LP8764-Q1: What fault could lead to below readings from the part?

Hi Mahdi,

The expert for this device is currently out of office. He will be back on 4-24-2024 and take a look at this then.

Best,

Michael Hernandez

Hi Mahdi,

With provided information, to me this looks SC fault on one of the bucks and UV fault is result of that which is causing the orderly shutdown. But in order to figure out which buck is causing the issue is hard to say at the moment since you mentioned that OSD is configured such that any moderate fault on any buck would shutdown all the rails. So, I would suggest to debug by isolating one by one rail if possible to figure out which one is causing the issue? Also please do check why you are having SC from layout or schematic or anything that could lead this fault. 

BR,

Ishtiaque 

Thanks for the reply, Ishtiaque! We have UCD9090 in the same system that does the sequencing and also monitors the voltages and logs the faults. Looking at the UCD logs, it shows that it saw the UV fault for buck3 first as the main cause of this fault. Knowing that, I have a possible scenario that could have happened here:

We know that the inductor that we are using for buck3 has I_sat=4.5A-5.1A, while its ILIM is set to 6.5A. In case the inductor current gets close to say 5.5A-6A, since the current slope will get sharper due to the inductor saturation, is there a chance that the input voltage (3.3V) gets a sharp dip and that leads to all 4 bucks going time almost at the same time (maybe buck3 a little faster)? and that's why we see SC/UV fault for all the bucks? Please note that this scenario has one cabeat and that is why we didn't get any VCCS_MON error? We have enabled VCCA_UV_THR and we have not masked it, and at the end, the read back from  INT_VMON: 0x00, so no interrupt or fault here, why? Could you think of any other possible scenario here that could lead to all 4 bucks showing SC/UV fault?

Can we change our config file such that the pmic does not turn off all the rails in such a fault case? the UCD will do that anyways, once it sees one buck going down.

Hi Mahdi,

Thanks for the more input.  Inductor current saturation is critical parameter and surely could cause destructive behavior by decreasing the inductance and increasing the current peaks at switch nodes. Do you know if the BUCK3 rail current is reaching to saturation level? If that is the case then you need to first address this issue. Try to replace the inductor with higher saturation current from the same family. 

I don't think saturation issue on buck output rail could affect input supply severely but I cannot be ruled out. however, with no, interrupts or any UV/OV fault on VCCA indicates that there was nothing severely affected input side.  

The PMIC configuration can be changed in such a way that it shutdown only that rail which causes the issue. However that would require NVM update. I do not know which device are you using. Is that our Catalog release or custom device specific to customer. 

BR,

Ishtiaque

Thank you, Ishtiaque,

Still the main question (that I brought up in my very first comment here) remains unanswered here that is what possible scenario could lead to the logs that shows SC/UV for all 4 bucks? We know that if the fault happens on only one buck, the PMIC turns off the other three buck SMPSs, but it only reports the fault for the affected buck and not for all 4 bucks. What could have happened here that leads to the above logs? Please help with possible scenarios.

Hi Mahdi,

Could please let me know what is the device OPN? because without knowing the device NVM we will have difficulty to understand the problem nature and device behavior. 

BR,

Ishtiaque