Author Topic: LDO Regulator Oscillation?  (Read 9023 times)

0 Members and 1 Guest are viewing this topic.

Offline esar

  • Contributor
  • Posts: 24
LDO Regulator Oscillation?
« on: January 25, 2014, 02:57:43 pm »
Hi all, this is my first post here, though I've been lurking for a while...

I've been designing a simple wireless dimmer for LED strips based around an Atmel Zigbit module and I recently got the first version of the PCBs back from China. After populating the board it's almost working perfectly, but while poking around with the oscilloscope probe I happened to notice a lot of noise on the power traces.

I've attached some screenshots, the first (with the sawtooth waveform) is from the output side of the AP1117 3.3V regulator that I'm using, the second is from the input side of the regulator. I've got a 10uF ceramic capacitor (C1210X106K3RACTU) on the input of the regulator and a 22uF ceramic capacitor (1206YD226MAT2A) on the output of the regulator, plus a 1uF decoupling capacitor for the Zigbit module only 10-20mm away from the regulator. I'm supplying the regulator with 12V (reducing this makes no difference).

Before I start desoldering things and making a mess of the board, I wanted to check if anyone recognises these symptoms. Is this likely to be the regulator oscillating because of the ESR of the 22uF output capacitor and/or 1uF decoupling capacitor being too low?

Strangely, if I briefly short circuit the supply, the oscillation stops and doesn't come back.

Thanks


 

Offline senso

  • Frequent Contributor
  • **
  • Posts: 951
  • Country: pt
    • My AVR tutorials
Re: LDO Regulator Oscillation?
« Reply #1 on: January 25, 2014, 03:12:39 pm »
Shouldn't you have more capacitance on the input side and not on the output of the regulator?
 

Offline esar

  • Contributor
  • Posts: 24
Re: LDO Regulator Oscillation?
« Reply #2 on: January 25, 2014, 03:34:13 pm »
Shouldn't you have more capacitance on the input side and not on the output of the regulator?

I don't know, the 22uF and 10uF values were taken straight from the "Typical Application Circuit" section of the AP1117 datasheet.
 

Offline gibbled

  • Regular Contributor
  • *
  • Posts: 102
  • Country: ca
  • VE7mumble
Re: LDO Regulator Oscillation?
« Reply #3 on: January 25, 2014, 03:38:57 pm »
Does the datasheet for the 1117 suggest ceramic caps for input and output?   I suspect ceramics have too little esr for your regulator. 

The 1117's I have specify tantalums for input and output caps.

 

Offline SeanB

  • Super Contributor
  • ***
  • Posts: 15500
  • Country: za
Re: LDO Regulator Oscillation?
« Reply #4 on: January 25, 2014, 03:47:24 pm »
At 12V that 10uF ceramic is going to be a heck of a lot lower in capacitance as well, there is a thread around about that fact.

This in fact....

https://www.eevblog.com/forum/projects/t22291/

 

Offline esar

  • Contributor
  • Posts: 24
Re: LDO Regulator Oscillation?
« Reply #5 on: January 25, 2014, 03:54:46 pm »
Does the datasheet for the 1117 suggest ceramic caps for input and output?   I suspect ceramics have too little esr for your regulator. 

The 1117's I have specify tantalums for input and output caps.

Yes, it specifically mentions that ceramic capacitors can be used and that they both must have an ESR between 0.15R and 0.5R. I've not been able to find out the expected ESR of the capacitors I've used, the datasheets for them don't seem to mention it.

I guess if this does look like typical LDO oscillation then it can't be much else, but if I change the input and output caps to something with a higher ESR, won't the remaining 1uF decoupling cap still make the combined resistance too low? or doesn't it work like that?


Here's what the AP1117 datasheet says:

Input Capacitors
To ensure stable operation, the input supply must be low impedance up to a frequency of a few MHz. This requires a closely placed input decoupling capacitor of 4.7µF minimum. This can be either ceramic or solid tantalum. The ESR of this capacitor must be less than 0.5?.

Output Capacitor
The AP1117 also requires a closely placed output capacitor as part of the device frequency compensation. As part of its improved performance over industry standard 1117, the AP1117 is suitable for use with MLCC (Multi-Layer Ceramic Chip) capacitors. A minimum output capacitor of 4.7µF ceramic X7R or 4.7µF solid tantalum is required. Aluminum electrolytic can be used but a minimum of 47µF is required. When using Aluminum electrolytic it is still recommended to also use a 1µF MLCC in parallel. The ESR of the output capacitors must be less than 0.5?. The AP1117 is stable when using the correct value of MLCC capacitors. When using MLCC capacitors X7R dielectric is recommended. Do not use Y5V dielectrics.

Capacitor Track Length
Both input and output capacitors must be placed close to the AP1117. PCB traces not longer than 10mm are recommended between the AP1117 and the capacitors.
 

Offline esar

  • Contributor
  • Posts: 24
Re: LDO Regulator Oscillation?
« Reply #6 on: January 25, 2014, 04:04:14 pm »
At 12V that 10uF ceramic is going to be a heck of a lot lower in capacitance as well, there is a thread around about that fact.

This in fact....

https://www.eevblog.com/forum/projects/t22291/

Thanks for the link, the input cap is a 25V X5R (as is the output cap), looking at the graph they show of capacitance vs voltage for a 4.7uF 25V X5R cap, and assuming it behaves similarly, then I'd guess my 10uF cap is still way above the 4.7uF minimum recommended for the AP1117, maybe somewhere between 7 and 8uF?
 

Offline tjaeger

  • Regular Contributor
  • *
  • Posts: 101
Re: LDO Regulator Oscillation?
« Reply #7 on: January 25, 2014, 04:08:03 pm »
Yes, it specifically mentions that ceramic capacitors can be used and that they both must have an ESR between 0.15R and 0.5R. I've not been able to find out the expected ESR of the capacitors I've used, the datasheets for them don't seem to mention it.
http://www.kemet.com:8080/webspice/Ceramic/CeramicCapSelection.aspx

C1210C106K4RAC is the closest to yours I could find on the website, it has an ESR of 134 m? at 1 kHz and 4.2 m? at 1 MHz.  Capacitance is around 9 µF at 3.3 V.
« Last Edit: January 25, 2014, 04:10:06 pm by tjaeger »
 

Offline tjaeger

  • Regular Contributor
  • *
  • Posts: 101
Re: LDO Regulator Oscillation?
« Reply #8 on: January 25, 2014, 04:21:05 pm »
The datasheet seems a little sketchy.  They say ceramic capacitors can be used but then they specify a minimum ESR of 150 m?.  Also note that they want a minimum load of 5 mA.  There are definitely much better LDOs out there.
 

Offline esar

  • Contributor
  • Posts: 24
Re: LDO Regulator Oscillation?
« Reply #9 on: January 25, 2014, 04:28:29 pm »
Yes, it specifically mentions that ceramic capacitors can be used and that they both must have an ESR between 0.15R and 0.5R. I've not been able to find out the expected ESR of the capacitors I've used, the datasheets for them don't seem to mention it.
http://www.kemet.com:8080/webspice/Ceramic/CeramicCapSelection.aspx

C1210C106K4RAC is the closest to yours I could find on the website, it has an ESR of 134 m? at 1 kHz and 4.2 m? at 1 MHz.  Capacitance is around 9 µF at 3.3 V.

Excellent , thanks, that's exactly what I've been searching for, they hid that well!
It looks like the K3 is even lower still, so that confirms it, the ESR is far too low. Now I just need to see if I've got anything better to hand.
 

Offline sleemanj

  • Super Contributor
  • ***
  • Posts: 2689
  • Country: nz
  • Professional tightwad.
    • The electronics hobby components I sell.
Re: LDO Regulator Oscillation?
« Reply #10 on: January 26, 2014, 02:28:35 am »
It looks like the K3 is even lower still, so that confirms it, the ESR is far too low. Now I just need to see if I've got anything better to hand.

Shove a few tens of milliohm resistance in series with your output ceramic and see how that looks.  This is a common way to use ceramics instead of a recommended tantalum on the output of a LDO.

See this brief app note from TI on the subject:
  http://www.ti.com/lit/an/slva214a/slva214a.pdf
~~~
EEVBlog Members - get yourself 10% discount off all my electronic components for sale just use the Buy Direct links and use Coupon Code "eevblog" during checkout.  Shipping from New Zealand, international orders welcome :-)
 

Offline esar

  • Contributor
  • Posts: 24
Re: LDO Regulator Oscillation?
« Reply #11 on: January 26, 2014, 11:15:54 am »
Shove a few tens of milliohm resistance in series with your output ceramic and see how that looks.  This is a common way to use ceramics instead of a recommended tantalum on the output of a LDO.

See this brief app note from TI on the subject:
  http://www.ti.com/lit/an/slva214a/slva214a.pdf

Great tip, thanks. I don't have any other suitable caps to hand so I'd resigned myself to waiting until I needed to order some more parts, but I do have some 0.1R 0805 resistors. As the attached picture shows, it's not pretty, but it has fixed the problem and the trace on the 'scope is now nice and flat.
 

Offline miceuz

  • Frequent Contributor
  • **
  • Posts: 384
  • Country: lt
    • chirp - a soil moisture meter / plant watering alarm
Re: LDO Regulator Oscillation?
« Reply #12 on: January 26, 2014, 09:45:44 pm »
I've seen a post on mightyohms blog about the similar issue on a bus pirate board.
http://mightyohm.com/blog/2013/02/simple-fix-for-bus-pirate-power-supply-oscillations/

He has solved it just by slamming in the most crappy electrolytic he was able to find  :-DD


Share me

Digg  Facebook  SlashDot  Delicious  Technorati  Twitter  Google  Yahoo
Smf