• State TI Thinks Resolved
  • Locked Locked
  • Replies 3 replies
  • Answers 2 answers
  • Subscribers 63 subscribers
  • Views 584 views
  • Users 0 members are here
Support feedback
Options
Options
Related

This thread has been locked.

If you have a related question, please click the "Ask a related question" button in the top right corner. The newly created question will be automatically linked to this question.

TPS7A7100: Reverse current protection diode selection when Vin<Vout

KLP KLP
Prodigy 60 points
Part Number: TPS7A7100
Other Parts Discussed in Thread: TPS7A37

Hi, 

I have read all the posting in https://e2e.ti.com/ regarding this issue but could not find a proper solution to it. 

What we understand is the need for a reverse bias diode connecting between Vin and Vout in order to protect the internal body diode of the LDO in the event when Vin << Vout. This is often the case for many situation since the preceding stage could well be a DC/DC with quick discharge enabled. However, since the body diode of the LDO is not characterized and stated in datasheet, it will be impossible to make educated selection of the external diode. 

In many of the posting, the only advice provided is to select schottky diode due to its low forward voltage. However, this parameter is affected by the forward current and the ambient temperature. An increase of ambient temperature from 25C to 70C usually increase the forward voltage of a schottky diode to >0.7V. Would this result in having the external diode not being activated when Vin < Vout? It will be critical for TI to assist us on making this decision.

In our circuit, we have also noticed the external diode causing the Vout to increase in tandem with the ambient temperature.

For example, suppose your LDO circuitry has Vin =15V, Vout=3.3V with Iout=5mA, the output will rise to 3.5V and above as the ambient temperature goes above 70C. This is due to the reverse current of the external diode. Higher ambient temperature increases the reverse current of the external diode. 

As such, we will have to find a diode with extremely low reverse current (range <10uA), low forward voltage (<0.4V) while being able to accept high pulsed forward current across temperature. Could TI assist us in shortlisting such a diode? Or would TI be able to at least provide a hard figure on the internal diode forward voltage?

Many thanks.

Best rgds,

KLP

  • 0
    TI__Mastermind 37005 points

    Hi KLP,

    You are correct that we do not have the parasitic body diode characterized for this older LDO.  We also make LDO's with reverse current protection, such as the TPS7A37, which may be a better fit for your application.

    We recommend limiting the reverse current to 5% of the rating of the device, which for this LDO would be 50mA.  If you have reviewed your reverse current and do not see more than 50mA, then we do not recommend any further action.

    If you need a parallel schottky diode and you are worried about leakage from the schottky diode, you can decrease the feedback resistors to draw some additional microamps on the load to offset this leakage current. 

    Thanks,

    Stephen

  • 0 in reply to Stephen Ziel
    Prodigy 60 points

    Hi Stephen, 

    We have done quite an extensive test comparing different schottky diodes. Any diodes with reverse current in mA will cause the LDO output to rise in tandem with the ambient temperature. Currently, only shunt diodes in the range of uA would be able to ensure it does not cause rise in Vout of the LDO. 

    Could you give a guidance on the maximum forward voltage of the shunt diode? Do you agree that with a forward voltage of 0.8V, the shunt diode will never be activated when needed? Many thanks!

    KLP

  • 0 in reply to KLP KLP
    TI__Mastermind 37005 points

    Hi KLP,

    Understood.  In my previous reply, when I said "reverse current" what I meant was the reverse current of the LDO not the parallel schottky diode.  So if your LDO reverse current is no more than 50mA then we would not suggest that a schottky diode is needed.

    Have you considered placing the diode in series with the input? If your Vin is 15V and your Vout is 3.3V, you can absorb the forward voltage drop. The reverse leakage current out of the series diode should be under 50mA. Like this:

    Thanks,

    Stephen