Errata for Chapter 3 of EMRFD

(31 July 09)

p3.4, middle col, 2.4 inches below top of text.  Remove the paren ) at the end of the sentence. (15May03)

p3.6, Table 3.2.   g(3) for the Wide U. Sp. filter should be 3.352 instead of 2.352.  (16 Sept 05)                

p3.7,  col 3, 2.9 inches up from text bottom.   Replace "texts.(ref 6)" with "texts such as Zverev.(ref 6)"   (15Jan04)    

p3.9, col 3,  1.9 inches up from text top.   Add a footnote after "Zverev's text."  The note should read "see ref 6.      (15Jan04)

p3.10, middle col, 4th line down from top:  Change "he" to "the" (24Feb03)

p3.10, left column,  4.2 inches up from text bottom:    Note the first few sentences in the paragraph beginning with "The sidebar.."  Just past the bold Table 3A that appears in text we find "(see sidebar on page 3.14)"     Move this expression to the beginning of the paragraph so that we now have "The sidebar equations (see sidebar on page 3.14) may be used ......This has been done done to form Table 3A .    The inductors ......"  (9Dec03)

p3.11, Fig 3.22 caption.   Change 0.4 mH to 0.4 uH.   (15Jan04)      

p3.12, sidebar Fig B.   It's "scraps" of circuit board material rather than scaps.   (16May03)

p3.14, Table 3A.  The inductors are all in uH (microhenry) rather than mH (millihenry.) (24Feb03)

p3.15, middle col, 1.3 inches down from top of text:  change "coil" like structure  to become "coil-like" structure.  (24Feb03)

p3.19, right column, 4 lines up from bottom:  Replace the existing text with "Normalized Q is q0=19.2, so..." (4May03)

p3.19, right column, 1.0 inch up from text bottom.   Replace k23=0.4512  with  k23=0.5412.    The filter presented in the example was designed with k23=0.4512 and is a nicely performing filter with a slightly narrower bandwidth than the desired one.   (4 Feb 05)           

p3.19, Fig 3.35.  This is not a report of an error, but rather, a revision to an existing method.    Jack Smith, K8ZOA, pointed out a very interesting omission that we made in the formula we use to calculate motional inductance from the frequency shift in the G3UUR type crystal oscillator.   The capacitance we had used was just the series C that is inserted by the switch, including the capacitance of the open switch.   This value should also include the crystal parallel capacitance, or holder C as it is sometimes described.   The figure below summarizes this:
 
The parameter C0 is the parallel capacitance of the crystal.    This is measured at low frequency with a device such as the AADE LC Meter.  This change is not a error that needed fixing.   However, I would recommend that anyone reading this mark it in their copy and use the refined equation.   It is typically a 5 to 10% kind of refinement.   (22may06)  (15Nov07)  

p3.21,  middle col, 0.5 inch down from text top.   Replace "shape is severely distorted" with " attenuation is markedly increased". (27July03)

p3.23, Fig. 3.44(C)  The vertical axis should not be dB, but ms for milliseconds.     [That needs to be lower case ms and not mS, which would be milli-Siemens, a unit of conductance and not time.   (corrected, 15Nov07)]  (5Mar03)   (15Nov07) 

p3.23, middle col, 3.6 inches down from top:  replace "termina-tion" with "termination" (24Feb03)

p3.23 and 3.24.  On these pages we describe the behavior of a "linear phase" crystal filter.   The use of the term linear phase is just an approximation here.  This filter was designed using the "linear phase with equal ripple error" k and q tables from Zverev.  These tables are an approximation, as is often done in network synthesis.  (Indeed, the Chebyshev filter is an approximation with ripples emulating a perfect brick wall filter.)  Also, the base filter is a low pass design.  Here, it undergoes a lowpass to narrow bandpass transformation when a practical filter is designed.   The approximations as well as the transformations lead to departure from the ideal of a perfectly flat group delay.   (This is more of a comment than an actual errata item.)  (29Apr03)

p3.24. Figures 3.46 and 3.47:  Exchange the two figures while leaving the captions in place.  That is, the Chebyshev filter is the one with a flat passband response and severe "wings" on the group delay. (24Feb03)

p3.24.  Fig. 3.48 should be the following:

  Fig 3.48 (same caption) (24Feb03)


p3.24 left column, 0.6 inches up from text bottom.   Replace the text beginning with "Fig 3.48" and ending in "5, and 10."   with  "Fig 3.48 is a special version with two equal resistors and a capacitor that is A times the value of the first.   Fig 3.49 shows representative responses for R=10K, C=.01 uF, and A=1, 2, 5, and 10. " (10Apr04)  

p3.24, middle column, text up 3.5 inches from text bottom.  Replace sentence beginning with "For A<=2 and"  with  "For A=2 and for equal R, the 3 dB cutoff frequency is given by"  (10Apr04)

p3.24, middle column.   Replace Eq 3.18 with
    (10Apr04)                  

p3.24, middle column, text following Eq 3.18.   Eliminate from "where A is the capacitor ratio, C2/C1"  to end of paragraph.   Replace with the following:
"For example, with R=10K-Ohm, C1=.01 uF, C2=.02 uF, the cutoff is 1125 Hz.  Eq 3.18 can be solved for R for any cutoff or capacitor, but is restricted to A=2, which is the Butterworth filter.  (10Apr04)                  


p3.25, Fig 3.49.   Add a sentence to the caption:  "The solid line corresponds to A=1 while highest peak is for A=10." (27July03)

p3.26, Fig 3.53.  Add a sentence to the caption:  "The solid line corresponds to A=1 while highest peak is for A=10." (27July03)

p3.26, left column, line 2 down from text top.   Replace "ungrounded" with "grounded"     (10Apr04)  

p3.26, left column, line 3 down from text top.   Replace "grounded" with "ungrounded"     (10Apr04)

p3.26, left column, 0.3 inch up from text bottom.  Replace "pure high" with "pure N=2 Butterworth"    (10Apr04)

p3.26, middle column, replace Eq 3.23 with
      (10Apr04)

p3.26,  add text just below Eq 3.23.   "This is identical to Eq 3.18."  (10Apr04)

p3.27, left col, 2nd line below equation 3.26.  Text should refer to Equation 3.25.  (5Mar03)   

p3.28, left col. 1.3 inches down from top of text:   "circ" should be "circuit" (24Feb03)   

p3.28, left col. 4.2 inches down from top of text:   "generated" should be "generate" (24Feb03)    

p3.28.  Left column, 1.6 inches down from text top.  Says "all all".   Should read at all frequencies.  (5Mar03, 10Apr09)  

p3.30.  Left column, 1.3 inches up from text bottom.  "It's" should be "its".  (31July09)  

p3.31, Fig 3.67.    The capacitors should be switched in the figure.   The new figure is
   (11Feb08)

p3.33, left column, 3.4 inches up from text bottom:   Note the Greek mu followed by i.   The "i" should be a subscript, indicating that this is an "initial" permeability.  (9Dec03)  

p3.34, middle column, 1.1 inches up from text bottom:  replace "later is -63" with "latter is -63"    (9Dec03)  

p3.34, middle col.  2.7 and 3.5 inches down from top of text.  Replaced "powder" with "powdered" (24Feb03)  

p3.36, left column, 0.7 inches up from text bottom:  replace "while Hybrid refers" to "while hybrid refers"   (9Dec03)  

p3.36, left col, 2.9 inches down from top of text:  add a hyphen to make text read "transformer-like" (24Feb03)  

p3.38, Ref 4.  The title for the paper is "...Using Ultraspherical Polynomials."  (27July03)

p3.39, Ref 18.  The complete reference is :  B. Carver, K6OLG, "High-Performance Crystal Filter Design," Communications Quarterly, Winter, 1993.    (27July03)