| Author |
Message |
Analog_starter
Joined: 15 Nov 2004 Posts: 118
|
31 Dec 2004 7:20 About regulator ripple |
|
|
|
Hi all,
I'v built a LDO voltage regulator circuit, 3.3V input, and 1.5V output. But there is very large ripple in the output voltage. How can I suppress this ripple?
Thanks & Best Regards
Analog_starter
|
|
| Back to top |
|
 |
IanP
Joined: 05 Oct 2004 Posts: 6347 Helped: 1505 Location: West Coast
|
31 Dec 2004 7:23 Re: About regulator ripple |
|
|
|
| You can try to connect smoothing capacitor..100µF-1000µF..depending on the load.
|
|
| Back to top |
|
 |
Tornado
Joined: 26 Apr 2002 Posts: 340 Helped: 1
|
31 Dec 2004 7:32 Re: About regulator ripple |
|
|
|
What frequency the ripple has? If is 100-120Hz then your input voltage is too
low do not add mutch capacitance on the output it may put the regulator in
overcurrent protection, around 10uF it's OK, just follow National Semiconductor and other manufacturers recomendations. Add capacitance on the input after your full wave rectifier. The minimum input - output delta is 3V in normal regs and about
1 volt and less on the low dropout types. I preffer the low drop types.
Tornado
|
|
| Back to top |
|
 |
Analog_starter
Joined: 15 Nov 2004 Posts: 118
|
31 Dec 2004 7:56 Re: About regulator ripple |
|
|
|
Thank you all.
Are these capacitors on chip? Or external load? This regulator targets as a SOC IP, so should I add so huge capacitance in my circuit?
And the ripple frequence is about 5Mhz.
What does it stand for:
"Add capacitance on the input after your full wave rectifier"
Thanks & Best Regards
Analog_starter
|
|
| Back to top |
|
 |
IanP
Joined: 05 Oct 2004 Posts: 6347 Helped: 1505 Location: West Coast
|
31 Dec 2004 8:03 Re: About regulator ripple |
|
|
|
Because the ripple frequency is around 5MHz you will not need to add any capacitor befor this regulator.
Also big capacitor (electrolytic) will not solve this problem. It will help, but you will have to add ceramic 100nF and maybe 10µF tantalum to cover more frequecies.
What may help is small (1µH) serial inductor and then these capacitors. It will be like otput low-pass filter.
|
|
| Back to top |
|
 |
andy1
Joined: 24 Jul 2004 Posts: 124 Helped: 2
|
31 Dec 2004 8:41 Re: About regulator ripple |
|
|
|
The best way to find out is to hook it up to spectrum analyzer and look for all the noise spectrum since the ripple may not be at just one frequency. There might be harmonics that exists at higher than 5mhz.
Once you find out about the noise spectrum, then add caps acordingly.
|
|
| Back to top |
|
 |
qutang
Joined: 25 Oct 2004 Posts: 315 Helped: 6
|
04 Jan 2005 14:05 About regulator ripple |
|
|
|
| ripple frequency input is usually at 100-120Hz
|
|
| Back to top |
|
 |
VVV
Joined: 26 Nov 2004 Posts: 1492 Helped: 278
|
04 Jan 2005 18:43 Re: About regulator ripple |
|
|
|
This sounds like oscillation to me. LDO's can oscillate more easily than regular linear regulators. Try changing the output capacitor, or adding a compensation capacitor/ network.
For more info, please read this document: www.national.com/an/AN/AN-1148.pdf
|
|
| Back to top |
|
 |
dreamteam
Joined: 05 Jan 2005 Posts: 13 Helped: 1 Location: France
|
05 Jan 2005 14:03 About regulator ripple |
|
|
|
| Agree with VVV, be aware of your LDO stability and check your phase margin. What decoupling capacitor do you use ? Is the value enough ?
|
|
| Back to top |
|
 |
staric
Joined: 26 Sep 2003 Posts: 79 Helped: 1
|
05 Jan 2005 15:12 About regulator ripple |
|
|
|
| Should be consider the stability. It should be closed loop. The phase margin is very important.
|
|
| Back to top |
|
 |