Current Regulation on a DC Buck Converter

[joelg11]

I'm an entry-level engineer, and I have a design question for the forum:

I'm designing a multiphase buck DC/DC converter which will convert roughly 65VDC to 52VDC at up to 150A. I'm using 4 phases, so inductors, etc will be designed to about 40A/phase, MOSFETS will be paralleled for less (about 13A)current, but that is unimportant.

The crucial point where this design deviates from the common design is that it is essential that we have current control rather than just a current limit. An external analog comparator signal is to dictate the current setpoint. I'm really wanting to use two Linear LTC3860 controllers, but again they only provide a current limit. Another option is to do a custom design using a PIC microcontroller, but I still need help with how to use the current sense circuit to control the PWM to provide current regulation.

Anyone have any ideas? I attached a rough schematic of one of the stages, with the current sense signals going off-sheet to my theoretical controller.

Thanks for your help!

 

[Zryn]

Are you familiar with PID Controllers?

Using a few Op Amps (filters, summing/difference, gain, integration) you can create a tidy user control scheme with an input potentiometer dial and a sense resistor.

A P Controller is generally sufficient to play around and learn how it works, but mostly PI is what you're aiming for.

 

[berkeman]

I'm an entry-level engineer, and I have a design question for the forum:

I'm designing a multiphase buck DC/DC converter which will convert roughly 65VDC to 52VDC at up to 150A. I'm using 4 phases, so inductors, etc will be designed to about 40A/phase, MOSFETS will be paralleled for less (about 13A)current, but that is unimportant.

The crucial point where this design deviates from the common design is that it is essential that we have current control rather than just a current limit. An external analog comparator signal is to dictate the current setpoint. I'm really wanting to use two Linear LTC3860 controllers, but again they only provide a current limit. Another option is to do a custom design using a PIC microcontroller, but I still need help with how to use the current sense circuit to control the PWM to provide current regulation.

Anyone have any ideas? I attached a rough schematic of one of the stages, with the current sense signals going off-sheet to my theoretical controller.

Thanks for your help!

Interesting design and questions. I'd be inclined to go with the uC controller, since you presumably will have a moderate amount of DSP to do to make this thing work. I'm not sure that a PIC will have the horsepower to do it, but it might.

How do you plan on managing the interleaving of the 4 stages? I'm not familiar with multiple parallel feeds in a buck converter topology. I'm guessing that each stage has its own inductor, right? What do you plan on doing with the current sense information? You cannot have both a set output voltage and a set output current at the same time, right?

 

[joelg11]

Thanks for the replies:

Zyrn: I'm not familiar at this point with PID Controllers, or P, or PI as you reference - I guess I have more homework to do. In the schematic I posted, I am using the DRC of the power inductor as my sense resistor, and I need to work out what you mentioned - the tidy scheme of amplifiers that will take this signal, compare it to a 0-5V reference to ultimately control the PWM on the switching circuit.

Interesting design and questions. I'd be inclined to go with the uC controller, since you presumably will have a moderate amount of DSP to do to make this thing work. I'm not sure that a PIC will have the horsepower to do it, but it might.

How do you plan on managing the interleaving of the 4 stages? I'm not familiar with multiple parallel feeds in a buck converter topology. I'm guessing that each stage has its own inductor, right? What do you plan on doing with the current sense information? You cannot have both a set output voltage and a set output current at the same time, right?

The interleaving of the 4 stages is what I am having a hard time conceptualizing. Each phase has its own LC on the output, and will be 90deg out of phase with the other. Some prefab controllers (LTC3860) have all of this built in - they sense output current on each phase, average it and compare it against a preset current limit - but this only provides a ceiling, not a current control.
I should explain that the current sense transformer in the attached schematic is only for input current - nothing on-board will be done with this information.
The plan is to have a voltage range with a limit; the voltage will be unregulated from 50-55V, but clamped at 55V if it tries to go above the range (current decreased). If it is within the allowable voltage range, the current must be controlled, with the input providing a constant current source to the load (input cannot rapidly change).

Please let me know if this does not make sense, I'm mulling it around in my head and becoming more confused the more I think of it :)

 

[Zryn]

PID controllers can be a bit of a struggle when you're learning them in a hurry, so a text of some kind would probably be more helpful, but here's a quick rundown.

Using control analysis (Laplace) on the first diagram, it can be seen that:

Output = Error * Gain
Error = Input - Output
thus: Output = (Input - Output) * Gain
rearranging: Output = Input * [ Gain / (1 + Gain) ]

Looking at the second picture, If the Output is connected to the first Op Amp to form a closed loop, it can be said that the inverting filter signal (from Op Amp 1) is summed (in Op Amp 2) with a user Input from a potentiometer. The resulting Error signal is then amplified by a Gain (Op Amp 3) and fed round in the closed loop, which will result in a follower (output follows user input) by the afore mentioned formula (there will always be some small difference in the signals, depending on the gain, but if the gain is too high, the signal will oscillate instead of settle to a final value).

Now, given a sense voltage (using a known resistance) you know the current that is actually in use. This sense voltage is fed through the feedback loop and turned into a control voltage which is some scale of the user input and fed into the control circuitry to increase/decrease duty cycle, frequency ... whatever you want ... to bring the actual signal closer to the user input signal.

The main problem I had with PID (and still have) is thinking/analysing of the Op Amps in an open loop configuration as just a filter summer and gain, compared to the PID requirement of a closed feedback loop which changes everything.

Hope this makes sense!

 

[Zryn]

Done!

I realized moments after posting that the pictures I uploaded had escaped into the interwebs somewhere, never to be seen again.

 

[joelg11]

Very helpful response Zryn. To write it out to make sure I understand - if I use a voltage sense ckt on the output node of one of my phases, I'll have the voltage (scaled down to a 0-5V signal).
I guess I would divide this by the DRC resistance of the output inductor (or use a low R current sense resistor) to determine the current? Let's say my user Input is a 0-5V signal coming from the main controller. I'll sum that with my output of OpAmp1 (through identical resistors). This will give me my error signal (expected in the range of ?) which I'll amplify to a 0-5V signal and connect it back to the input node.

That helps, now my conundrum is using this control voltage to adjust the PWM accordingly on 4-phases. I'd very much still like to use the LTC3860 (2 stacked) if it could take this input and do what I need to do with it - more thinking required.

 

[Zryn]

Looking at that PDF, there is a lot of mentions of built in error amplifiers and using the inductors DCR or a sense resistor for current/voltage control. You may have everything you need in that IC, and just need to figure out how to get it to do what you want (much easier said than done.

Also, COMP1 (pin 3) & COMP2 (pin 7) on page 7 appear to be the external Duty Cycle control pin for the output PWM waveform, so you can put in an external control voltage to modify the on time of the PWM and thus change the output of the MOSFET (by feeding that PWM into the LTC4449 gate drivers or any other type of gate driver to have the MOSFET on for longer or shorter).

So you have the sense resistor voltage, which will show ~1mOhms x 150A = 150mV maximum as the feedback signal.

Your user input is 0-5V and you want the current to range from 0-150A, therefore you need to scale the user input voltage, so that when your user requests, for example, 90A and you actually get 3V into your control circuit, the scaling would change this to 90mV which in the end represents the desired 1mV per A, and relates to the 0-150mV from the feedback signal.

These numbers are all made up to demonstrate that scaling may (or may not) also be required. As mentioned above, I'm not familiar with your IC, but it may be able to help you out and do all sorts of exciting things for you!