• State
  • Replies 6 replies
  • Subscribers 63 subscribers
  • Views 114 views
  • Users 0 members are here
Support feedback
Options
Options
Related

Original question:

UCC33420-Q1: UCC33420-Q1 enable/disable delay time

UCC33420: UCC33420 enable/disable behaviour

Nathan Kagi
Prodigy 30 points

Part Number: UCC33420

I am looking at a low-noise application using the UCC33420. To reduce noise during a particularly noise-sensitive measurement duration I want to turn the regulators off. I have been noticing, what seems to be, switching coming from the UCC33420 after the enable signal has gone low.

The schematic I was using is the image below. I have connected the solder bridge such that the input to the LDO is 5.5V.

image.png

Input voltage supplied from a benchtop power supply at 5V 3A.

CH1 - probed directly across the output using a pogo pin for ground for short measurement loop
CH3 - enable signal, the enable signal is a 2Hz signal from a function generator with 80% duty (80% high)

scope_6.png

This switching occurs at different times after the enable signal goes low. This seems to be dependant on where the burst mode switching of the UCC33420 lines up with when the enable signal is pulled low. You can see this in the image below, the short increase in CH is the chip's burst mode switching.

scope_8.png

I removed the LDO and probed on the output of the regulator itself and measured the FFT of the normal switching and the switching during shutdown. The frequencies are not the same.

"Normal" switching:

scope_7.png

 

Shutdown switching:

scope_9.png

Is there anything that I am doing wrong here? Is this expected behaviour of the chip? I have not replicated this exact measurement on this PCB with the UCC33420-Q1 but I have observed the same behaviour in a different design.

Any advice would be greatly appreciated.

  • 0
    TI__Intellectual 2450 points

    Hi Nathan,

    We have not checked this behavior of the IC. Could you provide the same data with 100% duty cycle EN signal. 

    Since you are using large capacitor at output and no load is connected, the output cap discharges slowly and changing burst frequency might be the cause for different switching frequency during shutdown. Try to put some load in your application and let us know what do you observe? 

    Thanks,

    Prateek

  • 0 in reply to Prateek Verma
    Prodigy 30 points

    Just to confirm, the current PCB has no LDO populated, the 1uF is still present on the PCB. The measurements were taken by measuring on the terminals of the second 22uF capacitor (the component proximity to the IC matches the schematic). I took measurements under three load conditions, no load, 1kR and 100R. The resistors were added on the 1uF cap. I have included 100% and 80% duty using the same 2Hz enable signal.

    Normal switching

    Normal to shutdown

    No load switching

    No load shutdown

    1kR switching

    1kR shutdown

    100R switching

    100R shutdown

    In general the switching frequency during normal operation "wobbles" slightly around the 65MHz. The burst PWM frequency does change adding load, as expected, but the shutdown behavior seems to have remained reasonably consistent.

  • 0 in reply to Nathan Kagi
    TI__Intellectual 2450 points

    Hi Nathan,

    I'll give you update after discussing it with design team

    Thanks,

    Prateek

  • 0 in reply to Prateek Verma
    TI__Intellectual 2450 points

    Hi Nathan,

    Just to confirm, you need to know whether below is the expected behavior of device or not right?

    Switching frequency slightly wobbling around 65MHz.

    And during shutdown there is a peak at around 67MHz

  • 0 in reply to Prateek Verma
    Prodigy 30 points

    Hi Prateek,

    Essentially yes. I would like to use this chip but if this is how the chip should behave then it is not suitable for my application. I was hoping to confirm that I was doing something wrong.

  • 0 in reply to Nathan Kagi
    TI__Intellectual 2450 points

    Hi Nathan,

    You have used large output cap (22uF || 22uF = 44uF), instead use output cap with 10uF to 33uF range

    And what is your noise requirement? Like how much suppression do you need at over what range of bandwidth.

    Thanks,

    Prateek