TPS3620: What happens if VOUT is powered externally while VDD and VBAT are not powered?

Hi Ben,

Thanks for the reply, that's good to know. Will any current flow out of VBAT? Or will the chip essentially think that it's "sourcing" power from VDD and keep the path to VBAT disabled? Also, assuming voltage and current limits are respected on the chip, can operating briefly (for a few minutes) in that way cause damage to the part?

Hi,

Just curious, it was mentioned that VBAT will be unpowered.  Is this always the case with the application?  The default path for the main switch is VOUT to VDD, but there could be a parasitic diode (body diode of the switch) to VBAT so might be a chance that VOUT will also pass current to VBAT if VBAT is unpowered.  Does the application include a battery powered voltage connected to the VBAT pin?  

If you have the part with you, will you be able to check to see if VBAT and VDD see any voltage if VOUT is at 3.3V?  There is a possibility that both VDD and VBAT will have current flow if there is a path to GND on VDD and VBAT.

Ben

Hi,

Are there any updates?

Ben

Hi Ben,

In this application, VBAT will normally be powered. The situation where VOUT would be powered while VBAT and VDD are unpowered would only occur once when programming the microcontroller on the board.

As a short-term solution, I found a way to apply 3.3V to VBAT and VOUT at the same time. In theory, I think this is okay: the chip will see 3.3V on VBAT and 0V on VDD, so it will connect VBAT to VOUT via an 8 ohm resistance pathway. 3.3V is also being applied to VOUT via a lower impedance path, so most current will flow through that path instead of through the TPS3620. Does this solution sound reasonable to you?

In the next few days, I want to try applying 3.3V only to VOUT and seeing if there is current flowing out of VDD and VBAT, or just VDD. If it's possible to safely apply 3.3V to only VOUT, that would be the best solution.

Hi Ben,

Yes, I have a bit of an update. Please see my other reply.

Thanks!

Hi, 

Is this 8ohm pathway external?  When the IC is in the battery backup mode, the internal switch (15-ohms) is turned on and is in parallel with your 8ohm resistor.  As long as you are okay to have current flowing through both pathways, the setup should be good to go.  

During normal operation, there will be no current flow through the 8ohm resistor because the voltage potential across the 8-ohm resistor is the same.

Keep me posted with your findings.  Thanks!

Ben

Hi Ben,

No, when I was talking about the 8ohm pathway, I was referring to the TPS3620's internal pathway. The datasheet says it has a typical value of 8ohms, so that's all I meant. The external pathway will have a resistance of less than 1ohm. 

And thanks, I will!

Thank you for the clarification.  Again, keep me posted with your results.

Ben

Hi,

Just following up to see if you have any new updates regarding your application,  Thanks!

Ben

Hi Ben,

Sorry, nothing yet. I'll keep you posted though!

Thanks for the update!

Ben

Hi,

Just pushing this thread back to the top.  Please let me know if you have any new updates.  If you think you might not have any updates for the next several weeks, my suggestion is to close the thread by clicking on "resolved" and opening a new thread when you are ready to present new information.  The threads usually lock themselves after a month if there is no activity.  Let me know what works for you.  Thanks!

Ben

Hi Ben,

That sounds like a good plan. I appreciate all the help and I'll start a new post next time I have an update.

Hi,

Keep me posted when you have new results!  

Ben

Hi

I have a circuit which uses the TPS3620 and has a 220uF capacitor on VOUT. When VDD is removed there is about 0.5 second delay before the TPS3620 switches over to VBAT. Therefore could this be caused by current flowing from the 220uF capacitor into VDD and keeping this supply voltage up until the cap discharges below VIT. When this happens the TPS3620 then switches over to VBAT?

We are thinking of putting a diode in series with between VOUT and the 220uF capacitor but it is not an ideal solution.

Br

Edward

Hi Edward,

That definitely sounds like a plausible scenario. What exactly is your question? Are you trying to eliminate to the 0.5 second delay?

Is this the same application Ben was working on?

I can't tell if these 2 inputs are related or not. Do you have an update for Ben? Ben is out and I am trying to figure out this thread..

Hi Ajay,

Edward's question is not related to my original question. I think he posted to this thread because it is related to the issue he is dealing with. I do not have any updates for Ben currently.

Thanks

Thanks for the confirmation. Let us know when you have the udpates. I will let Edward answer your question and then let Ben follow up with the answer. if necessary, I will chime in or get someone else to help.

Hi Edward,

Please create a new thread for your topic.  Thanks,

Ben