• State TI Thinks Resolved
  • Locked Locked
  • Replies 1 reply
  • Answers 1 answer
  • Subscribers 63 subscribers
  • Views 1011 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.

LM337L: Output Capacitor Requirements

Chuchen Wang
Expert 3470 points
Part Number: LM337L
Other Parts Discussed in Thread: LM337

Hi,

My customer has the following questions:

For the LM337 for 3.3 V output < 100mA load application:

1- Could ceramic capacitor of 10uF OK for the design on the output regulator for output stability?
2- If a ceramic capacitor is not recommended why?
3- Could be a 10uF Tantalum capacitor OK for the design on the output instead of 1uF recommended on the application note?  Datasheet shows tantalum cap must be 1uF
4 - What is the range of the ESR that can be acceptable?

As seen below, tantalum type will be used at C29 per your recommendation. However, C46 is a ceramic type which is mounted at the microprocessor VCC thru a small inductor L1 per microprocessor manufacturers’s recommendation.

The reason for this question came up is the product has been using two 10uF tantalum caps (C29 and C46).  But recently we tried to change to 10uF ceramic and it starts showing 300mV ripple on the VCC, especially using the Y5V type.  Therefore we wanted to know why.  

Thanks!

Chuchen

  • 0
    TI__Guru 55940 points
    Hi Chuchen,

    LM337 was released almost 40 years ago well before ceramic capacitors became common. As such it was designed, tested, and characterized for use with a tantalum capacitor.

    Ceramic capacitors likely have too little ESR to be used by themselves and achieve a stable output; however, you could add a small resistor in series with the capacitor to boost the ESR.

    If you choose to attempt a ceramic capacitor with a series resistor, it is highly recommended that when you prototype your application you perform your worst case transients for the regulator while observing the output. This is a simple way to check the stability of the linear regulator as described in the following application report:

    www.ti.com/.../slva381b.pdf

    Very Respectfully,
    Ryan