Products Mentioned

LTspice Production
LTspice® is a powerful, fast, and free SPICE simulator software, schematic capture and waveform viewer with enhancements and models for improving the simulation...
LTspice on Analog.com

Thread Details

  • State Verified Answer
  • Replies 3 replies
  • Subscribers 11 subscribers
  • Views 245 views
  • Users 0 members are here

Download the Latest Version of LTspice

Options
Related Content

        

            Post
             Go back to editing
        

    

    


    

    









		


















	
	
			
	
			
		
	
	
		

	    
	    
	    		
Bug report in latest LTSpice Version (X64) 26.0.1

		
	
				

			

									
					    RZetopan
					
							

            
            			
				
									
											
			
			
				
			
			
			    	
			                    
                
				                on May 2, 2026 
				
			
			
		

		
		                    	    

        	    
Thread Summary

                The user encountered issues with parameter definitions in LTSpice, where .PARAM Gk=1/GBr^2 incorrectly resulted in Gk=GBr^2 and .PARAM Gk=1/(GBr^2) caused a division by zero error. The solution was to use the correct exponentiation operator ** instead of ^, e.g., .PARAM Gk=1/(GBr**2). The user confirmed that this resolved the problem.
                        	    

                
                     AI Generated Content
                
    	    

             
				
        		

		    
		    		    

		      		            		        
Category: Software

    		        		        		        		        
Software Version:  (X64) 26.0.1

    		    		    

		    		    
		    			

			    
.PARAM GBr=100  (Starting definition)




.PARAM Gk=1/GBr^2 actually results in Gk=GBr^2. Apparently, exponentiation overrides the previous division completely, even cancelling it out!








.PARAM Gk=(1/GBr)^2 works correctly, as does Gk=1/GBr/GBr, and this pair: .PARAM Ga=1/GBr  .PARAM Gk=Ga^2.








Also, a very likely related error:




.PARAM Gk=1/(GBr^2) actually results in a division by zero error message!




It appears that trying to perform squaring immediately after division breaks things, while doing division "separately" (including enclosed in parentheses) works. If this info is insufficient, I can supply the full circuit model.  I am modeling a PMDC motor with a gearbox.  The simulation works very well using my "correct" workarounds shown above.




RZetopan

			

			
									
		                             
                        

    

                                            
                 (x64) 26.0.1
                
                                                    
                 software
                
                                                    
                 LTspice
                
                                
            

    
    
		


				
				        
        		

			
						
									
												
			
			
                        
                                    
                        
                        
                        
            
			            
						
						
			
			
			
			
		


        
        
        
    
					

			





















	

	

					



			


				
	









	
    
	

  • 
	
    



	
		
	
	



    
	
    
		
    
							
					
    
		
    
		    
		    		    
			
				
									
							        
			
			
			
            
						
                            
            
            				
            
                on May 3, 2026 3:35 PM
            
            
						
			            		
    
		
										0
					
	
    

		
		
    
		
    
                                
                                                            
    
					
    ^ is a boolean operation. It you mean to square the number, use **. ^ is not for exponents :)

    try this:
    

    .PARAM Gk=1/(GBr**2)
    
			
    

		
			
	      

		
		
	
    
	
    
		
    
			
		
    
	
    
	
    
		
						
						
				
						
				
				
		
		
    
			
		
    
	
    

    

    • 

	

  • 
	
    



	
		
	
	



    
	
    
		
    
							
					
    
		
    
		    
		    		    
			
				
									
							        
			
			
			
            
						
                            
            
            				
            
                on May 3, 2026 10:30 PM
            
            
												
						in reply to Gudni
					
										
			            		
    
		
										0
					
	
    

		
		
    
		
    
                                
                                                            
    
					
    Shades of FORTRAN!!  That never occurred to me!  That worked, but the parentheses were unnecessary. It does not matter if 1/GBr or GBr gets squared in this specific instance.  Thank you so much. By the way, I tried looking up the LTSpice mathematical operations list before filing my false bug report. It would be nice if there was a list available within the LTSpice "help" function. Of course, I never thought to check the boolean functions.
    
			
    

		
			
	      

		
		
	
    
	
    
		
    
			
		
    
	
    
	
    
		
						
						
				
						
				
				
		
		
    
			
		
    
	
    

    

    • 

	

  • 
	
    



	
			
		
	



    
	
    
		
    
							
					
    
		
    
		    
		    		    
			
				
									
							        
			
			
			
    •  Analog Employees  
    
            
						
                            
            
            				
            
                on May 4, 2026 10:30 AM
            
            
						
			            		
    
		
										0
					
	
    

		
		
    
		
    
                                
                                                            
    
					
    Hi  ,
    

    From the help:
    

    Numerical parameters support the following operations, grouped in reverse order of precedence of evaluation:
    

    


    
Operand
Description

    

    
&
Convert the expressions to either side to Boolean, then AND.

    

    
|
Convert the expressions to either side to Boolean, then OR.

    

    
^
Convert the expressions to either side to Boolean, then XOR.

    

    
 
 

    

    
>
True if expression on the left is greater than the expression on the right, otherwise false.

    

    
<
True if expression on the left is less than the expression on the right, otherwise false.

    

    
>=
True if expression on the left is greater than or equal the expression on the right, otherwise false.

    

    
<=
True if expression on the left is less than or equal the expression on the right, otherwise false.

    

    
 
 

    

    
+
Floating point addition

    

    
-
Floating point subtraction

    

    
 
 

    

    
*
Floating point multiplication

    

    
/
Floating point division

    

    
%
Remainder of floating point division, same as mod(x,y) function

    

    
 
 

    

    
**
Raise left hand side to power of right hand side, only real part is returned, e.g., -2**1.5 returns zero, not 2.82843i

    
    

    

    mike
    
			
    

		
			
	      

		
		
	
    
	
    
		
    
			
		
    
	
    
	
    
		
						
				
						
						
				
				
		
		
    
			
		
    
	
    

    

    •