Products Mentioned

LT3045 Recommended for New Designs
The LT3045 is a high performance low dropout linear regulator featuring LTC’s ultralow noise and ultrahigh PSRR architecture for powering noise sensitive...
Datasheet
LT3045 on Analog.com

Thread Details

  • State Verified Answer
  • Replies 4 replies
  • Subscribers 39 subscribers
  • Views 140 views
  • Users 0 members are here
Related Content

        

            Post
             Go back to editing
        

    

    


    

    









		


















	
	
			
	
			
		
	
	
		

	    
	    
	    		
LT3045 current limit overshoot at startup (10 mA setting)

		
	
				

			

									
					    Atsushi123
					
							

            
            			
				
									
											
			
			
				
			
			
			    	
			                    
                
				                on May 18, 2026 
				
			
			
		

		
		                    	    

        	    
Thread Summary

                The user observed an output current transient exceeding 10 mA during startup of the LT3045 with a 10.1 μF output capacitor. The final answer confirmed that the current limit function may not take immediate effect, leading to a charging current that can temporarily exceed the set limit. The user's actual measurement showed an inrush current of about 50 mA for 10 μs, which is considered normal for the capacitor charging. The engineer recommended reviewing the attached data for further insights.
                        	    

                
                     AI Generated Content
                
    	    

             
				
        		

		    
		    		    

		      		            		        
Category: Hardware

    		        		            		    
Product Number:  LT3045  

    		        		    		    

		    		    
		    			

			    


Hello,


 


I am using the LT3045 with the current limit set to 10 mA.


 


During startup, I observed that the output current may temporarily exceed 10 mA immediately after power is applied. The output has a 10.1 µF capacitor connected.


 


I would like to ask the following:


 


1. Is it possible that the current limit function does not take effect immediately after power-up, allowing a transient current higher than the set limit?


 


2. If this is generally not expected, could it still happen due to part-to-part variation?


 


3. What startup control or external circuit/design method is recommended to reliably avoid this behavior?


 


4. Is this temporary current increase considered normal charging current for the output capacitor, or does it indicate that the current limit loop has not fully taken control yet?


 


Any guidance would be greatly appreciated.


Best regards,

			

			
									
		                             
                        

    

                                            
                 hardware
                
                                                    
                 current-limit
                
                                                    
                 inrush-current
                
                                                    
                 lt3045
                
                                                    
                 LDO Linear Regulators
                
                                
            

    
    
		


				
				        
        		

			
						
									
												
			
			
                        
                                    
                        
                                                    
                        
            
			            
						
						
			
			
			
			
		


        
        
        
    
					

			





















	

	

					



			


				
	









	
    
	

  • 
	
    



	
		
	
	



    
	
    
		
    
							
					
    
		
    
		    
		    		    
			
				
									
							        
			
			
			
    •  Analog Employees  
    
            
						
                            
            
            				
            
                on May 19, 2026 9:59 PM
            
            
						
			            		
    
		
										0
					
	
    

		
		
    
		
    
                                
                                                            
    
					
    

    Hello Atsushi123,
    

     
    

    Let me confirm your problem correctly.
    

    We simulated the behavior at start up with current limit of 10mA.
    

    Is your concern circled in yellow portion?
    

    

      
    

    

    Regards,
    

     
    

    
			
    

		
			
	      

		
		
	
    
	
    
		
    
			
		
    
	
    
	
    
		
						
						
				
				
				
				
		
		
    
			
		
    
	
    

    

    • 

	

  • 
	
    



	
		
	
	



    
	
    
		
    
							
					
    
		
    
		    
		    		    
			
				
									
							        
			
			
			
            
						
                            
            
            				
            
                on May 19, 2026 10:40 PM
            
            
						
			            		
    
		
										0
					
	
    

		
		
    
		
    
                                
                                                            
    
					
    Hi Satoru, 

    Thank you for your confirmation. 

    Our concern is based on actual measurement results rather than simulation. 
    During startup, we observed an inrush current of approximately 50 mA lasting for about 10 µs. 
    This transient current exceeds the configured current limit of 10 mA, which is the behavior we are concerned about. 

    However, this phenomenon does not correspond to the portion circled in yellow. 

    Best regards,
    
			
    

		
			
	      

		
		
	
    
	
    
		
    
			
		
    
	
    
	
    
		
						
						
				
				
				
				
		
		
    
			
		
    
	
    

    

    • 

	

  • 
	
    



	
			
		
	



    
	
    
		
    
							
					
    
		
    
		    
		    		    
			
				
									
							        
			
			
			
    •  Analog Employees  
    
            
						
                            
            
            				
            
                on May 22, 2026 6:22 AM
            
            
						
			            		
    
		
										+1
					
	
    

		
		
    
		
    
                                
                                                            
    
					
    

    Hello Atsushi123,
    

     
    

    We made an experiment by modifying EVkit-DC2491A.
    Modification point is,
    1:Set current limit 10mA
    2:Set Cset small to 0.01uF.
    3:Short PGFB to groud.
    

    The purpose is, the output current to be controlled by the ILIM programming and not something else, so that is why PGFB was shorted  to ground and made CSET really small.   Shorting PGFB to ground keeps the 2mA fast start current on all the time, instead of the 100uA precision reference current.  That is why the LT3045 did not regulate at 3.3VOUT but instead VOUT increased all the way to VIN.
    

    The VOUT voltage ramp and COUT can be used to calculate IOUT:   I=C * dv/dt so I=10uF*5V/6.5ms = 7.7mA.  The blue trace shows the input current which is a few mA higher than the output current.  There is a schematic in the report that shows COUT is 10uF.
    Please have a look excel file attached.
     
    

    Regards,
    

     
    

     LT3045 mod DC2491A 15kILIM.xlsx
    

    

    

    

    
			
    

		
			
	      

		
		
	
    
	
    
		
    
			
		
    
	
    
	
    
		
						
				
						
				
				
				
		
		
    
			
		
    
	
    

    

    • 

	

  • 
	
    



	
		
	
	



    
	
    
		
    
							
					
    
		
    
		    
		    		    
			
				
									
							        
			
			
			
            
						
                            
            
            				
            
                on May 26, 2026 1:05 AM
            
            
						
			            		
    
		
										0
					
	
    

		
		
    
		
    
                                
                                                            
    
					
    

    Hi Satoru,
    

    Thanks a lot for checking this on actual hardware.
    

    We understand your explanation that IOUT is controlled by ILIM and is around 10 mA, based on both the measurement and the C·dv/dt calculation. The behavior with PGFB shorted and CSET minimized also makes sense to us.
    

    We’ll take a closer look at the attached data and follow up if needed.
    

    Thanks again for your support.
    

    Best regards,
     Atsushi
    

    

    
			
    

		
			
	      

		
		
	
    
	
    
		
    
			
		
    
	
    
	
    
		
						
						
				
				
				
				
		
		
    
			
		
    
	
    

    

    •