be achieved with - power it can handle in a specific ume. In short, you have to abuse
i his cost and size constraints for a frequency range for a specific your speakers (and your ears) be-
liven model. Special high- tempera- amount of time. fore disaster is likely to strike. If you
ure materials such as voice -coil Listing specifications that way don't ask for trouble, it probably
vire, voice -coil forms, and cements gives rise to a somewhat complex, won't happen. it
www.americanradiohistory.com
www.americanradiohistory.com