UC1875-SP: External Slope Compensation Using Discrete Components

Vince Jison

Part Number: UC1875-SP
Other Parts Discussed in Thread: UC1843B-SP

Hello,

In my previous thread for the UC1875-SP (https://e2e.ti.com/support/power-management-group/power-management/f/power-management-forum/1292405/uc1875-sp-configuring-slope-and-ramp-pins-for-cmc-with-slope-comp) it was explained that external circuitry is needed to support CMC slope compensation.

To do this, I've found two methods (UCC3802D article):

(1) The slope comp ramp can be added to the reference voltage (VREF pin)
(2) The slope comp ramp can be added to the current sense signal (CS+ pin)

For (2), I've come up with the following scheme:

The sawtooth signal generated at the RAMP pin drives the NPN. By way of the resistor divider R1,R2 , a ramp voltage is also generated and that signal is AC coupled into the CS+ pin by means of capacitor C1. I wanted to verify this was the correct approach.

10 months ago

0 Daniel Hartung 10 months ago

TI__Mastermind 26295 points

Hey Vince,

Your implementation is similar to how one would do it on the UC1843B-SP.
Note that the RAMP signal generated here needs a capacitor in order to be formed, and SLOPE needs a resistor.

Thanks,
Daniel

0 Vince Jison 10 months ago in reply to Daniel Hartung

Prodigy 30 points

Hi Daniel,

Thank you for the feedback. As I begun implementing the changes, I decided to look further into the UC1875-SP's functional block diagram. I now think my implementation for external slope comp may be flawed. I originally assumed CS+ was part of the PWM control loop. However, I see the following in the block diagram:

Is my understanding of the datasheet correct that the CS+ pin is only used for checking for over current? It does not look like it's part of the PWM LATCH logic so injecting a slope comp ramp into the CS+ pin may no longer work.

0 Daniel Hartung 10 months ago in reply to Vince Jison

TI__Mastermind 26295 points

Hey Vince,

CS+ is meant for overcurrent events correct.

Thanks,
Daniel

0 Vince Jison 10 months ago in reply to Daniel Hartung

Prodigy 30 points

Hi Daniel,

Understood, thanks for confirming. A slope comp ramp will then need to be injected into the RAMP pin (input to the PWM comparator). To do that, I've made the following changes, please confirm if this new approach is fine for injecting slope comp into the RAMP pin:

This is a modified circuit taken from the UCC3802D's approach to slope comp:

https://www.edn.com/new-slope-compensation-method-stabilizes-switchers/

+1 Daniel Hartung 10 months ago in reply to Vince Jison

TI__Mastermind 26295 points

Hey Vince,

What you have provided seems more reasonable to work.
I dont believe you should ground slope as there may be a current mirror you will damage.
Note that you will likely run into issues actually getting the correct amplitude due to the large variation in ramp that could be generated.

Thanks,
Daniel

0 Vince Jison 10 months ago in reply to Vince Jison

Prodigy 30 points

Hi Daniel,

Thank you for your help and feedback, I will implement the slope comp circuit in my schematic.
As for grounding the SLOPE pin, this is meant to disable the ramp generator since I am now using the RAMP pin as an input:

0 Daniel Hartung 10 months ago in reply to Vince Jison

TI__Mastermind 26295 points

Hey Vince,

Sounds good!

Thanks,
Daniel

Power management

Power management forum

UC1875-SP: External Slope Compensation Using Discrete Components