• State Verified Answer
  • Replies 3 replies
  • Subscribers 39 subscribers
  • Views 859 views
  • Users 0 members are here
Options
Related 

        

            Post
             Go back to editing
        

    

    


    

    









		











	
	
			
	
			
		
	
	
		

		
About estimating duty cycle



		

			

									
						Tarzan
					
							

			
				
									
											
							 		
		        	            							
			
			    	
			                    
                
				                on Jul 15, 2016 
				
			
			
		


				
		

		    		    
		    
		    	
			
Hi there, I have a question about On duty about step-down converter.

 
According to the basic principle, on duty is estimated like Duty=Vout/Vin, but the results of the measurement result is a bit different.

 
I'm using ADP2301-Eval board and checking ON-duty like the attached condition.

 
http://www.analog.com/media/en/technical-documentation/user-guides/UG-179.pdf 

 
But the calculated value and the measurement vale is not nearly equal.

 
Could you give me some advice on the reasons ?

 

 
Best regards,

 
Tarzan

								


				
				
		

			
						
									
									
						
						
			
			
			
			
		


					

			





	














	

	

					



			


				
	









	
    
	

  • 
	
    



	
			
		
	



    
	
    
		
    
							
					
    
		
    
			
													0
							
			
				
									
							        
            
                                    			
			                            
                        
                on Jul 18, 2016 9:05 AM
            
            
					
    
	
    

		
		
    
					
    Tarzan,
    
 
    A more correct estimate of duty cycle is (Vout/Vin)*(1/e) where e is the efficiency factor. 
    
 
    Energy will be lost in the series resistance of the inductor (Vin must stay applied longer to achieve the same current ramp as for an ideal inductor.)  Channel resistance in the switching FETs, plus the DF of the output capacitor(s) also detract from the overall efficiency.
    
 
    The datasheet for your regulator should have an efficiency vs output current plot that can provide you a good starting point for including 1/e in your initial design calculations.  Actual efficiency depends on many factors. Highly complex models in a Spice simulator can get pretty close, but to know the actual value usually relies on measurements taken from the actual implementation. Consider also that the actual value will not be a constant, but will vary depending on the load current.
    
 
    Based on your calculated D = 0.55 and your measured D=0.646, you can assume the efficiency is ~85.1%.
    
 
    Hope this helps.
    
			
    

		
		
		
		
	
    
	
    
		
    
			
		
    
	
    
	
    
		
						
						
						
						
				
				
		
		
    
			
		
    
	
    

    

    • 

	

  • 
	
    



	
		
	
	



    
	
    
		
    
							
					
    
		
    
			
													0
							
			
				
									
							        
            
                                    			
			                            
                        
                on Jul 19, 2016 9:19 AM
            
            
					
    
	
    

		
		
    
					
    Hi MBrookes,
    
 
    Thank you for your answer!
    
 
    I really appreciate it.
    
 
    Best regards,
    
 
    Tarzan
    
			
    

		
		
		
		
	
    
	
    
		
    
			
		
    
	
    
	
    
		
						
						
				
						
				
				
		
		
    
			
		
    
	
    

    

    • 

	

  • 
	
    



	
		
	
	



    
	
    
		
    
							
					
    
		
    
			
													0
							
			
				
									
							        
            
                                    			
			                            
                        
                on Dec 4, 2018 5:29 AM
            
            
					
    
	
    

		
		
    
					
    I think I am also facing this issue....  Any suggestion?
    

    Download Psiphon
    
			
    

		
		
		
		
	
    
	
    
		
    
			
		
    
	
    
	
    
		
						
						
				
						
				
				
		
		
    
			
		
    
	
    

    

    •