how to remove negetive spike?

[navidrct]

hi
i have a tps54331 and build 4V DC with it.
in digital scope when i see single shot of it , it has negative spike around -250 mv during start up.
i have a output filter like this in output in attachment
how can i remove negative spike?

[rs20]

A schottky diode in parallel with the capacitors?

[navidrct]

A schottky diode in parallel with the capacitors?

ill test it but my negativespike is -250 mv . is there any schottky  that work with -250mv?

[rs20]

You're correct, a schottky is unlikely to turn on a 250mV. Why do you care about a -250mV spike? No IC will be damaged by that.

[navidrct]

Why do you care about a -250mV spike? No IC will be damaged by that.

its for RF Device that its important to them.

[Phoenix]

Can you upload a picture from the oscilloscope and show where on the circuit you are probing?

[rs20]

its for RF Device that its important to them.

You need to give more detail. Is the spike occuring once during start up, or is it correlated to the switching frequency of the TPS54331?

And what specifically are you worried the -250mV spike will do? Damage your device, or just introduce noise?

[navidrct]

Can you upload a picture from the oscilloscope and show where on the circuit you are probing?

actually no because its in RF department and i don't have any access to it . but i probing the output of filter

[navidrct]

its for RF Device that its important to them.

You need to give more detail. Is the spike occuring once during start up, or is it correlated to the switching frequency of the TPS54331?

And what specifically are you worried the -250mV spike will do? Damage your device, or just introduce noise?

its during start up . and im worried about damaging devices.

[rs20]

but i probing the output of filter

Where's your ground lead? Are you sure you're picking up a real signal?

its during start up . and im worried about damaging devices.

I'm not really qualified, but I call [citation needed] on the claim that devices will get damaged, even fancy RF voodoo. Damage occurs due to current flowing through parasitic diodes, overheating them, and creating a short circuit that renders the device inoperable. If the voltage isn't high enough to even turn a Schottky diode on, that sequence of steps is averted at the first step?

But assuming your concerns are legitimate, it is odd that you're getting a negative spike. Is it purely negative, or ringing that goes both positive and negative?

[navidrct]

but i probing the output of filter

Where's your ground lead? Are you sure you're picking up a real signal?

its during start up . and im worried about damaging devices.

I'm not really qualified, but I call [citation needed] on the claim that devices will get damaged, even fancy RF voodoo. Damage occurs due to current flowing through parasitic diodes, overheating them, and creating a short circuit that renders the device inoperable. If the voltage isn't high enough to even turn a Schottky diode on, that sequence of steps is averted at the first step?

But assuming your concerns are legitimate, it is odd that you're getting a negative spike. Is it purely negative, or ringing that goes both positive and negative?

yes im sure in probing.
its ringing in positive as well

[rs20]

Well, I suppose your LC filter is ringing, then? Is it at the right frequency for that to be the explanation? What frequency is the ringing at?

Perhaps you could use diodes in parallel with the inductors to clamp any ringing.

[navidrct]

Well, I suppose your LC filter is ringing, then? Is it at the right frequency for that to be the explanation? What frequency is the ringing at?

Perhaps you could use diodes in parallel with the inductors to clamp any ringing.

its during start up and take 120 us to settle
i don't know if diode work because its just -250 mv

[rs20]

Diodes in parallel with the inductors. If the DC-DC is outputting 4V, and the output after the filter is -250mV, that's 4.25V across the series pair of diodes in that configuration, clearly more than enough to turn them on.

And what is the frequency of the ringing, not the time it takes to settle?

[navidrct]

Diodes in parallel with the inductors. If the DC-DC is outputting 4V, and the output after the filter is -250mV, that's 4.25V across the series pair of diodes in that configuration, clearly more than enough to turn them on.

And what is the frequency of the ringing, not the time it takes to settle?

spike is in start up time and is one time thing and its freq will be 1/120us

[navidrct]

I won't care about it, if I were you.
At such voltage, even it managed to turn on some RF devices, who cares?
As long as there is no lasting effects (such as latch up), or dielectric breakdown (which 250mV is too low), I can accept it.
Avalanche breakdown is never a problem, if the current is limited, and it does not trigger a latch up.

That is my opinion too but they didn't accepted it

Sent from my HTC Desire 820ts using Tapatalk