voltage controlled power supply

[FrankBuss]

I need a circuit which outputs 0-24V, with an input of  43-57V. Max output current is 2A. The output voltage should be controlled by a microcontoller. Doesn't need to be fast, like settling in 0.1s. It will drive a small motor, so needs probably some protection for the back voltage when the motor is turned off.

I was thinking about using this IC: https://www.ti.com/lit/ds/symlink/lm5116wg.pdf and the circuit on page 11. The output MOSFETs are very much overdesigned, but this probably keeps things cool.

For the regulation, ChatGPT suggested I should dimension FB to output 24V, and then just apply a voltage from 0-1.2V to SS with a DAC, and it will scale linearly. Will this work? Does it go down to 0? Or should I use some other IC?

[Doctorandus_P]

What sort of motor? Normally Brushed DC motors are controlled with PWM. The motor inductance and inertia smooth it out.

[Benta]

Yeah, I'd just let the MCU do the PWM calculation based on input voltage and desired output voltage and take it from there, using the timers in the MCU.

[FrankBuss]

Sounds interesting with the PWM, I'll ask the client what motors they plan.

But in general, would be nice to have a voltage regulated output. The datasheet says "The LM5116WG will regulate the FB pin to the SS pin voltage or the internal 1.215V reference, whichever is lower.". So looks like it should work with the SS pin. Of course, the min "on" time probably wouldn't allow to regulate it all the way down to 0 V I guess?

[PCB.Wiz]

For the regulation, ChatGPT suggested I should dimension FB to output 24V, and then just apply a voltage from 0-1.2V to SS with a DAC, and it will scale linearly. Will this work? Does it go down to 0? Or should I use some other IC?

Nearly, like all AI creations.
You need to inject some slight current from above Vref in order to regulate close to zero.
Usually a DAC a bit higher than Vref is used, or you can use 2 reference sources if your only  Vref is the regulator one.

[FrankBuss]

Right, I'll buffer the DAC output with a small opamp, this should overcome the 10 uA internal current source at SS.

Maybe I should combine it with a nice PWM driver like this: https://www.allegromicro.com/-/media/files/datasheets/amt49502-datasheet.ashx Then I can set the output voltage of the regulator IC to the max allowed motor voltage, and regulate the speed with PWM. This would be most flexible and allow also tiny 3V motors, without worrying about low limits of the voltage regulator IC.

[FrankBuss]

I searched a bit more, and looks like with a synchronous regulator, I don't need external MOSFETs or diode, and it still stays cool, even worst case at 2A. Found this nice part:
https://www.digikey.de/de/products/detail/renesas-electronics-corporation/RAA211820GSP-HA0/19235009
But the soft start trick can't be used with it, because it switches to the internal reference after the startup. But I found this article how to inject current into the FB pin to adjust it:
https://www.analog.com/en/resources/technical-articles/a-3step-approach-for-designing-a-variable-output-buck-regulator.html
So this should work then for setting the max voltage, depending on the motor. And then I'll use also this H bridge:
https://www.digikey.de/de/products/detail/power-integrations/BRD1263C-TL/9743341
This will be the most flexible: I can set the voltage to like 3V for tiny 3V motors, and then regulate it to very slow speed with PWM. If I would use a fixed 24V supply, I wouldn't feel comfortable to control a 3V motor with it, even if the PWM gets averaged. Just some but in the firmware or microcontroller crash, and the motor would go really crazy

edit: the 600V h-bridge might be a bit overkill, I think I'll use this one: https://www.infineon.com/assets/row/public/documents/10/49/infineon-btn7030-1epa-datasheet-en.pdf also no flyback diodes needed anymore, and I can even implement a "brake".