datasheets.com EBN.com EDN.com EETimes.com Embedded.com PlanetAnalog.com TechOnline.com  
Events
UBM Tech
UBM Tech

Characterization of the Power Management Unit

Authored on: Mar 13, 2012 by Sudhansu Sekhar Mishra, Tarun Rehani, Neeraj Jain

Technical Paper

0 6
More InfoLess Info
The paper covers the basic architecture of linear and switching mode regulators, block diagram level descriptions of the PMU, the important electrical parameters for PMU characterization, characterization methodology, instruments used for characterization, some best practices for characterization in terms of the board guidelines, test points, and debugging the issues. This will help any characterization novice trying to understand post-Si characterization and hobby readers, especially undergraduate college students, looking to make a career in Post-Si characterization.
6 comments
write a comment

No Avatar

ashish672 Posted Mar 16, 2012

This is an amazing piece of work. Ash - PHD Cambridge University U.K

reply

sudhansu sekhar mishra Posted Mar 17, 2012

Thank You Ashish for your comments.

reply

No Avatar

Light-bear Posted Mar 22, 2012

I feel that your comments about transition times of "nanoseconds" for loads or inputs are not representative of real-world systems, or of real-world tests. I wondered about the use of a square wave for PSRR, since the classical PSRR test is with sine wave coupling? Short circuit current is not simply about pass transistor RdsOn, it more usually incorporates current limiting and/or foldback limiting, so that the pass transistor RdsON is not 'visible' Standby current is NOT the currrent when no load is drawn, it is the current when the PMIC is in Standby mode, as defined by the device... the no-load current is more likely the Iq specification... I would never recommend using a resistor at the input of a supply to monitor inrush current... the added resistance affects the power supply impedance, and thus the performance of the DUT... I much prefer a GOOD current probe, with high bandwidth and good resolution. That all said, there is still some good 'meat' in your article :-)

reply

sudhansu sekhar mishra Posted Mar 22, 2012

1. The load transition times of nanaoseconds is essentially the boundary conditions for the STABILITY tests. It certainly doesn't have to do anything with the real world systems or use cases. The ringing observed on the regulator output due to these fast transitions would give an estimate of the relative stability of the regulator. 2. Short circuit current is also a test to verify the current limiting feature of the LDO. Very true. 3. Standby current could more or less be interchangeably used with the "Quiescent current" of the LDO. I referred it in this context rather than the "STANDBY MODE" current of the DUT. 4. Using a high BW current probe is definitely an alternative way to measure the inrush current. Nice to have pointed it out. Thank you for the feedback.

reply

No Avatar

Light-bear Posted Mar 22, 2012

Yes, a fast edge, of course, is the hard test for stability... but pretty hard to dive nanosecond load signals with ANY amount of output capacitance ;-) I guess I would have been happier if you had showed both Iq and Istdby tests separately... I found the Tek TCP0030 current probe to be a 'must-have' for validating DC/DC converters and LDO's. Thx

reply

No Avatar

TarunRehani Posted Mar 22, 2012

Well Current Probes are the best options if you can afford it ,using a series of 0.5ohm/1ohm is the alternative and cheaper way to measure the In-rush current. and I think 0.5ohm/1ohm of resistance will not affect the impedance of the supply. Did we mention Square wave for PSRR measurements?? Oops, our mistake ,yes it is done with sine wave. Thank you for the feedback. Thankyou for your feedback.

reply

Please Login

You will be redirected to the login page

×

Please Login

You will be redirected to the login page

×

Please Login

You will be redirected to the login page