Hi Team,
I read the datasheet of this device and it does not mention what to do if SI pin is unused. Should I keep it floating or connected to GND? Is it possible not to use sense?
I hope you can help clarify.
Regards,
Marvin
This thread has been locked.
If you have a related question, please click the "Ask a related question" button in the top right corner. The newly created question will be automatically linked to this question.
Hi Team,
I read the datasheet of this device and it does not mention what to do if SI pin is unused. Should I keep it floating or connected to GND? Is it possible not to use sense?
I hope you can help clarify.
Regards,
Marvin
Ditto, I have the same question. Apparently, when you keep it floating you may incur some additional leakage current. Or, at least that is what we measured.
So, looking forward to the answer and also the answer to:
Why does leaving SI unconnected result in about 150uA of additional Iq ?
Our current plan is to just connect SI to Vout directly, as that seems to eliminate the 150uA additional Iq.
Thanks,
JJ
Hi Marvin,
I do not see any concerns with floating or grounding these pins.
I ran some tests in the lab and the device operated fine with <1mA deviation in GND current for both the floating and grounded operation of the Sense. Therefore, I can say I do not see any problems with just floating it. Alternatively grounding the pins may improve thermal performance slightly.
Regards,
John
Hi John,
I agree if you are looking at ~1mA levels of current then there is no problem with floating SI. However, the device has 18uA of Iq and we are seeing 150uA additional when we float SI. Our best "guess" would be that SI floats below 1.12V threshold due to internal design, so this results in SO being actively driven low which results in the additional 150uA. This seems to correlate to the datasheet that states there is a 30k internal pullup on SO. 5V/30k = 167uA. Once you verify my observations, it would be good to add some comments to datasheet so folks know to tie SI to Vin or Vout if unused to ensure you don't incur the extra 150uA of Iq. Many Automotive designs have strict Iq specs and 150uA additional will kill them when not planned. Thanks, JJ
Hi JJ,
Yes, I did not measure the Iq spec with high accuracy, but it is expected to see an increase in Iq due to the pull up. interacting with a floating resource that is inherently unstable. I can run additional testing tomorrow, however your explanation aligns with my own on the matter, and is most likely one of the causes for an Iq increase.
Generally speaking this device is recommended for applications specifically where SI and SO are being utlized as functions the customer is paying for, and if not desired, other parts would be better suited for the application. However we understand with the past two years of device shortages, many customers had to select devices that may not have checked every single box that was asked for, and hence why parts are being used in unintended ways.
Regards,
John
Hi JJ,
It looks like there is an IQ increase, although we would not know how consistent the increase is across process variation as I only tested one device, I will make a note for the datasheet revisions moving forward. We strongly recommend utilizing the SI pin tied to input as then the device will follow EC table specifications with no problem, SO does not need to be utilized, but tying SI to the input rail would allow the LDO to operate normally as if it was being fully utilized.
Regards,
John