jaijeet / mapp Goto Github PK

Issue was filed mistakenly on the VAPP repo: see here

makes an unknown for the ground node!

Findable by following help MAPP

Hello,

I have been trying to install MAPP on my Windows PC. However, the snapshot-link provided in the wiki is not working anymore. Could you maybe update the link? Or is there a new way to install MAPP?

resistor is v-iR = 0 in ModSpec

simplenl;
DAE = STA_EqnEngine(MOD);
SWPO = dcsweep(DAE, [], 'VDD:::E', -1:0.1:1);
dcsweep: no initguess provided for DC sweep initial point;
using zero vector.

In NR>NR_tianshi (line 584)
In NR (line 276)
In QSS>QSSsolve (line 214)
In dcsweep (line 195)
Warning: Matrix is singular to working precision.

NR: Jacobian is singular, aborting.
QSSsolve: failed after 2 iterations
QSSgetSolution: Warning: solution not valid.
QSS failed at VDD:::E=-1
re-running with NR progress enabled
Iter 1: |x|=0
x=[0,0,0,0,0]

In NR>NR_tianshi (line 584)
In NR (line 276)
In QSS>QSSsolve (line 214)
In dcsweep (line 210)
Warning: Matrix is singular to working precision.

NR: Jacobian is singular, aborting.
QSSsolve: failed after 2 iterations
Error using dcsweep (line 212)
aborting DC sweep due to QSS failure.

%% I was trying to learn how's the accuracy of initial guess could affect the iteration time, so at the first
%% time I put a +/-n% variation of final result of DC value as initial guess but it does not show much %% progress, then I directly put final result of DC value to the initial guess of the same circuit at next
%% run.
%% It should converge immediately, but it does not and start oscillate then settle down, the total
%% iteration number is still 10. we suspect it is because the limiting function and try to debug it, perhaps %% any helps could be expected from your side? (The MATLAB code below is for copy and run)

%% It is a example code from A1oDAEAPI-MATLAB_'op.m' %%
DAE = MNA_EqnEngine(current_mirror_ckt());
DC = op(DAE, 'dbglvl', 2);
feval(DC.print, DC);
% the code above is the DC analysis(Newton-Raphson method) with all zeros as initial guess

init = DC.solution; % get the final DC value after 10 iterations
clear DAE; clear DC; % clear all workspace except for 'init' which would be used for later initial guess

DAE = MNA_EqnEngine(current_mirror_ckt());
DC = op(DAE,init, 'dbglvl', 2);% the initial guess of DC value there should be the same as final answer

MOD = MVS_1_0_1_ModSpec_wrapper(); MEO = model_exerciser(MOD);
vdb=0; vgb=-2:0.1:2; vsb=0; vbsi=0;
MEO.plot('dimplicit_equation_2_fi_dvdib', vdb, vgb, vsb, -1.3:0.01:-1.1, vbsi, MEO); % plots look fine, no problems
MEO.dimplicit_equation_2_fi_dvdib(vdb, vgb, vsb, -1.3:0.01:-1.1, vbsi, MEO) % numerical data fine, no NaNs
MEO.dimplicit_equation_2_fi_dvdib(vdb, vgb, vsb, -1.2, vbsi, MEO) % shows NaNs!

This is in MATLAB R2016b

all model exerciser functions should check for NaNs in their results and, if found, announce immediately.

1. Neither model_exerciser nor model_dc_exerciser have any help!
2. both should be enhanced to give the user access to internalNode and internalUnk unknowns, for printing and plotting

Running the tests in step 4 of "00-READ-THIS-FIRST" produced a large number of errors. I eventually traced these errors to a number of dummy functions that had not correctly linked to the correct terminal functions. These functions were

• vecvalder/@vecvalder
  • double
  • jac
  • val
  • valder
• DAEPI-MATLAB
  • basicdiode
  • EbersMoll_BJT
  • DSintSH_Ntype

As a quick work around (without switching to my Linux partition), I simply rewrote the offending functions by hand. I have not looked too deeply into the compiler process, but I'm guessing that MATLAB 2017 has deprecated the functionality of simply listing the target function in the wrapper.
Thanks for creating this very useful package,
Matt

I'd like to try out MAPP and VAPP on Windows with MATLAB,
which means I can't use the install scripts.
My first problem is that after adding the directory
vecvalder and all its subdirectories to MATLAB (with addpath),
vecvalder is still hidden, see below.

-marcel

D:\MAPP\vecvalder

ls
. 0-README @vv1 @vv3 configure.ac test_2.m tests vv4
.. 00-VERSION @VV2 Makefile.in test_1.m test_3.m utils

test_1
Undefined function 'vecvalder' for input arguments of type 'double'.
Error in test_1 (line 9)
x_vecvalder = vecvalder(x_rand, 'indep');
path
MATLABPATH

D:\MAPP\vecvalder
D:\MAPP\vecvalder\tests
D:\MAPP\vecvalder\utils
D:\MAPP\vecvalder\vv4
D:\MAPP\vecvalder\vv4\vv4-MATLAB
D:\MAPP\vecvalder\vv4\vv4-tests

...

If I installed MAPP correctly, doc/workshops/2017-01-30--IITM-course is empty.

hello! I have searched through the MAPP source code on Github, but I was unable to find any functions related to optical device modeling. Could you please let me know if MAPP has any support for optical device simulation, and if so, where I can find the relevant files or documentation?

Using add_subcircuit now gives an error. The error occurs because of the following line

(line 282)
cktnetlist.elements = {the_element};
It seems like changing it to

        if isfield(cktnetlist, 'elements')
            cktnetlist.elements = {cktnetlist.elements{:}, the_element};
        else
            cktnetlist.elements = {the_element};
        end

fix the issue

Too easy now to make mistakes and not be able to debug easily: eg, nl.groundnodename = {'gnd'};

to reduce confusion against model_exerciser - and make it more clear that model_dc_solver requires dc solution (a simulation algorithm), as opposed to model_exerciser(), which just exercises the model's API functions.