proteus simulation problem

Status
Not open for further replies.
You must check "hide common pins" in TLC393 properties and erase the floating wires.
I don't know why, but Proteus don't like of multiple connections in common pins.

After that, there is a convergence problem due errors in U6 (charger) block.
 
Wow it is solved. and its a good input for me "You must check "hide common pins" in TLC393 properties and erase the floating wires.
I don't know why, but Proteus don't like of multiple connections in common pins." I will remember this. thanks.

any clue to to solve the other convergence problem (U6) .

---------- Post added at 20:58 ---------- Previous post was at 20:44 ----------

It got resolved after I changed method to "GEAR". At least this moment things seem to be working.
Thank you all. Hope no new problem pops up now. wish me luck friends.
 

What are the steps that need to be taken for "To many iterations without convergence". what should be done?
Do I upload DNS file too ?

 

What are the steps that need to be taken for "To many iterations without convergence". what should be done?
First, you should try to identify the source of numerical instability to see what can be done.

Many people change the algorithm or modify the parameters, but I never had good results with this technique. I usually add in the simulation (and only for the simulation) dummy elements such as resistors and snubbers.
Sometimes the problem are simply high-impedance nodes, that are solved (in the simulation) with dummy high value resistors (1Mohm. 1Gohm ..).
Sometimes the problem are parasitic ringins, that are solved (not always) with dummy snubbers.
Sometimes.............
And sometimes the only solution is change the simulator (I love LTSpice :smile.

Do I upload DNS file too ?
Would be better.
 

Well here is the DNS file. you can use previous hexfile for microcontroller.

me too love LTspice. Its very flexible. lot of good support. the only but is, there is no support for microcontroller.

I am making note of all the teachings and advises I am getting from you all. I am planing to make single msword document and upload here at the end of my project. so that it will help every reader in solving the issues.

Edit:- I would like to learn, How to diagnose where the snubber or extra resistor is needed. and also how to know which value would work.
 

Attachments

  • Atmega8 Sinewave Inverter.zip
    32 KB · Views: 121
Last edited:

As I said before, there are errors in the block corresponding to U6.

The OA is supplied by default with + / - 10V (in a real situation would burn by the 24V on inputs), depending on the model used for OA that may be irrelevant or cause numerical problems due to excessively negative base polarization of the 2N2222.

Also, the loop gain of OA+2N2222 is too high , and may generate parasitic oscillations and probably numerical instabilities.


The right way is change the power rail of U6 (eg 0 - 30V). If you like, you can see the influence of the other factors decreasing the voltage gain (from 1E5 instead of 1E6) or adding a resistor and a diode to ground in the base the 2N2222 to prevent excessive reverse polarization.
 
as you said ,Just by changing supply voltage to 0-30V solved the problem. Thanks Eduardo.

Also I will be trying other things said by to you.
 

I am trying to simulate UC3845. The spice model I am using works well with LTspice. but when I try to run it on proteus it gives errors. I am attaching error snapshot and spice file. pls let me know what changes do i need to make.

before I tried this i read the help of protues on attaching spice. But facing error.
 

Attachments

  • uc384x.txt
    15.1 KB · Views: 111
  • error.png
    31.3 KB · Views: 116
Last edited:

i tried simulating but there seem to be problem... ie error could you please help
 

hi Guys ! I have also problems with Proteus and i dont understand WHAT IS FLOATING WIRES in Proteus ! Thanks for ur help !!!
Post here also my design with HEX file. I think my idea and algoritm is ok but Proteus did not want to show me the result
 

Attachments

  • Debug.rar
    32.7 KB · Views: 116
Last edited:

Status
Not open for further replies.
Cookies are required to use this site. You must accept them to continue using the site. Learn more…