• State
  • Locked Locked
  • Replies 3 replies
  • Subscribers 63 subscribers
  • Views 225 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.

Original question:

LMS1585A: Capacitor and ripple

LMS1585A: Instability of LDO using Ceramic output capacitors

Peter Doyle
Prodigy 10 points
Part Number: LMS1585A

Tool/software:

We are using this device (3.3V Fixed) in a design that makes use of X7R output capacitors. We have been experiencing intermittent instability which led me to research this LDO and came across AN-1482. This note seems to suggest the use of ceramic output caps can cause instability in LDOs that are not designed for them.

My question is, does the LMS1585A fall into this category ? 

Is there a suitable replacement capable of ceramic output caps ?

Is there a way to find LDOs that support ceramic caps? Since not all datasheets contain this information.

  • 0
    TI__Guru 50025 points

    Hi Peter, 

    Peter Doyle said:

    My question is, does the LMS1585A fall into this category ? 

    I believe so. It's not super obvious from the datasheet and I had to look through it to find a hint, but in the EC table it is mentioned that tantalum capacitors were used for characterization. 

    Peter Doyle said:
    Is there a suitable replacement capable of ceramic output caps ?

    I think LM317 is the closest to working, but the KCS package it comes in isn't exactly the same. 

    Peter Doyle said:
    Is there a way to find LDOs that support ceramic caps? Since not all datasheets contain this information.

    Pretty much any device that was released in the last 10 years will have more details for the stability requirements and will be ceramic-stable. Avoiding old devices is the easiest way to avoid the ones that are not designed for ceramic caps. However, sometimes it's not possible to if you have a particular application need, such as a high-current device that can handle a mid-VIN range input voltage like the LMS1585A. We have other 5-A devices, but the newer ones are low-VIN devices. In the case that you can't avoid the old device, it's usually acceptable to use ceramic caps with a series resistor to artificially add the required ESR.

    Regards,

    Nick

  • 0 in reply to Nick Butts
    Prodigy 10 points

    Thanks for your response Nick, I will look at replacing the LDO in the longer term. The current design uses 4x10uF X7R caps (a bit ambitious as it turns out) in parallel on the output, do you think if we increase the ESR by using a single MLCC ? or revert to a Tantalum Polymer would be the best approach with the current layout?

  • 0 in reply to Peter Doyle
    TI__Guru 50025 points

    Hi Peter,

    Even a single MLCC has much lower ESR than tantalum, so I would expect the device to be unstable in this case as well. Replacing the caps with a dry tantalum-type capacitor is the best you can do without a layout change.

    Regards,

    Nick