Post Go back to editing

LT3798 V_EN_UVLO

Category: Hardware
Product Number: LT3798
Hello, 
We have a 12V out power supply that keeps restarting its switching cycles. We configured the Vin to the LT3798 (U_SW_VIN on schematic) Vin Enable/Rising as 17.5V, and Vin Disable/Falling as 8.57V using the equation below from the data sheet. We dropped our originally designed turn off voltage from 10V to 8.57V because we were having start up issues under light loads because the Tertiary winding was not able to supply enough power to keep V_U_SW_VIN from dropping out (Figure 1 Below). This helped with start up issues, but we now have an intermittent start up issue where the Turn On Voltage (17.5V) and the Turn Off Voltage (8.5V) look correct on initial power up, but the LT3798 starts switching again at only 10V (instead of 17.5V) when rising after the initial start up failure (Figure 2 Below). Can you help us figure out why it begins switching again at 10V instead of 17.5V? Is there another fault scenario we should be looking at? 
 
Note: 
1. We can sometimes get the supply to work by probing Pin 12 (EN/UVLO). We are trying to determine if the added probe capacitance filters some of the noise.
2. Our Nas turns ratio is 1:1. With our output being only 12V, this does not leave a large margin with the tertiary flyback voltage being only 12V. We are looking into changing this turns ratio, but this is the first time we've come across this problem in years. 


Fixed Screenshots
[edited by: cboschert at 7:07 PM (GMT -4) on 17 May 2024]
Parents
  • Update: It is likely that we were dropping below the INTVcc dropout voltage so the gate driver would shut off, but we were not dropping below the UVLO threshold. Therefore, the internal regulator would drop out, but the overall chip would not go into UVLO. Vin would ramp up to ~10V and the internal regulator would start up again and drop out at ~9V. Here is the waveform of the Vref pin of the LT3798 that illustrates the Linear Regulator Dropping out : 

Reply
  • Update: It is likely that we were dropping below the INTVcc dropout voltage so the gate driver would shut off, but we were not dropping below the UVLO threshold. Therefore, the internal regulator would drop out, but the overall chip would not go into UVLO. Vin would ramp up to ~10V and the internal regulator would start up again and drop out at ~9V. Here is the waveform of the Vref pin of the LT3798 that illustrates the Linear Regulator Dropping out : 

Children
No Data