Require a Solar Array model in Pspice, [not LTSpice]

Thread Starter

MikeJacobs

Joined Dec 7, 2019
156
So I have a desire to model a solar array in pspice. Not Lt Spice.

I cant find a way to do it. Has anyone done this.

And i should clarify when i say model, i mean i want to build a working model in pspice.

aka, i want to build the model with two terminals + and -
Then i want to be able to hook up a resistor to it and watch the IV curve.

So when its open circuit, i want to see the open circuit voltage and when its shorted i want to see 0V but the full rated current of the panel.

How can i build such a model in pspice

Thanks
 
Last edited by a moderator:
Hi Mike, this can certainly be done in PSpice. Do you have a model for the solar array already? If so, you can turn that into a block and place it between your source and resistor and check out the IV curve. Please reply with what you already have in terms of a solar array model and we'll see what needs to be done from there.
 

Papabravo

Joined Feb 24, 2006
14,244
Spice is spice. If it can be done one one, it can be done in the other. It is likely that one or more of the dependent sources are part of the solution. You should know about and understand these "wee beasties".
 

Thread Starter

MikeJacobs

Joined Dec 7, 2019
156
Hi Mike, this can certainly be done in PSpice. Do you have a model for the solar array already? If so, you can turn that into a block and place it between your source and resistor and check out the IV curve. Please reply with what you already have in terms of a solar array model and we'll see what needs to be done from there.

Hi Thanks

All I have done so far is tried to take a model from Lt spice and duplicate it
However, I cant make it work.
The model in LT uses a special LT part called a linear dependnent current source that does not exist in pspice.



Here is the video I was trying to model.
But since I don't have the linear current source I don't know how to make this work in pspice.
I would really be interested to see this work.

Any ideas?
 

Thread Starter

MikeJacobs

Joined Dec 7, 2019
156
Spice is spice. If it can be done one one, it can be done in the other. It is likely that one or more of the dependent sources are part of the solution. You should know about and understand these "wee beasties".
That's correct, my issues is LT has a part that can be exploited where pspice does not
 
Hi Thanks

All I have done so far is tried to take a model from Lt spice and duplicate it
However, I cant make it work.
The model in LT uses a special LT part called a linear dependnent current source that does not exist in pspice.



Here is the video I was trying to model.
But since I don't have the linear current source I don't know how to make this work in pspice.
I would really be interested to see this work.

Any ideas?
The underlying PSpice model should work fine using the IDC and the E parts from the default PSpice libraries. You can also use the Subparam part if you want to pass properties from the top level down to the lower level of hierarchy.

solar cell PSpice.jpg

At the top level, if you want to sweep the DC sources, you can use the DC Sweep option and that will allow you to do multiple sweeps of each the current and voltage sources. Choose the range that you want them to run up to and you can then probe the current to see the same things that are being shown in the video attached.
 

Papabravo

Joined Feb 24, 2006
14,244
Pspice has E, F, G, and H cells. What it lacks are the behavioral sources. That is an annoyance, but one that can be overcome.
 

eetech00

Joined Jun 8, 2013
1,965
Hi Thanks

All I have done so far is tried to take a model from Lt spice and duplicate it
However, I cant make it work.
The model in LT uses a special LT part called a linear dependnent current source that does not exist in pspice.



Here is the video I was trying to model.
But since I don't have the linear current source I don't know how to make this work in pspice.
I would really be interested to see this work.

Any ideas?
pspice has F and E devices.
You should be able to duplicate that in pspice.
 

Thread Starter

MikeJacobs

Joined Dec 7, 2019
156
The underlying PSpice model should work fine using the IDC and the E parts from the default PSpice libraries. You can also use the Subparam part if you want to pass properties from the top level down to the lower level of hierarchy.

View attachment 214461

At the top level, if you want to sweep the DC sources, you can use the DC Sweep option and that will allow you to do multiple sweeps of each the current and voltage sources. Choose the range that you want them to run up to and you can then probe the current to see the same things that are being shown in the video attached.
So I think we are really close but I still cant make it work.
So like I said above I want to see open circuit voltage and short circuit current.

In the previous video I have seen, the person in the video set that current source to be a linear relationship of E.
I don't see you doing that here. You called it Fcell?

The gain being 36 meaning you have created a 36V cell.
So when I try this I get zero volts on the output. But atleast it compiles which is more then I can say normally.
Correct me if im wrong but to see the ISC you would need to put a current source of whatever amperage you want on your ISC terminal correct?

Can you help me get over the finish line?
Thanks
 

Thread Starter

MikeJacobs

Joined Dec 7, 2019
156
So I think I have it sort of working.
I can see my cell votlage
But like I said above I cant get the model to work like it should

So if I were to short the outputs I should see my short circuit current and ~ 0 V Now I cant really short anything in Pspice or it wont converge so lets say you put a .001 R at the end effectively shorting it.

Instead I get my constant voltage in this case 36V and a huge current. 36/.001 36KA
What I should have is 0V and my short circuit current.

So how can I make the model behave like a real cell.
 
Last edited:

Thread Starter

MikeJacobs

Joined Dec 7, 2019
156
Pspice has E, F, G, and H cells. What it lacks are the behavioral sources. That is an annoyance, but one that can be overcome.
Abd by behavioral source you mean a linear controlled current source which is what I need.
So in other words you cant build this model in pspice.
 

Papabravo

Joined Feb 24, 2006
14,244
Abd by behavioral source you mean a linear controlled current source which is what I need.
So in other words you cant build this model in pspice.
You need to read the definitions for E, F, G, H sources they have a polynomial mode which can be used.
This mode is deprecated in LTspice, but is part of spice AFAIK
 

Thread Starter

MikeJacobs

Joined Dec 7, 2019
156
And this is exactly my point, this does not work like im specifying.
The video has no explination as to why thre is a magic voltage supply under the load.
I have no idea why you would be connecting a voltage source to the negative terminal

Is there a way to do this?
What is that voltage source?
 

Thread Starter

MikeJacobs

Joined Dec 7, 2019
156
Can anyone answer this KEY question. That was the whole point of the post
Why is there a power supply connected to the load when the solar model output should be powering the load solely.

This model simply does not behave in the manor i described in multiple locations

I want to see OV when shorted and ISC,
I want to see OCV when open.

I also want to use it functionally in other circuits with this behavior. And that is why i have no idea why the author of the video and your model above are adding in this mystery voltage source.
If i am going to add a mystery voltage source what good is the model. You dont have another voltage source in real life.
 

Papabravo

Joined Feb 24, 2006
14,244
Can anyone answer this KEY question. That was the whole point of the post
Why is there a power supply connected to the load when the solar model output should be powering the load solely.

This model simply does not behave in the manor i described in multiple locations

I want to see OV when shorted and ISC,
I want to see OCV when open.

I also want to use it functionally in other circuits with this behavior. And that is why i have no idea why the author of the video and your model above are adding in this mystery voltage source.
If i am going to add a mystery voltage source what good is the model. You dont have another voltage source in real life.
The answer to the question is that a voltage source and a power supply are not the same thing. The purpose of the voltage source is to mimic the behavior of the photovoltaic source as a function of irradiance, temperature, and any other parameters you care to add to the model. If the output characteristics of the model and the actual device match, this should be the initial test. What you can do with the model going forward will dictate whether it is sufficient or requires a revision. You can go through the math involved if you have the stones for it, but is that really going to help. Certainly you realize the inherent duality of current and voltage sources, where each of those things can be converted in to the other with associated algebraic gymnastics.
 
Top