Control circuit for Power Converters
There is always a mixed response when comparing power management ICs (PMIC) and microcontrollers in the application of power converters (I mean as pwm controllers). What are the advantages and disadvantages of each? What high level requirements are relevant in deciding which is more appropriate in a particular instance?
2 answers
You are accessing this answer with a direct link, so it's being shown above all other answers regardless of its score. You can return to the normal view.
These are my general thoughts: -
- Easiest - power management IC (PMIC)
- Smallest footprint - PMIC usually
- Cheapest - usually PMIC (for a certain minimum performance level)
- Highest performance - PMIC will outshine all others
- Best reliability - PMIC
- Best stability - PMIC
- Quickest reaction time to load changes - PMIC
- Quickest reaction time to incoming voltage changes - PMIC
- Least start-up overshoot - PMIC (especially boost converters)
And now the bad things: -
- Unreliability - microcontroller
- Instability - microcontroller
- Noisiest - microcontroller
- Poor reaction time to load changes - microcontroller
- Poor reaction time to input voltage changes - microcontroller
which one is better in the long run and also how to mitigate its drawbacks?
Use the better one - if the better one still has drawbacks then extra circuitry may be required to overcome limitations but, without a specific application/job in mind this cannot be answered.
I'll assume by "PMIC" you mean a dedicated switching power supply chip. Such a chip contains the PWM generator and receives output voltage feedback at a minimum. They may also include a driver for an external FET as the switching element, include the switching element directly, include the diode with synchronous rectification circuitry around it, maybe a shutdown input, a power-good output, and other additional features.
Dedicated chip advantages
- Does its job without external intervention.
- Doesn't need a regulated supply already to run. Runs directly from the raw input supply.
- High speed, since everything is in dedicated logic.
- Built-in compensation. As long as you use the right parts according to the datasheet, stability, transient response, regulation performance, and other parameters have already been tweaked for you.
- Very little additional circuitry may be required. For the most integrated chips, all you need to add is the inductor and output capacitor.
- Cheaper than a microcontroller.
Microcontroller running switching power supply advantages
- Cheaper and lower footprint if a microcontroller with a spare PWM module is already there anyway.
- Much more flexible. Since the controller is in firmware, it can do more than is reasonable in dedicated silicon. Examples include feed-forward of the input voltage, non-linear control over parts of the range, run-time tweaking of operational parameters based on measurements, etc.
- Easy to implement high level management, telemetry, LIN, IIC, drive a display, etc.
General tradeoffs
If you just want a 5.0 V or 3.3 V power rail to run stuff on your board from maybe a 9 to 24 V input, then a dedicated switcher chip is usually the best choice. They are small, cheap, and just work.
If you are implementing a power supply as an end onto itself, then something more sophisticated than a dedicated chip is often useful. High end power supplies with 10s of Watts or more output require enough other parts that the difference in cost from a switcher chip to a micro is minimal in the overall scheme. Even if you just implement a controller with standard compensator, you may want telemetry, external digital control, driving of status indicators, electronic fuse, etc.
I have had designs where both dedicated chips and a microcontroller were used to implement different supplies. In several recent designs, I've used dedicated chips to make the 3.3 V and 5.0 V supplies from the industrial 24 V power bus, and a spare PWM generator in the micro to implement an isolated supply.
The PWM output controlled the switch driving the input of a small transformer. The resulting voltage on the isolated side was compared to a threshold, and the result sent back via opto-isolator. That drove the shutdown input to the PWM. Once set up in firmware, it runs completely in hardware.
This was basically a pulse on demand scheme. Those result in more ripple than more sophisticated control schemes, but still good enough for many purposes. They are also inherently stable with good transient response.
1 comment thread