I am designing a high efficiency power supply which requires a clean +-5V rail for a matrix of Op-amps providing a target +60dB gain on an input signal. However I am running into a few issues with getting the dual output SEPIC to work on the LT1374-SYNC as per the application note 'DN100' and the bottom of the datasheet of the 1374. I have also made a small modification by adding a voltage divider between the positive output and the FB pin, this should allow the output voltage to be adjustable (I have tested this with the standard LT1374 eval board), I actually need the SEPIC to have an output voltage around +-6.3 V as I will pass this into dual linear regulators to get the final +-5V.
I have implemented the circuit on perfboard and I am testing it with 12V (cannot be changed) and it seems to be working quite well, developing the +ve and -ve voltages, however I am not able to get below -5.5 V i.e to my target at -6.3 V and the negative voltage stops changing linearly after -5V. For example when I adjust the potentiometer further the positive voltage continues to adjust up to 10 V however the negative voltage remains stuck in the region of -5 V. When loading the -ve rail with say 10 to 20 Ohms, which should result in 300mA to 600mA being drawn, the voltage on the rail drops drastically to -2 or -3V, however the app note states that the max load should be 1.5A for the 1376 model, so I should expect it to be able more for the 1374 model rated at 3A.
Initially I thought it was because of the 'C4' capacitor in the app-note being rated at 10V and the output being +-5V that I quickly thought it might be the capacitor limiting the voltage, I installed a 47uF rated at 16V but that didn't make a difference and it would seem that the full voltage range is not imposed on the capacitor anyway. I have used the exact same component values as indicated by the schematic at the bottom of the LT1374 datasheet, though having more and bigger capacitors at the output.
Edit: Changing the input voltage from 12 V upwards, adjusts the magnitude of the negative voltage but doesn't change the positive voltage, if I go to 15V for example I am able to achieve the +-6.3V, it effectively changes this non-linear breakdown point indicated by the black arrow. The fact that I can obtain these voltages tells me that I do not have an issue with any of the voltage ratings on my components. I am still having issues with the negative rail being able to sustain load and I am still not able to get -6.3V with an input voltage of 12V only. Please see the below plot:
Any help or guidance will be appreciated as it is first time designing/working with a system that generates -ve voltages and working with switching power supplies.
Best Regards, Jarez
Attached and embedded the graph properly.
[edited by: Jarez95 at 2:18 PM (GMT -4) on 20 Oct 2020]