OSD3358-512M-BSM powers up from USB but not VIN_AC

[madhu.a]

I am a bit stumped with this one – we have a custom OSD3358 board. It powers up fine via the USB port – which tells me the rails are coming up fine? But if I supply 5V from an external power supply (a lab supply with a 3A current limit), the board does not power up. The 5Vin rail shows a dip every 40ms down to 1.5V – almost as if the TPIC powers up, sees a large load on the rails and resets. But this does not make sense as the board powers up fine via the USB port. Am I missing something obvious here?
Thanks for any help provided!

[EricT]

This is similar to what happen to me when trying to power the board from 5V supply and 6ft cord.
“The 5Vin rail shows a dip every 40ms down to 1.5V – almost as if the TPIC powers up, sees a large load on the rails and resets”

Adding bulk capacitance to our board fixed the issue.

[madhu.a]

Thank you Eric.
I have a 220uF tantalum at the input to the VIN_AC power rail and another 220uF on the SYS_5V(snippet of the schematic attached). Did you have to add more?
Appreciate you taking the time to reply.

Attachments:

1. 

   Processor_Power.png
   

[EricT]

That should be good, I have about 200uF on ours now.. is the forward voltage drop across Q2 causing an the issue? maybe try bypassing your protection circuit.

also.. Just making sure, your lab supply is not over current’ing?

[madhu.a]

I checked to make sure I have 5.1V after the protection circuit (bumped up the lab supply to 5.2). I see just a 18mA draw on the lab supply. I have attached scope capture of the VIN_AC rail probed right at C1. Can’t figure this out…
Thanks for your help!

[madhu.a]

sorry – the .bmp file was rejected. Trying a .png file..

Attachments:

1. 

   SDS00001.png
   

[Neeraj Dantu]

Madhu,

One thing to make sure is the T_rise_time of the PMIC. The TPS65217C(http://www.ti.com/lit/ds/symlink/tps65217.pdf) requires the input voltage to go from 100mV to 4.5V with in 50ms.

Can you try a different bench power supply with a shorter rise time to make sure that is not the issue?

As you can see in Figure 11 of the datasheet, switching transistors are responsible for switching the power input between VIN_AC and VIN_USB. The SYS rail then powers the rest of the DC/DC converters and LDOs. So, there should be no difference in behavior of these two power rails except for inrush(inrush on VIN_USB can be limited to 500mA. Inrush on VIN_AC can go as high as 1.5A).

You can also dig in a little more and analyze the power up to see where the fault occurs. See https://octavosystems.com/app_notes/osd335x-sm-power-application-note/ for power up sequencing.

Best,

Neeraj

 

 

[madhu.a]

Thank you for the tips Neeraj. I am going to take the measurements you suggested and get back to you.

[madhu.a]

That was it. The rise time from my lab supply was around 60ms to go from 0 to 4.5V. Just enough to screw things up looks like. A different supply with a 30ms rise time took care of the issue. I can now see the ‘CCC’ on my debug powered either from USB or VIN_AC.
Thanks for all the help guys!

[Neeraj Dantu]

Great!

Thanks for letting us know.

Best,

Neeraj

[Author]

Posts