2024-10-03 07:09 AM
I am currently working with the STEVAL-VP26K01B evaluation board, which utilizes the VIPER26K IC, as part of an electric car project for Formula Student. During my study of the board's design, I noticed that the feedback is taken from a resistor divider between the output and the source of the power MOSFET.
Intuitively, I would have expected the feedback to be referenced to ground, but this is not the case in the design. Could you please provide clarification or insights into the rationale behind this design choice? Understanding this would greatly enhance my comprehension of the system.
2024-10-10 12:20 PM
Welcome @venk2004, to the community!
The VIPER265K used in the STEVAL-VP26K01B is primarily a flyback controller, but here it works as a non-isolated buck, which makes it difficult to recognise the GNDs. From the VIPER265K point of view, GND is not the output GND (here connected to AC_IN), but on its own pin GND, on which C4, C5, C6, C7, D4, R6 and R7 are also connected. This is a tricky circuit that uses a flyback regulator as a buck.
If you think of the the pin GND of the VIPER265K as an internal signal, you can perhaps better recognise the principle. The buck pushes a current through the coil L and charges Cout, whereby the current flows against AC_IN, which is connected to the negative pole of the output. As soon as the target voltage, which is measured via the voltage divider R5/R6 and FB, is reached, the buck switches off, causing the magnetic field of coil L to collapse and allowing the current to continue flowing through L and C7 - but this time via the freewheeling diode D4. Once the hysteresis is undercut, the buck switches on again and the whole process is repeated.
Incidentally, the eDesignSuite shows the circuit that I have inserted in the following image in a slightly more classic way. For a better understanding, I have circled
Note that the output voltage is taken between Vout (top right) and the red-circled, dashed GND.
Hope that makes it a bit clearer?
Regards
/Peter