Products Mentioned

HMC981LP3E Production
HMC981LP3E is an active bias controller that automatically adjusts the gate voltage of an external amplifier to achieve constant bias current. It can be...
Datasheet
HMC981LP3E on Analog.com

Thread Details

  • State Verified Answer
  • +1 person also asked this people also asked this
  • Replies 9 replies
  • Subscribers 70 subscribers
  • Views 706 views
  • Users 0 members are here 

        

            Post
             Go back to editing
        

    

    


    

    









		


















	
	
			
	
			
		
	
	
		

	    
	    
	    		
Required configuration resistors for HMC981LP3E

		
	
				

			

									
					    kdrain
					
							

            
            			
				
									
											
			
			
				
			
			
			    	
			                    
                
				                on Jan 21, 2026 
				
			
			
		

		
		                    	    

        	    
Thread Summary

                The user asked if resistors are required in both feedback paths when adjusting Vneg and Vgate_threshold on the HMC981LP3E. The support engineer confirmed that resistors are not always required in both paths, and the behavior observed (Vgate being pulled to Vneg) is expected under no-load conditions. The user verified the configuration by connecting the target amplifier (Mini-Circuits AVA-183MP+), and the issue resolved. The Vgate_threshold was slightly off, but the overall setup worked as intended.
                        	    

                
                     AI Generated Content
                
    	    

             
				
        		

		    
		    		    

		      		            		        
Category: Hardware

    		        		            		    
Product Number:  HMC981LP3E  

    		        		    		    

		    		    
		    			

			    
When adjusting Vneg and Vgate_threshold on the HMC981LP3E, are resistors always required in both feedback paths, or can Vgate_threshold be adjusted while Vneg remains at its default value of -2.5V?


I want to use the HMC981LP3E to bias a Mini-Circuits AVA-183MP+. The bias requirements in its datasheet say it needs -2V Vgs to pinch off the channel, so I want Vgate_threshold set to -2V. The note in the spreadsheet calculator says I need a threshold at least 300 mV more positive than Vneg, so I thought no adjustment would be necessary for Vneg. I added a 1.5 Meg resistor to the R6 position on the HMC981LP3E eval board, but on startup the value of Vgate is pulled all the way down to Vneg(-2.5V). The examples I have seen for configuring this part tend to adjust Vneg and Vgate_threshold in tandem, but AN-1363 doesn't specify that this is required. Is my problem just that I didn't choose do adjust Vneg up by 100 mV along with Vgate_threshold?




To be clear, I don't have the amplifier connected because I don't want the excessive negative voltage to damage the gate, so I've considered that the missing headroom between Vgate_threshold and Vneg could just be the expected behavior under a no load condition. If that's the case, is there a recommended way simulate an amplifier under control and verify the HMC981LP3E is properly configured without connecting it to an expensive amplifier?


 

			

			
									
		                             
                        

    

                                            
                 hardware
                
                                                    
                 RF Amplifiers Bias Controllers
                
                                                    
                 hmc981lp3e
                
                                                    
                 rf amplifiers
                
                                
            

    
    
		


				
				        
        		

			
						
									
												
			
			
                        
                                    
                        
                                                    
                        
            
			            
						
						
			
			
			
			
		


        
        
        
    
					

			





















	

	

					



			


				
	









	
    
	

  • 
	
    



	
		
	
	



    
	
    
		
    
							
					
    
		
    
		    
		    		    
			
				
									
							        
			
			
			
    •  Analog Employees  
    
            
						
                            
            
            				
            
                on Jan 23, 2026 11:04 AM
            
            
						
			            		
    
		
										0
					
	
    

		
		
    
		
    
                                
                                                            
    
					
    Hi Kyle, 
     
    The resistors are not required in both feedback paths, as long as the VGATE threshold is 300 mV more positive than VNEG. You should be fine with just placing 1.5 Mohms between VGATE and VGATEFB.
     
    Are you applying an enable signal to the HMC981? In the HMC980 datasheet, whose function is very similar to the HMC981, there is mention that “when EN is pulled down to ground, the HMC980LP4E discharges VDRAIN and VG2 down to ground and pulls the VGATE pin to VNEG. VGATE goes down to VNEG regardless of the threshold level setting of VGATEFB”. 
     
    The lack of amplifier on the output could also be causing this. Let me investigate this on my end and get back to you.
     
    Thanks,
    Joe
    
			
    

		
			
	      

		
		
	
    
	
    
		
    
			
		
    
	
    
	
    
		
						
						
				
				
				
				
		
		
    
			
		
    
	
    

    

    • 

	

  • 
	
    



	
		
	
	



    
	
    
		
    
							
					
    
		
    
		    
		    		    
			
				
									
							        
			
			
			
            
						
                            
            
            				
            
                on Jan 23, 2026 11:27 AM
            
            
						
			            		
    
		
										0
					
	
    

		
		
    
		
    
                                
                                                            
    
					
    I was leaving the EN pin floating since it's internally pulled high according to the datasheet. I have used the chip in previous designs and this is how I always configure it. Never noticed an issue before. I can double check my setup to make sure something isn't pulling that pin low though.
    

    I am 99% sure that I am installing the config resistor R6 in the correct spot, but can you give me a sanity check that the component locations marked up below are accurate? There are no silkscreen reference designators on the board for R4-7, so I had to manually trace out the Gerber files. I figure accidentally putting the resistor in the R7 slot would adjust Vgate_threshold in the wrong direction and could cause the issue I'm seeing.
    

     
    
			
    

		
			
	      

		
		
	
    
	
    
		
    
			
		
    
	
    
	
    
		
						
						
				
				
				
				
		
		
    
			
		
    
	
    

    

    • 

	

  • 
	
    



	
		
	
	



    
	
    
		
    
							
					
    
		
    
		    
		    		    
			
				
									
							        
			
			
			
    •  Analog Employees  
    
            
						
                            
            
            				
            
                on Jan 23, 2026 2:29 PM
            
            
						
			            		
    
		
										0
					
	
    

		
		
    
		
    
                                
                                                            
    
					
    Hi Kyle, 
     
    Leaving EN floating is fine since as you said it is internally pulled high. Regarding the resistors, the designators you made are all in the correct spots.
     
    Joe
    
			
    

		
			
	      

		
		
	
    
	
    
		
    
			
		
    
	
    
	
    
		
						
						
				
				
				
				
		
		
    
			
		
    
	
    

    

    • 

	

  • 
	
    



	
		
	
	



    
	
    
		
    
							
					
    
		
    
		    
		    		    
			
				
									
							        
			
			
			
            
						
                            
            
            				
            
                on Jan 28, 2026 4:40 PM
            
            
						
			            		
    
		
										0
					
	
    

		
		
    
		
    
                                
                                                            
    
					
    Hi Joe,
    

    Was there any feedback regarding whether the behavior I am seeing is the expected no-load performance or not?
    
			
    

		
			
	      

		
		
	
    
	
    
		
    
			
		
    
	
    
	
    
		
						
						
				
				
				
				
		
		
    
			
		
    
	
    

    

    • 

	

  • 
	
    



	
		
	
	



    
	
    
		
    
							
					
    
		
    
		    
		    		    
			
				
									
							        
			
			
			
    •  Analog Employees  
    
            
						
                            
            
            				
            
                on Jan 29, 2026 4:41 PM
            
            
						
			            		
    
		
										0
					
	
    

		
		
    
		
    
                                
                                                            
    
					
    Hi Kyle, 
     
    Sorry for the wait. Are you able to make scope measurements on your HMC981?
     
    I managed to do a test with the HMC981 without an amplifier connected on the output. For reference, on this test, I used a 1 Mohm resistor for R6, to set a threshold of -1.8V. I triggered on VDD and took three scope shots - Figure 1 measuring VGATE, Figure 2 measuring VNEG, and Figure 3 measuring VGATE after some time.
     
    Figure 1:
     
     
    Figure 2:
     
     
     
    Figure 3:
     
     
     
    In both shots, CH1 is VNEG, CH2 is VGATE, and CH3 is VDD. The activity seen in the scope shots is the application of VDD. You can see that to start off with this test, VNEG is 0, VGATE is some positive voltage, and VDD is slightly. Once VDD goes high, VNEG goes to ~ -2.5 V and VGATE closely follows, also going to ~ -2.5 V. After ~45 ms, VGATE starts ramping up to ~ 2.45 V. 
     
    I will note the starting points of VGATE and VDD are not expected - when I tested by applying VDD first and then applying VDIG, this issue was not present.
     
    I will test next by trying the HMC981 with an amplifier, the HMC637ALP5E.
     
    Thanks,
    Joe
    
			
    

		
			
	      

		
		
	
    
	
    
		
    
			
		
    
	
    
	
    
		
						
						
				
				
				
				
		
		
    
			
		
    
	
    

    

    • 

	

  • 
	
    



	
		
	
	



    
	
    
		
    
							
					
    
		
    
		    
		    		    
			
				
									
							        
			
			
			
            
						
                            
            
            				
            
                on Jan 29, 2026 5:03 PM
            
            
						
			            		
    
		
										0
					
	
    

		
		
    
		
    
                                
                                                            
    
					
    Joe, 
    

    This behavior looks very similar to what I was seeing. Vgate gets driven pretty much all the way to Vneg and at some point rises to a positive (presumably maximum) value. That ending value makes sense to me because the control loop would be trying to open the drain-source path to reach the current set point and never get there, so it would just saturate.
    

    I will get some scope captures of the waveforms and share them here so we can compare.
    

    

    Thanks,
    

    Kyle.
    
			
    

		
			
	      

		
		
	
    
	
    
		
    
			
		
    
	
    
	
    
		
						
						
				
				
				
				
		
		
    
			
		
    
	
    

    

    • 

	

  • 
	
    



	
		
	
	



    
	
    
		
    
							
					
    
		
    
		    
		    		    
			
				
									
							        
			
			
			
            
						
                            
            
            				
            
                on Jan 30, 2026 2:59 PM
            
            
						
			            		
    
		
										0
					
	
    

		
		
    
		
    
                                
                                                            
    
					
    Joe,
    

    This is the no load performance I am seeing with 1.5 Meg resistor in the Vgate feedback path. C1 is Vdd and C2 is Vgate. The delay before Vgate rises and saturates is a lot less in my case, but otherwise I think we are seeing the same thing.
    

     
    

    I also got my hands on another HMC981 configured for a different amplifier and measured it with and without an amplifier attached.
    

    This is the no load response. C1=Vneg, C2=Vgate
    

     
    

    This is the response with the amplifier attached. C1=Vneg, C2=Vgate.
    

     
    

    

    Because the no load performance tracks with what you and I both saw previously, I was feeling lucky and hooked up my target amp, and it looks fine. I believe that noise at the transition is from my power supply hitting current limit briefly because I set it too low. 
    

    C1 is Vdd and C2 is Vgate.
    

     
    

    

    Vgate_threshold looks more like -1.75V than the -2V I was expecting, but that is a separate issue. Not sure what to make of it, but maybe it's related to the supply current limit triggering.
    
			
    

		
			
	      

		
		
	
    
	
    
		
    
			
		
    
	
    
	
    
		
						
						
				
				
				
				
		
		
    
			
		
    
	
    

    

    • 

	

  • 
	
    



	
		
	
	



    
	
    
		
    
							
					
    
		
    
		    
		    		    
			
				
									
							        
			
			
			
    •  Analog Employees  
    
            
						
                            
            
            				
            
                on Feb 12, 2026 12:26 PM
            
            
						
			            		
    
		
										0
					
	
    

		
		
    
		
    
                                
                                                            
    
					
    Hi Kyle, 
     
    I am happy to hear that your test with your target amp worked. Have you had a chance to test again without the supply current limit triggering? I haven't been able to test with a 1.5 Mohm resistor to get -2V threshold, but with 1 Mohm for -1.8V threshold, I was seeing -1.8 V during testing.
     
    Thanks,
    Joe
    
			
    

		
			
	      

		
		
	
    
	
    
		
    
			
		
    
	
    
	
    
		
						
						
				
				
				
				
		
		
    
			
		
    
	
    

    

    • 

	

  • 
	
    



	
			
		
	



    
	
    
		
    
							
					
    
		
    
		    
		    		    
			
				
									
							        
			
			
			
    •  Analog Employees  
    
            
						
                            
            
            				
            
                on Feb 25, 2026 1:58 PM
            
            
						
			            		
    
		
										+1
					
	
    

		
		
    
		
    
                                
                                                            
    
					
    Closing this thread due to no further activity
    
			
    

		
			
	      

		
		
	
    
	
    
		
    
			
		
    
	
    
	
    
		
						
						
						
				
				
				
		
		
    
			
		
    
	
    

    

    • 

	





	
			








	






































     



















		        
        

            
Before You Switch

            
        

        

        

            Switching languages will make ADI Explorer unavailable. Resume your session by switching back to English and reopening ADI Explorer.