DRV8874: Use as an active full-wave bridge rectifier

Here's discrete version I found online

:

Here's an example of a FET controller.  I'm trying to do the same thing but with low voltage DC and integrated FETs:

This is close, but WAY too many parts:

OK, so this is what I want to do, but with integrated FETs and operational down to 8V:

Hey Steve,

Long time! Indeed, an intriguing approach. However, because of the way the DRV8874 is designed to work I doubt it will support this implementation. Seems like you want to rectify only and not have a smooth DC output - no smoothing capacitor. Even if the DRV receives VM via the FET body diodes the device has an UVLO (under voltage lockout) at around 4.2 / 4.3 V. As the rectified AC on VM goes from 0 to peak the device won't become active until after 1 ms, VM > UVLO.

What is the anticipated frequency of the AC input and what would be the load on VM (V_RECT)? We have never come across anyone using this device in this manner. Perhaps you may want to get an EVM and play with it? 

Regards, Murugavel 

Actually, don't need AC- this is for a DC bus, so what do you think about that use case?

I guess that would keep the VM > UVLO threshold all the time. As long as the absolute max. ratings on these pins are not exceeded - may be worth trying. Would the AC input be a sine or a square wave? May have to derive the EN_IN1 and PH_IN2 control signals from the AC input and drive in independent half-bridge mode.   

It's not an AC input- only DC.  The issue is that the wires could be hooked up backwards and the full wave rectifier simply "untwists" the input without adding much loss.

Yes, the PH_IN2 pin needs to be driven according to which input is actually high so that the bridge runs properly.  I wish the other part had integrated FETs- would be slam dunk in that case.