TPS65023: What is the LDO reverse leakage spec?

Hi David,
Thanks for your interest in TI devices!
I checked the datasheet, I don't see the specs you're asking. What I can tell is that usually should be a diode reverse leakage current; it's about few uA depends on the reverse voltage.
I've assigned your question to the device expert; usually you can expect an answer in one business day.

Thanks Phil for your reply. I'm hoping you're correct and its only a few uA (or less). Reverse voltage would be 2.6V from a Supercapacitor. Will wait to hear ...

David,

Since the actual spec is not given in the datasheet, I can only say this for certain: The LDOs are wired such that an excessive or dangerous amount of current when VCC = 0V and a voltage within the normal LDO output voltage range is applied externally at the LDO output pin.

I have reached out to some designers on our team to determine if we can provide an estimate of this value, but this is typically a long holiday week in the US. I do not expect to have an answer before Monday 11/26 at the earliest. It is also possible the answer will be that we cannot provide this value.

Can you provide a schematic that will help me understand why you need a value for this un-listed spec?

Brian,

From the TPS65023 datasheet "The LDOs also have reverse conduction prevention. This allows the possibility to connect external regulators in parallel in systems with a backup battery."

I'd like to charge a 3V backup supercapacitor to 2.6V (max rating at 85C) directly (via a current-limiting resistor) from a TPS65023 VLDO output programmed (with I2C) to 2.6V. If the leakage current of the VLDO output on the TPS65023 is too high (when powered down) I will have to add a low-reverse-current diode and these have fairly large forward voltages (Schottky diodes have poor reverse current). If TI is going to make the claim above it should include a leakage current spec for the TPS65023 VDLO outputs (when powered down).

Note that the supercapacitor will be feeding a TPS7A0512 1.2V LDO, with 1.4V min input, that will feed the 1.2V RTC_CVDD supply of an OMAP-L138 processor. I want to charge the supercapacitor as close to 2.6V as possible to maximize RTC operation time when power is lost.

Thanks,

Dave

Dave,

Thank you for the clear explanation for your concern and need to have an exact value on ILEAK into the LDO when VCC=0V and VLDO=2.6V. When I am able to contact a designer who can lookup or simulate this value, he/she will appreciate your explanation and quantified conditions.

I agree that if the datasheet says "The LDOs also have reverse conduction prevention" then an exact spec value should also be provided.

For newer devices, like the TPS65218D0, which has reverse current protection on the CC pin, the reverse leakage current (ILEAK) is specific as 500nA under certain Test Conditions.

However, on the TPS65023 (initially released in 2006) there is a mention of a feature without the spec to "back it up", so to speak. This is unfortunate and I disagree with this approach, but the situation we find ourselves in cannot be remedied by agreeing on what should have been done.

As I mentioned in my previous reply, I have reached out to some designers on our team to determine if we can provide an estimate of this value, but this is typically a long holiday week in the US. I do not expect to have an answer before Monday 11/26 at the earliest.

Brian,

Thank you for the thorough follow-up. This is exactly what I needed.

Dave Brent