INT- LMZ10503 NOISE ON ENABLE PIN

Hi Nicola,

We haven't seen any issue comes from this before. The EN voltage of this part does show coupled SW waveform on top of the DC value when it is pulled up to VIN, due to the high impedance nature of this pin.

Since the noise comes from the switching action, it is not going to be there before the module is enabled and operating. Since the EN rising and falling thresholds are much lower than the UVLO of VIN, such noise on EN is not going to false disable the module as well.

Yang

Dear Yang,

thanks for your quick answer. The possible issue the customer is concerned, derives from the fact that he set the threshold at 1.76V with a Vin around 5V.

Is it possible, in this case, that the coupled switching activity can cause some issues at startup? For example a false disable can occur when the part starts switching and due to the resistor divider this goes to disable the part?

I was thinking that maybe a false disable can occur but once the VIN is raised up to its nominal value, this can result in false disable only if the noise can exceed the hysteresis value...

Please let me know what you think about it.

 

Thank you and kind regards,

 

Nicola.

Nicola,

I tried the same EN divider ratio as mentioned in your post. The module starts up well over temperature. My observation is that in the beginning of the soft start, the additional noise on EN voltage is added on top of the DC value, while after soft start, the noises shift to below the EN DC voltage. I haven't figured out the reason behind it. It could be the part operate in diode emulation mode during soft start and force PWM after soft start.

Since the EN noise in the beginning of soft start is not lowering the EN voltage, it is not going to false disable the part.

I still recommend to give it more room in the EN DC level if the application allows.

regards,

Yang

Dear Yang,

Thank you again. From the measurements I made, it seems that the noise is centered around the EN DC value, but maybe it is not accurate. ( CH1 enable, CH2 coupled sw node, CH4 Vin)

In any  I made some measurements as well to understand if this noise is able to affect the robustness of the enable, simply setting the divider ratio so that the noise results across the enable H and L threshold as you can see in the following and I couldn't see any issue. My feeling is that the internal comparator bandwidth is not high enough to follow this oscillation, so that the enable is not affected in the end.

Just to conclude: I honestly don't see the reason behind the customer choice to put the enable th below the UVLO H and L threshold :) !

Waiting for your comment I thank you again for your efficient support.

Kind Regards,

Nicola.

I agree with you that the magnitude of the coupled noise may vary given how it is probed. The magnitude on your scope shot is higher than what I've seen. I saw peak to peak magnitude about 400mV.

The EN falling threshold is around 1V. So the noise on top of the DC EN value at 1.76V should not trigger EN falling threshold.

Regards,

Yang

Dear Nicola and Yang,

I have encountered exactly the same noise in EN pin. However, this noise also couples to FB pin which causes regulation problems in the output. To be more clear, output voltage diverts significantly ( + / - 30 %), even though the circuits are exactly the same. Is there any development regarding the issue ?

Best Regards,

Sertkan.