Forums

TUTORIAL - Example 1, Using Femm to Calculate Component Values of Simple Microstrip to use in LTspice/PSpice

Started by RobertMacy May 5, 2015
I offer to share with you a TUTORIAL but don't know how to get it into  
your hands.

The TUTORIAL is about 15kB, zipped, and describes the step-by-step process  
of using FREE femm 4.2 to analyze a simple PCB structure to calculate  
values to then use in circuit simulation, LTspice/PSpice. The sequence can  
be modified for use in analyzing really complex structures, structures  
where there are NO formulas. [I know there are many formulas and online  
calculators for 'normal' structures that can provide similar values, BUT  
they deprive you of gaining any understanding.]

The simple structure in this example is a 50 ohm microstrip implemented  
using 1 oz copper on a four layer 1 inch wide 5 inch long FR4 PCB  
operating at 10MHz. The trace is 16 mils wide, copper 1.4 mils thick, and  
dielectric separations of 9 mils each with the overall PCB thickness  
approx 32 mils. The stack up is:
Layer 1  Trace
Layer 2  GND
Layer 3  Vcc
Layer 4  Trace

An exmaple, of gaining insight:

If the microstrip uses only the top two layers, evrything is 'normal' and  
as expected. Current flows into the SIGANL trace and is RETURNed through  
the GND plane. Of course no one would do that, you would always keep the  
next layer, Vcc, and use it as a pseudo-ground plane.

BUT! Now something very interesting happens. The RETURN current splits  
between GND plane and Vcc plane, where the current in GND plane is in the  
correct direction, but slightly MORE than what you put intn SIGNAL! and  
there is a slight bit of that 'extra' current flowing in the Vcc plane,  
BUT IN THE SAME DIRECTION AS THE SIGNAL??!!

No book formula I've ever found will show you this, so there is  
'education' in getting down into details.

As I said, this is a step-by-step tutorial, with descriptions,  
explanations, and dedicated femm drawings for each step culminating in  
LTspice schematics of the final model. Hopefully, educational/useful for  
someone out there.

Regards,
Robert
On a sunny day (Tue, 05 May 2015 03:36:33 -0700) it happened RobertMacy
<robert.a.macy@gmail.com> wrote in <op.xx5rq8hc2cx0wh@ajm>:

>An exmaple, of gaining insight:
Ms. Maple?
On Tue, 05 May 2015 03:36:33 -0700, RobertMacy <robert.a.macy@gmail.com>  
wrote:

>> ...snip....
ARRRGGGG!!!! I sent the WRONG file to any of you! should have been the TUTORIAL_Example01.zip Throw away the TUTORIAL_StepByStep.zip !!! I'll go back and do again! ...sigh Regards to all, Robert
On Tue, 05 May 2015 03:36:33 -0700, RobertMacy
<robert.a.macy@gmail.com> wrote:

>I offer to share with you a TUTORIAL but don't know how to get it into >your hands. > >The TUTORIAL is about 15kB, zipped, and describes the step-by-step process >of using FREE femm 4.2 to analyze a simple PCB structure to calculate >values to then use in circuit simulation, LTspice/PSpice. The sequence can >be modified for use in analyzing really complex structures, structures >where there are NO formulas. [I know there are many formulas and online >calculators for 'normal' structures that can provide similar values, BUT >they deprive you of gaining any understanding.] > >The simple structure in this example is a 50 ohm microstrip implemented >using 1 oz copper on a four layer 1 inch wide 5 inch long FR4 PCB >operating at 10MHz. The trace is 16 mils wide, copper 1.4 mils thick, and >dielectric separations of 9 mils each with the overall PCB thickness >approx 32 mils. The stack up is: >Layer 1 Trace >Layer 2 GND >Layer 3 Vcc >Layer 4 Trace > >An exmaple, of gaining insight: > >If the microstrip uses only the top two layers, evrything is 'normal' and >as expected. Current flows into the SIGANL trace and is RETURNed through >the GND plane. Of course no one would do that, you would always keep the >next layer, Vcc, and use it as a pseudo-ground plane. > >BUT! Now something very interesting happens. The RETURN current splits >between GND plane and Vcc plane, where the current in GND plane is in the >correct direction, but slightly MORE than what you put intn SIGNAL! and >there is a slight bit of that 'extra' current flowing in the Vcc plane, >BUT IN THE SAME DIRECTION AS THE SIGNAL??!! > >No book formula I've ever found will show you this, so there is >'education' in getting down into details. > >As I said, this is a step-by-step tutorial, with descriptions, >explanations, and dedicated femm drawings for each step culminating in >LTspice schematics of the final model. Hopefully, educational/useful for >someone out there. > >Regards, >Robert
Now posted on the Simulations Tools & Macros Page of my website. ...Jim Thompson -- | James E.Thompson | mens | | Analog Innovations | et | | Analog/Mixed-Signal ASIC's and Discrete Systems | manus | | San Tan Valley, AZ 85142 Skype: Contacts Only | | | Voice:(480)460-2350 Fax: Available upon request | Brass Rat | | E-mail Icon at http://www.analog-innovations.com | 1962 | Statistics means never having to say you are certain