Solar cell Pspice

[jo]

Bonjour, je suis enseignant et à la recherche de modélisations sur
Pspice (sur lequel je débute) de cellules Photovoltaiques.
Je souhaiterais faire un modèle qui ressemble à la modélisation
courante (source de courant, diode, Rs et Rshunt) mais je me heurte à
la resistance interne de la diode dans Pspice qui déforme la
caractéristique I=f(U) attendue.
Je recherche donc un modèle de diode semi parfaite ( tension de seuil
mais pas de résistance série de la diode) ou une technique pour en
faire une (Peut on toucher au paramètres des diodes sous Spice , J'ai
cru comprendre que le logiciel I/Cap4 pourrais peut être m'aider).
Une autre possibilté de modélisation plus évoluée vue ici :
http://www71.homepage.villanova.edu/kalyan.rapolu/photovoltaic_system_modeling_us.htm
serais peut-être envisageable mais dans ce cas je ne comprends pas
tellement comment déterminer mes paramètres et surtout comment
fonctionne cette source Gpoly.

Si quelqu'un à un quelconque renseignement pouvant éclairer ma
lanterne. Merci.

 

[Joerg]

Bonjour Jo,

Can't write in French but understand it un petit peu. The only link that
comes to mind is this, hopefully it helps a little:
http://chuck-wright.com/apppv/pv_spice.html

Cordialement, Joerg

 

[Klaus Kragelund]

Use the model editor in PSpice

Regards

Klaus

 

[jo]

Thanks a lot for your answer. I seen it before but what can I do with
the netlist how can I use it. I would like to draw the same
caracteristics and play with differents parameters.
So I often saw netlist in forum telling about Pspice, but as I am a
beginner I don't know what to do with this.
May I use it in schematics and how? Usually I try to draw the same
circuit and try to get the simulations I expect.
The answer of Klaus show me a part of Pspice I had never used before.
It will certainly help me but I have a problem for creating the .slb
file.
If ever you've got more links once again thank you.

Yours Sincerely

 

[Jim Thompson]

Thanks a lot for your answer. I seen it before but what can I do with
the netlist how can I use it. I would like to draw the same
caracteristics and play with differents parameters.
So I often saw netlist in forum telling about Pspice, but as I am a
beginner I don't know what to do with this.
May I use it in schematics and how? Usually I try to draw the same
circuit and try to get the simulations I expect.
The answer of Klaus show me a part of Pspice I had never used before.
It will certainly help me but I have a problem for creating the .slb
file.
If ever you've got more links once again thank you.

Yours Sincerely

See "SUBCKT.SLB"

on the Subcircuits & Symbols page and

"SubcircuitImportByNetlist.pdf"

on the S.E.D/Schematics page of my website.

I use these techniques to import clients' circuits directly into my
schematics rather than having to redraw circuitry that I'm not
redesigning.

[CAUTION: I know it works with PSpice Schematics... wouldn't hold my
breath trying to make OrCAD Capture do the same.]

...Jim Thompson

 

[jo]

Hello,
I had never used it before. I tried to use it but now I have a new
problem I modified the parameters of a 1N4002 diod and I changed the Rs
param to 1e-6 but that's not enough to have a nice caract like this one

current
!
! !
! !
====== ---------!--------.voltage
0,7V
Which param? which value?

Yours Sincerely




..

Yours Sincerely

 

[Jim Thompson]

Hello,
I had never used it before. I tried to use it but now I have a new
problem I modified the parameters of a 1N4002 diod and I changed the Rs
param to 1e-6 but that's not enough to have a nice caract like this one

current
!
! !
! !
====== ---------!--------.voltage
0,7V
Which param? which value?

Yours Sincerely


Yours Sincerely

What are you trying to do... make an IDEAL diode characteristic?

...Jim Thompson

 

[jo]

Great, I'm discovering the wide possibilities of Pspice
I'll try this as soon as I can maybe after Easter.
Each test is very difficult for me, I made many mistakes, sometimes it
works but sometimes no and I don't know why.

Thanks.

 

[jo]

Yes nearly I still want to have the voltage of 0,7 V because the model
of a solar cell is this one:
http://www.pvresources.com/en/solarcells.php
and I think that the diod is just here for the non linearity of the
I=f(U) caract. The ideal caract should look like a square but because
of the differnt param of the diod I still have a slope in my model.
...Jo

Did you tried vinaiger to prepare a sausage
Excuse me for the mistakes, It's a long time I didn't practice.

 

[Jim Thompson]

Yes nearly I still want to have the voltage of 0,7 V because the model
of a solar cell is this one:
http://www.pvresources.com/en/solarcells.php
and I think that the diod is just here for the non linearity of the
I=f(U) caract. The ideal caract should look like a square but because
of the differnt param of the diod I still have a slope in my model.
...Jo

Did you tried vinaiger to prepare a sausage
Excuse me for the mistakes, It's a long time I didn't practice.

Jo, In your follow-up posts you need to leave some quoted text so we
have a clue as to what you are talking about.

...Jim Thompson

 

[Joerg]

Hello Jo,

Great, I'm discovering the wide possibilities of Pspice
I'll try this as soon as I can maybe after Easter.
Each test is very difficult for me, I made many mistakes, sometimes it
works but sometimes no and I don't know why.

If you haven't worked with SPICE a lot it would make sense to subscribe
to the Yahoo LTSpice group. It's free and over there in the file section
you will find lots of information. There are many example projects where
people made their own models, created sub-circuits and so on. Often
there is more than one way to solve a particular problem in SPICE.

Regards, Joerg

 

[Fred Bartoli]

Hello Jo,



If you haven't worked with SPICE a lot it would make sense to subscribe
to the Yahoo LTSpice group. It's free and over there in the file section
you will find lots of information. There are many example projects where
people made their own models, created sub-circuits and so on. Often
there is more than one way to solve a particular problem in SPICE.

Sure. I already gave Jo 4 or 5 ways to create his almost ideal diode, but it
seems he wants some more

Apart from the obvious N/IS tweaking, a funny one was to put an ordinary
diode on the secondary of an ideal transformer.

 

[Chris Jones]

Yes nearly I still want to have the voltage of 0,7 V because the model
of a solar cell is this one:
http://www.pvresources.com/en/solarcells.php
and I think that the diod is just here for the non linearity of the
I=f(U) caract. The ideal caract should look like a square but because
of the differnt param of the diod I still have a slope in my model.
...Jo

Did you tried vinaiger to prepare a sausage
Excuse me for the mistakes, It's a long time I didn't practice.

The diode should be a p-n diode characteristic where the voltage is
something like I=Is*(e^(q*U/kT)-1), so it will not be really square. Solar
cells ARE big diodes, so the diode is the right model, not some imaginary
square function. If you make it really square then this will be a less
accurate model. Also note that Is depends on temperature, and normally
makes the temperature coefficient of the forward voltage have the opposite
sign from what you would expect from just looking at the T in the formula.

Chris

 

[jo]

Solar cells ARE big diodes.
No problem

If you make it really square then this will be a less accurate model

The aim of my circuit is not to be a realistic model but a simple
circuit easily understood by my students (16-18 year old). And they
just know the ideal diode and don't understand the function e^.
So I try to show them a model where they can measure the different
parameters like loss of power in Rs and Rsh and when a cell is in the
sadow.
Jo