You ask - we answer...

 a nice swoosh

If YOU have an RF/Microwave question, a circuit design technique to share, or a question about our software, please ask us by e-mail. Don't be shy! We'll do our best to either answer directly or point you in the right direction! If we can't answer, we're quite sure one of our site visitors will be able to. Let's shareour knowledge and learn new things.

Also, if you have a design tip that will help other engineers, please e-mail us. We will post it HERE, along with your name and e-mail address.



Question:     "Bondwire inductance calculation?"

As a young engineer, when I asked this question of an "old timer," I was told: "Use 20 nanohenrys per inch." This is what I did until I began to design MIC's (microwave integrated circuits). I then found the need to be able to calculate bondwire inductance much more accurately!

As I later learned, the figure I was given was an approximation for AWG #18 buss wire. If you look at the sample bondwire calculation in our Utilities+ program, you'll see that the inductance of a 1 mil bond wire, 125 mils long, 50 mils above ground, operating at 10 GHz is about2.954 nH. The "20 nH/inch" approximation I once used would have given 2.5 nH, a figure about 29% lower. A hybrid MIC is quite expensive to prototype. You can easily see by this that if you're designing one, and (maybe), it contains dozens of bondwires, you're "first cut" will likely have to be redone several times before it performs close to what theory and circuit design software would predict.

Please see Utilities+ Ad or our on-line User Manual:  manuals/utilpg7.html
======================

 Question:     "What is an L-Match Network?"

It is a two-element impedance matching network having an inductor and a capacitor, an "L & a C." One component is in series and the other in parallel, hence the name "L-Match."

Please see our Utilities+ Ad or our on-line User Manual:  manuals/utilpg10.html
======================

Question:   "What is .s2p format?"

As you know, s-parameters are complex numbers having both magnitude and angle. Back when this parameter set was new, there was a need for a standard format, so that engineers, software developers, test equipment suppliers, et al, would all "be on the same page." S2P format is where the operating frequency, F, and the 4 s-parameters are given in the following order in a data file: F, s11M, s11P, s21M, s21P, s12M, s12P, s22M, s22P with "P" (phase) always in degrees, and "F" (frequency) usually in MHz, but sometimes, depending on vendor, in GHz.

For a description of our s-parameter enhancement and calculations software, please see our SData+ Ad.
======================

Question:   "Where can I find a Form ZY-01-N Smith Chart?"

You're in luck!  :-) Click on the following like to see one in full color: Form ZY-01-N. Careful tho - if you have a dial-up connection. The file size is about 2.85 MB. For those who are unaware of this type of Smith Chart, its actually called an "Immittance Chart" as it overlays an impedance chart atop an admittance chart, each in a differentcolor. This is, in our opinion, the most USEFUL  rf engineering design chart EVER devised! You can plot a load impedance on it, followed by the conjugate of the source impedance, and then, literally, "connect the dots," by traveling clockwise, or counter-clockwise, on circles of either constant resistance or constant admittance, and easily devise a first-cut impedance matching network.

Our SmithMatch program can really simplify (make easy) the design of impedance matching networks. Read thru  the tutorial style on-line SmithMatch User Manual. We're sure you'll find it very interesting!
======================

Question:   "A 'how to primer' on directional coupler design?"

Simple equations for the design of a directional coupler are given in many text books; we've listed two below for your convenience. The equations let you input the degree of coupling (dB). This is the EASYpart.  :-) Now, just "turn the crank," and output the normalized coupled-line even and odd mode impedances, Zoe/Zo and Zoo/Zo.

Once you have these two normalized impedances, simply use our MStrip+ program to calculate the physical line dimensions, i.e., the width and spacing of the coupled lines for the type board material you are using.

Using this program, you can easily synthesize lines in either microstrip or stripline for a coupled-line bandpass filter too!

Reference #1: Microwave filters, impedance-matching networks, and coupling structures, by Matthaei, Young, and Jones. McGraw-Hill Book Company, 1964. Library of Congress Catalog Card Number 64-7937.

Reference #2: Stripline Circuit Design by Harlan Howe, Jr. Artech House, Inc. 1974. Dedham, MA. Library of Congress Catalog Card Number 73-81242.
======================

Question:   "How can I reduce insertion loss in a microstrip or stripline filter"

First, pick the vendor and the dielectric material you use with regard to the frequency range over which the filter will be used. The LAST thing you want to do is to use "FR-4" at 4-8 GHz! The losses would be measured in dBs' rather than in tenths of dBs'. Cheap FR-4 junk works fine for power supplies but little else.  :-)  Next, having chosen a good quality low-loss material from the vendors spec sheet, if you want the lowest insertion loss, specify you want "rolled copper" instead of electro-deposited copper. The surface roughness of rolled copper is a lot less than for the electro-deposited type, and will, in many cases, reduce insertion loss by as much as 50%. Naturally, rolled copper costs more, but, its rare that you get something for nothing! Also, if possible, (based on the level of signal passing thru the filter), use 1/4 oz. or 1/2 oz. copper instead of the usual 1 to 2 oz. copper. As a suggestion, I've had excellent results using 30 mil Rogers RT/duroid dielectric material with 1/4 oz. rolled copper plating for stripline side-coupled bandpass filters in the 1 to 8 GHz range.

Home Page  |   Site Map

Microwave Software
San Clemente, CA  USA

support@microwavesoftware.com

Copyright 1984-2009 Microwave Software
All Rights Reserved Worldwide