• State Resolved
  • Locked Locked
  • Replies 6 replies
  • Answers 2 answers
  • Subscribers 62 subscribers
  • Views 128 views
  • Users 0 members are here
Support feedback
Options
Options
Similar topics

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.

LM120QML: Stability With Ceramic Output Capactors

Keith Jenkins
Prodigy 20 points
Part Number: LM120QML

I am working on refreshing a design that utilizes an LM120H-12/883 (the military grade LM320).  The existing design utilized an old leaded (non-SMT) 1uF ceramic capacitor as well as a solid tantalum 100uF capacitor as the output filtering (and compensation).  I note from the data sheet for the LM120QML that is linked from the LM120H-12/883 product page that a 1uF tantalum or a 25uF aluminum electrolytic is recommended for the output capacitor, and the specific absence of ceramic capacitors recommended for this purpose.  I am aware that other linear regulators of this type that take their output from the collector of a bipolar transistor can exhibit stability issues with ceramic output capacitors due to those capacitors' low ESR.  As part of the design refresh, the output capacitors will be migrated to SMT devices, including the 1uF ceramic, which may likely result in a lower value ESR for that capacitor than the existing design.  Also, due to changes in other parts of the circuit, the output current drawn from this regulator is likely to decrease significantly, likely shifting the load pole of the regulator.  Is there a risk that the revised regulator circuit will have stability issues due to the low ESR of an SMT ceramic output capacitor and the decrease in the output current drawn?

  • 0
    TI__Expert 4700 points

    Hi Keith,

    To make sure I'm understanding correctly, are you saying your new design looks like this? I'm wondering if the 100uF solid tantalum capacitor is still on the output as well.

    Regards,

    Kelsey

  • 0 in reply to Kelsey Farr
    Prodigy 20 points

    Kelsey,
    Yes, the 100uF capacitor is as you have shown it, across the output and parallel to the 1uF ceramic.  I am concerned as the result of information in application notes such as AN-1482 LDO Regulator Stability Using Ceramic Output Capacitors, which indicates that ceramic capacitors can still result in instability even when paralleled with larger non-ceramic output capacitors (see in particular section 7 of that application note).

  • 0 in reply to Keith Jenkins
    TI__Expert 4700 points

    Hi Keith,

    I agree, and I'm a bit surprised there aren't any stability issues with the current design using a ceramic capacitor. I recommend removing the ceramic cap and using only the tantalum cap. 

    The datasheet of this device also recommends connecting a high current diode from output to input to protect against momentary input shorts if the output capacitance in excess of 100 μF, so if you don't plan to reduce the 100 μF tantalum cap, consider adding a diode to protect against reverse current  in the event of LDO input shorting to GND as well. Relevant app note on preventing reverse current HERE, in case you're interested.

    To clarify a point in your original question, though, there isn't an expected difference in the behavior of a leaded ceramic cap vs a SMT ceramic cap of the same value. Regardless, in this case the ceramic cap should not be used.

    Regards,

    Kelsey

  • 0 in reply to Kelsey Farr
    Prodigy 20 points

    Kelsey,
    Thanks for the clarification in regards to ceramic output capacitors, and also for the information regarding reverse currents.  We will take both points under advisement.

  • 0 in reply to Keith Jenkins
    Prodigy 20 points

    One additional question, would there be any harm in converting the 1uF capacitor to a solid tantalum capacitor to improve the -12V supply filtering above the ESR zero frequency of the 100uF capacitor?

  • 0 in reply to Keith Jenkins
    TI__Expert 4700 points

    Hi Keith,

    This question is a bit more difficult to answer. In short, this could potentially cause stability issues. The ESR of the two caps in parallel can be calculated based on the frequency range of interest -- I found this article online that has equations and some calculation examples to help you with that. 

    Unfortunately, however, we don't have the ESR stability range available in our databases for this legacy device, so I'm not able to offer an ESR range off-the-bat for you. Based on the equivalent ESR you calculate for the caps in parallel, though, you should be able to ballpark if it's in a reasonable/non-ceramic cap range. Feel free to reach out again if you need more guidance here.

    Regards,

    Kelsey