TPS650864: Inquiry about design materials

Hello JH,

Thank you for reaching out.

1. Unfortunately, I only have Altium for creating sch/layout, so unfortunately no DSN file. The device file can be downloaded for Cadence Allegro from here, though I will note that the default is TPS65086401, so the part number will need to be changed to delete the trailing '1': https://vendor.ultralibrarian.com/TI/embedded/?gpn=TPS650864&package=RSK&pin=64&sid=01766a2102bf0021842cdc747bea03072001b06a00bd0&c=1
   
   
2. We do not have any simulation models for TPS650864. For BUCK3/4/5, the inductor should be a fixed 470 nH. For BUCK1/2/6, section 9.2.2.1.1 Selecting the Inductor provides the equations recommended for inductor selection based on the desired ripple current.

Hi JH,

1. The bucks have an internal setting for compensation assuming 470 nH and 1 uH - the original idea was that some orderables might have one or the other. Ultimately, all orderables ended up with the 470 nH so we generally recommend 470 nH. It's not really a hard requirement, just a general recommendation.

2. Good question. The PMP12004-HE was done by a central team, likely they chose it to reduce ripple voltage. It seems to have worked for them so I wouldn't object to using it here.

3. I generally recommend a small boost converter like TPS61040. If cost or space is a high concern, LM2665 or LM2767 can potentially be used but they can be a bit weak for start-up. I have a board with LM2767 in particular on it and it takes 3 or 4 cycles to fully charge LDO5P0. I never noticed anything not working but it just looks a little bad seeing it try to start up, fail, and repeat a few times before staying on. 

Hi JH,

LMR62014 seems plenty powerful to support the VSYS pin of the PMIC.