Author Topic: LDO Output Cap - Changing Tantalum to Ceramic  (Read 1133 times)

0 Members and 1 Guest are viewing this topic.

Offline PsiTopic starter

  • Super Contributor
  • ***
  • Posts: 10385
  • Country: nz
LDO Output Cap - Changing Tantalum to Ceramic
« on: February 17, 2020, 03:08:58 am »
What sort of tests should i do to an LDO to confirm it is not operating right on the edge of stability.
And confirm it won't go unstable in the field.

I have a LM2937-5.0 reg with 33uF 6.3V tantalum in a product and it has been 100% reliable.
https://www.digikey.com/product-detail/en/avx-corporation/F930J336KAA/478-8172-1-ND
https://www.digikey.co.nz/product-detail/en/texas-instruments/LM2937IMP-10%2FNOPB/LM2937IMP-10%2FNOPBCT-ND

But i have been sourcing and soldering the tantalums on myself because tantalums scare me a bit and i don't trust china PCB fabs to put quality tantalums on my boards.

I want to change to using general ceramics that i can trust china to source and place.
Maybe multiple ceramics in series for higher ESR,  or a cap + resistor.

I have seen this reg go unstable with a single ceramic in the past, so i know it can be an issue.
I want to make sure it has been fully tested and is stable, what tests should i run?

Thanks.
« Last Edit: February 17, 2020, 03:12:30 am by Psi »
Greek letter 'Psi' (not Pounds per Square Inch)
 

Offline Paul Bryson

  • Contributor
  • Posts: 21
  • Country: us
Re: LDO Output Cap - Changing Tantalum to Ceramic
« Reply #1 on: February 17, 2020, 03:44:57 am »
What I do to check regulator stability is to apply a step load change and a step change of the input voltage.  You'll want to check at various loads from minimum expected load to maximum.
I want there to minimal change in the output voltage and little to no ringing.
 
The following users thanked this post: T3sl4co1l

Offline ANTALIFE

  • Frequent Contributor
  • **
  • Posts: 516
  • Country: au
  • ( ͡° ͜ʖ ͡°)
    • Muh Blog
Re: LDO Output Cap - Changing Tantalum to Ceramic
« Reply #2 on: February 17, 2020, 03:47:40 am »
How about looking at the power up/down transients at "no-load" and 10mA, 100mA, 1000mA...
 
The following users thanked this post: T3sl4co1l

Offline T3sl4co1l

  • Super Contributor
  • ***
  • Posts: 22436
  • Country: us
  • Expert, Analog Electronics, PCB Layout, EMC
    • Seven Transistor Labs
Re: LDO Output Cap - Changing Tantalum to Ceramic
« Reply #3 on: February 17, 2020, 05:31:28 am »
Doesn't obviously look like it should have problems.  They do make a point of saying "less than 3 ohms", which includes very small values.  It's not obvious how small though, or if other conditions can exacerbate it (like input or output side resonances).

In any case, the dumb and obvious cure is to simply put a resistor in series with the ceramic cap.  I've done it plenty of times myself. :)

Tim
Seven Transistor Labs, LLC
Electronic design, from concept to prototype.
Bringing a project to life?  Send me a message!
 

Offline PsiTopic starter

  • Super Contributor
  • ***
  • Posts: 10385
  • Country: nz
Re: LDO Output Cap - Changing Tantalum to Ceramic
« Reply #4 on: February 17, 2020, 06:50:58 am »
It's not obvious how small though

It does spec a min ohms in another page of the DS

The output capacitor, COUT, must have a
capacitance value of at least 10 µF with an ESR of at least 10 mΩ, but no more than 3 Ω
Greek letter 'Psi' (not Pounds per Square Inch)
 
The following users thanked this post: T3sl4co1l

Offline I wanted a rude username

  • Frequent Contributor
  • **
  • Posts: 662
  • Country: au
  • ... but this username is also acceptable.
Re: LDO Output Cap - Changing Tantalum to Ceramic
« Reply #5 on: February 17, 2020, 07:00:16 am »
That would explain the instability then ... and carry little penalty for putting a 10 mOhm resistor in series.

But it seems (at least thinking about a manufacturer like JLCPCB/LCSC) more likely if you spec a Kemet 33 uF you will get a genuine one ... than if you spec some generic 33 uF MLCC it will really have 33 uF at your actual operating voltage and temperature.  ;)
 

Offline David Hess

  • Super Contributor
  • ***
  • Posts: 17428
  • Country: us
  • DavidH
Re: LDO Output Cap - Changing Tantalum to Ceramic
« Reply #6 on: February 17, 2020, 06:59:16 pm »
The LT1375 datasheet has a good description of how to check the small-signal feedback loop stability which is a little easier than the large-signal load response.  A low frequency square wave source is coupled into the output through a 50 ohm resistor and AC coupling capacitor.  Then the small-signal response can be measured at the output using an oscilloscope.  This should be done while sweeping the load current as described in the linked datasheet.

A large-signal load response test using a switched load like ANTALIFE suggests can also be important as the two tests do not exactly measure the same thing.
 
The following users thanked this post: Psi, exe, ANTALIFE

Offline PsiTopic starter

  • Super Contributor
  • ***
  • Posts: 10385
  • Country: nz
Re: LDO Output Cap - Changing Tantalum to Ceramic
« Reply #7 on: February 17, 2020, 11:07:29 pm »
cool. Thanks
Greek letter 'Psi' (not Pounds per Square Inch)
 


Share me

Digg  Facebook  SlashDot  Delicious  Technorati  Twitter  Google  Yahoo
Smf