# Forum: FPGA, VHDL & Verilog method for modeling circuit

Rate this post
 0 ▲ useful ▼ not useful
HI every body
i want to model a C-HB for real time simulation and implement this into
FPGA boards.
i read some paper about MNA , but i xplorer method for modeling circuit

thanks

 Author: Jürgen S. (engineer) Posted on: 2014-08-07 22:03

Rate this post
 0 ▲ useful ▼ not useful
Mohammad wrote:
> i want to model a C-HB for real time simulation
I guess that less than 1% in here might guess what you mean by that,
since several definitions came to my mind already.

Assuming you are meaning a coil's HB-curve, you found somebody who did
that, and if you are meaning the concentration of hemoglobin, too.

> and implement this into FPGA boards.
done

> i read some paper about MNA , but i xplorer method for modeling circuit
I am not sure if MNA was the right approach in this case, I did that the
discrete way in FPGAs for my, but here it depends on your detailed task.
In another case, analytical node analysis was the right, but this was
done in software.

Although FPGAs are fine to process Matrix operations required for MNA,
I'd consider to do this in C language to be able to modify algorithms
and handle the depth of iterations  more easily. Moreover recursive
processing is a commonly required method when doing simulation and this
refers even more to MNA. You will need strong knowledge and concepts
regarding pipelined virtual processing to be able to realize interative
(and even recursive processes) in an FPGA effectively.

Rate this post
 0 ▲ useful ▼ not useful
Jürgen Schuhmacher wrote:
>> i want to model a C-HB for real time simulation
> I guess that less than 1% in here might guess what you mean by that,
> since several definitions came to my mind already.
>
> Assuming you are meaning a coil's HB-curve, you found somebody who did
> that, and if you are meaning the concentration of hemoglobin, too.
>
>> and implement this into FPGA boards.
> done
>
>> i read some paper about MNA , but i xplorer method for modeling circuit
> I am not sure if MNA was the right approach in this case, I did that the
> discrete way in FPGAs for my, but here it depends on your detailed task.
> In another case, analytical node analysis was the right, but this was
> done in software.
>
> Although FPGAs are fine to process Matrix operations required for MNA,
> I'd consider to do this in C language to be able to modify algorithms
> and handle the depth of iterations  more easily. Moreover recursive
> processing is a commonly required method when doing simulation and this
> refers even more to MNA. You will need strong knowledge and concepts
> regarding pipelined virtual processing to be able to realize interative
> (and even recursive processes) in an FPGA effectively.

my main problem is that finding a reference ?
reference for MNA or ADC etc

 Author: Lothar Miller (lkmiller) (Moderator) Posted on: 2014-08-09 22:25

Rate this post
 0 ▲ useful ▼ not useful
Mohammad Mothermohammad wrote:
> reference for MNA or ADC etc
Don't use TLAs! Write complete and meaningful sentences instead.
Otherwise no one even understands the question. And given that, how can
you expect a useful answer?

 Author: Jürgen S. (engineer) Posted on: 2014-08-10 12:41

Rate this post
 0 ▲ useful ▼ not useful
Mohammad Mothermohammad wrote:
> my main problem is that finding a reference

No your main problem is, that you still did not describe, WHAT you
want to model. This is the first thing to clarify and HOW to model it,
was the second question. Please refer to my first sentence in the
posting above. Once we know what you mean by C-HB there will be people
regarding WTCs. People won't google for UTAs just to find out, what you
intend to do. In germany for instance MNA is not a commonly used term
for ECSs, but they use the german EKA.

Anyway, possibly MNA was an approach if there are current branches to
investigate which was the case for the assumed magnetic problem. To find
the corresponding BCE's we again need to know the focus and also the
parameters.

In my FMT finally only voltage response was required depending on the
frequency input, so the equations simplified to second order
differential equations which could be ported into a C-language model as
well as an ABM block for pSPICE, to make use of a given simulator. Case
verification (which will probably your next question after having a
solution) was performed by MCA using existent SPICE models for these
cases which were particularly adapted to the frequency. You can get an
impression here:

http://home.arcor.de/juergen.schuhmacher/japarameters.html

The other case, I described above, where RTA was performed by an FPGA,
made use of modified (discrete) equations based on common KCLs.

LOA
------------------

ABM = Analog Behavioral Model
BCE = Branch Constitutive Equation
ECS = electrial circuit simulation
ENA = Enhanced Node Analysis
KCL = Kirchhoffs Circuit Laws
LFL = Lothar's Forum Laws
LOA = List of Abbrevations
MCA = model coverage analysis
MNA = Modified Node Analysis
RTA = real time analysis
TLA = three letter abbreviation
UTA = unknown technical abbrevation
WTC = word to clarify

-------------------
EKA = Erweiterte Knotenanalyse

: Edited by User
 Author: Fathermohammad (Guest) Posted on: 2014-08-19 12:41

Rate this post
 0 ▲ useful ▼ not useful
Lothar Miller wrote:
>> reference for MNA or ADC etc
I really wonder how these guys want to be engineers or scientists one
days without being able to use google or start a scientific research
process. If people are not able to explain their questions the right
way, google will also be of no help.

• $formula (LaTeX syntax)$