To interpret the accuracy... Because the free 2D solver
(http://ataitec.com/free_2d_solver/) is based on BEM, another true BEM solver
may claim to be more accurate if it gives slightly less (<<1%) impedance. A
true FEM solver should not give higher impedance than the above free 2D solver.
Ching-Chao
-----Original Message-----
From: si-list-bounce@xxxxxxxxxxxxx [mailto:si-list-bounce@xxxxxxxxxxxxx] On ;
Behalf Of Dennis Han
Sent: Monday, May 14, 2018 1:46 PM
To: Ching-Chao Huang; si-list@xxxxxxxxxxxxx
Subject: [SI-LIST] Re: 2D Field Solver Comparisons
The BEM (boundary-element method, very similar to method of moments,
MoM) solvers are HyperLynx, Free 2D Field Solver, Simbeor SFS, Si9000,
TNT, mdtlc, atlc, atlc2, and Trace Analyzer. The FEM (finite-element
method) solver is TNT. TNT claims to use both methods.
I removed the tabs from the impedance table below to make it easier to
read since SI-LIST converted the tab character to junk.
Dennis
On 5/14/2018 2:49 PM, Ching-Chao Huang wrote:
A BEM solver provides an upper bound and an FEM solver provides a lower bound
to the true impedance. It should be easy to converge to less than 1% or 0.5%
error.
Regards,
Ching-Chao Huang
-----Original Message-----
From: si-list-bounce@xxxxxxxxxxxxx [mailto:si-list-bounce@xxxxxxxxxxxxx] On ;
Behalf Of Dennis Han
Sent: Monday, May 14, 2018 10:37 AM
To: si-list@xxxxxxxxxxxxx
Subject: [SI-LIST] 2D Field Solver Comparisons
This topic came up again and it might be worth seeing some of my
comparisons of the free 2D field solvers. As a contractor, I sometimes
have access to field solvers that are not free and sometimes the free
solvers are all I have; that's why I like to keep my comparisons up to date.
I'm using HyperLynx as the reference. Simbeor SFS will still calculate
Zdiff after the trial license has expired. I have a trial license for
Si9000 right now. All three use conductor roughness and the others don't.
mdtlc uses atlc has its solver (don't know the version though) and I
exported the stackup from mdtlc for use with atlc and atlc2. It's
interesting that mdtlc and atlc don't give the same results, so mdtlc
must use an older version of atlc.
Conductor thicknesses are from IPC-6012B, tables 3-7 and 3-8, class 3,
finished averages, which includes plating for the microstrip.
Comments welcomed.
Dennis
Physical definitions (all units micron):
Conductivity defined at 55°C (50022962 S/m), except for mdtlc and atlc,
which fix the value at 20°C.
All impedance calculations at 5 GHz if a frequency is required.
Off-center Edge-coupled Stripline
bottom width = 101, top width = 88.3, spacing = 149, thickness = 13.399,
no coplanar ground fill
core thickness = 125, er = 4.1, loss tangent = 0.019, f = 5 GHz
prepreg thickness = 160, er = 3.9, loss tangent = 0.019, f = 5 GHz
Solver Zdiff Zcm
HyperLynx VX2.3_1 100.1 29.8
Free 2D Field Solver 1.0 100.0 29.4
TNT 1.2.2 100.9 30.0
mdtlc 0.2.1.92 102.7 30.3
atlc 4.4.4 102.4 30.3
atlc2 1.02 102.0 not calculated
Trace Analyzer 2.0 100.1 27.7
Simbeor SFS 2017.02 100.8 not calculated
Si9000 18.03 100.6 29.6
Edge-coupled Microstrip
bottom width = 180, top width = 154.6, spacing = 150, thickness =
39.408, no coplanar ground fill
prepreg thickness = 260, er = 4.0, loss tangent = 0.019, f = 5 GHz
solder mask thickness = 35, er = 3.5, loss tangent = 0.032, f = 1 GHz
Note: **some solvers model solder mask as a dielectric layer and not a
conformal coating (use thickness 35 + 39.408 = 74.408)
Solver Zdiff Zcm
HyperLynx VX2.3_1 106.3 49.2
Free 2D Field Solver 1.0 ** 103.9 52.8
TNT 1.2.2 ** 102.8 48.6
mdtlc 0.2.1.92 107.8 47.1
atlc 4.4.4 105.7 46.8
atlc2 1.02 107.9 not calculated
Trace Analyzer 2.0 ** 103.0 48.6
Simbeor SFS 2017.02 105.2 not calculated
Si9000 18.03 107.3 49.4
------------------------------------------------------------------
To unsubscribe from si-list:
si-list-request@xxxxxxxxxxxxx with 'unsubscribe' in the Subject field
or to administer your membership from a web page, go to:
//www.freelists.org/webpage/si-list
For help:
si-list-request@xxxxxxxxxxxxx with 'help' in the Subject field
List forum is accessible at:
http://tech.groups.yahoo.com/group/si-list
List archives are viewable at:
//www.freelists.org/archives/si-list
Old (prior to June 6, 2001) list archives are viewable at:
http://www.qsl.net/wb6tpu
