Hey all,
I recently measured an undershoot on the switching node of the MP4569GN-Z:
I am converting 34V input to 5V output. On the datasheet the maximum ratings are described as -0,3V below Ground.
Is that an issue and if yes, how can I fix this?
Kind Regards
Stefan
Hi Stefan,
Thank you for utilizing Technical Forum to debug this issue, our sincere apologies on delayed response.
Can you please to provide a schematic? and kindly let me know which node you are probing for the ringing waveform with multiple cycles.
Kind Regards,
Nouman
It is totally normal and happens with all buck converters not just ones with an M in the part number. The source of this is the output inductor that will attempt to force current to flow no matter what. As soon as the ON time ends the top side switch turns off attempting to interrupt flow of current in the inductor, the inductor voltage swaps polarity from and the voltage at the switch node goes as negative as it needs to go to keep current flowing in the inductor.
Before the lower switch turns on the current path is the body diode of the lower FET 0.7V give or take, once the lower switch turns ON that undershoot falls to the rds ON of the FET times the current.
So the problem here is you are too observant, the datasheet is BS if everybody was as persnickety as you they would have to rewrite the datasheet to reflect the reality of the circuit operation.
Some parts from some sources will allow a transient undervoltage on this pin to -1 or so ( reflective of the circuit reality) Most datasheets from most manufactures do only allow -.3 or -.7 on this pin and if you go look at their demo boards you will see this spec violated.
Kind of an unsatisfactory situation, it is nothing to worry about in reality but I feel your pain.
BTW are your scope probes compensated
Hello nouman,
an engineer from you answered me already and told me, thats a totally fine situation.
Hello jshannon,
thanks for your comment. As you mentioned, many datasheets are too extreme. Everyone writes, that the for buck, the swn is not be allowed to be below -0.3V. However, an engineer from MPS wrote me and said, its fine. About your question: Yes, the probe is compensated. I am even using a 1 GHz scope with a IsoVu probe (Isolated Probe for fast switching and low cmc error).
Again thank you both for the answers!
Kind regards
Stefan Weller
