Products Mentioned

AD9102 Production
The AD9102 TxDAC and waveform generator is a high performance digital-to-analog converter (DAC) integrating on-chip pattern memory for complex waveform...
Datasheet
AD9102 on Analog.com

AD9106 Production
The AD9106 TxDAC® and waveform generator is a high performance, quad digital-to-analog converter (DAC) integrating on-chip pattern memory for complex waveform...
Datasheet
AD9106 on Analog.com

Thread Details

  • State Verified Answer
  • Replies 4 replies
  • Subscribers 26 subscribers
  • Views 2061 views
  • Users 0 members are here
Options 

        

            Post
             Go back to editing
        

    

    


    

    









		


















	
	
			
	
			
		
	
	
		

	    
	    
	    		
AD9106/AD9102 clock issue

		
	
				

			

									
					    yorkpa
					
							

            
            			
				
									
											
			
			
				
    
			
			
			    	
			                    
                
				                on Jun 29, 2015 
				
			
			
		
    
		
		                    	    
    
        	    
    Thread Summary
    
                The user experienced SPI communication failure on the AD9106 when the DAC clock (TI CDC421 156 MHz LVPECL) was enabled. The issue was resolved by using an LVDS clock (IDT ICS844021-01 at 166.66 MHz) and powering the AD9106 at 1.8 V, which corrected the CLDO voltage from 1.4 V to 1.8 V.
                        	    
    
                
                     AI Generated Content
                
    	    
    
             
				
        		
    
		    
		    		    
		    			
    
			    
    I have a really strange issue with SPI communication on the AD9106.
    
 
    When I disable my DAC clock (TI CDC421 156 MHz LVPECL), SPI communication (330 kHz) works great, but when I turn the DAC clock on, I lose SPI communication-- I can no longer read from the registers.
    
 
    How could the DAC clock affect SPI communication?
    
 
    Perhaps I'm not terminating my DAC clock or AC-coupling it properly, but I think it's fine:
    
 
    
 
    I see the common mode voltage properly sitting at 0.9 V and I see a 500 mV peak-to-peak swing from CLKP to CLKN
    
 
    One thought is that there might be a setup time issue that arises with using the faster clock.
    
 
    Or, if I'm not terminating it correctly, there's some sort of over-current issue?
    
 
    I'd really appreciate some insight into what might be going on.
    
 

    Pete
    
			
    
			
									
		                             
                        

    
    
                                            
                 ad9102
                
                                                    
                 lvpecl
                
                                                    
                 spi problem
                
                                                    
                 spi
                
                                                    
                 clock and timing
                
                                        
                                            
                
                
                                        
                                            
                
                
                                        
                                            
                
                
                                
                    Show More
            
    
    
    
		
    

				
				        
        		
    
			
						
									
												
			
			
                        
                                    
                        
                        
                        
            
			            
						
						
			
			
			
			
		
    

        
        
        
    
					
    
			
    


    
















    
	
    
	
    
					



			








	
	
	
	



Parents
	
      
	

    • 
	
      



	
		
	
	



      
	
      
		
      
							
					
      
		
      
		    
		    		    
			
				
									
							        
			
			
			
        
            
						
                            
            
            				
            
                on Jul 1, 2015 8:14 PM
            
            
						
			            		
        
		
										0
					
	
        

		
		
        
		
        
                                
                                                            
        
					
        Thanks, Larry. I think there's still something strange going on with the DAC clock.
        
 
        SPI works great if the DAC clock is disabled or if TRIGGER_BAR is disasserted.
        
 
        I lose SPI if I enable the DAC clock and assert TRIGGER_BAR. Then, if I disassert TRIGGER_BAR, SPI is not restored, but if I turn off the DAC clock, SPI comes back. So strange!
        
 
        Do you have any insight as to why this would happen? I seriously think it's a clock issue. I'm trying to use the clock circuit in the picture above. I'll probably end up trying to use a different clock, but I'd really like to know why/if this one is fouling things up.
        
 

        Pete
        
			
        

		
			
	      

		
		
	
        
	
        
		
        
			
		
        
	
        
	
        
		
						
						
				
						
				
				
		
		
        
			
		
        
	
        

        

        • 

	
      



Reply

        


      • 
	
        



	
		
	
	



        
	
        
		
        
							
					
        
		
        
		    
		    		    
			
				
									
							        
			
			
			
          
            
						
                            
            
            				
            
                on Jul 1, 2015 8:14 PM
            
            
						
			            		
          
		
										0
					
	
          

		
		
          
		
          
                                
                                                            
          
					
          Thanks, Larry. I think there's still something strange going on with the DAC clock.
          
 
          SPI works great if the DAC clock is disabled or if TRIGGER_BAR is disasserted.
          
 
          I lose SPI if I enable the DAC clock and assert TRIGGER_BAR. Then, if I disassert TRIGGER_BAR, SPI is not restored, but if I turn off the DAC clock, SPI comes back. So strange!
          
 
          Do you have any insight as to why this would happen? I seriously think it's a clock issue. I'm trying to use the clock circuit in the picture above. I'll probably end up trying to use a different clock, but I'd really like to know why/if this one is fouling things up.
          
 

          Pete
          
			
          

		
			
	      

		
		
	
          
	
          
		
          
			
		
          
	
          
	
          
		
						
						
				
						
				
				
		
		
          
			
		
          
	
          

          

          • 


        


Children
	
        No Data