Ripple Cancellation

This is the place for any magazine-related discussions that don't fit in any of the column discussion boards below.
User avatar
Sambuchi
Posts: 366
Joined: Tue Jan 18, 2005 1:01 am
Location: Orlando FL
Contact:

Ripple Cancellation

Post by Sambuchi » Wed Nov 04, 2009 9:33 am

I have a 5V power supply that draws .06A

There is a 15mV 2.5 MHz AC ripple on this supply voltage that I want to get rid of.

What can I do here?

User avatar
MrAl
Posts: 3862
Joined: Fri Jan 11, 2002 1:01 am
Location: NewJersey
Contact:

Re: Ripple Cancellation

Post by MrAl » Wed Nov 04, 2009 10:25 am

Hi,


Did you try putting a cap across the output, 0.01uf or 0.1uf or both?
That's the general idea.
LEDs vs Bulbs, LEDs are winning.

psycho
Posts: 388
Joined: Thu Jan 10, 2008 8:13 pm
Location: Northwest Indiana
Contact:

Re: Ripple Cancellation

Post by psycho » Wed Nov 04, 2009 10:42 am

Yes... This is SOP. Put a .1uF cap between the output and ground and it should clear it up. The most common values of caps are .1uF, .01uF and 1uF. If you plan on using IC's alot, get a bunch of these. It is a good idea to put a .1uF cap between EVERY IC's power supply pins. I.e. each chip will have it's own .1uF cap between it's VCC & GND as close (physically) to the chip as possible.

Hope it helps,
Kevin

User avatar
Sambuchi
Posts: 366
Joined: Tue Jan 18, 2005 1:01 am
Location: Orlando FL
Contact:

Re: Ripple Cancellation

Post by Sambuchi » Wed Nov 04, 2009 11:51 am

Yeah.... I've tried to use different value caps and no luck cleaning this up.

I found this nice site...
http://www.wa4dsy.net/filter/hp_lp_filter.html

I have some sensitive circuits on board that cant handle the output of this dc to dc converter.
and using an external power supply the board works great.

sghioto
Posts: 322
Joined: Wed Dec 26, 2007 5:16 pm
Location: Colorado
Contact:

Re: Ripple Cancellation

Post by sghioto » Wed Nov 04, 2009 1:15 pm

I would install a "pie" filter, 100 uh choke in series with a couple of .001 disc caps to ground on both sides of the choke.

Steve G.

User avatar
sofaspud
Posts: 531
Joined: Wed Feb 13, 2002 1:01 am
Location: San Antonio, TX
Contact:

Re: Ripple Cancellation

Post by sofaspud » Wed Nov 04, 2009 2:49 pm

2.5MHz ripple from a power supply? I don't recall ever running across anything like that. Are modern switchers
running that fast, or is this some "non-standard" app? The aforementioned LC filters and an LDO regulator with
good ripple rejection are the only options I'm aware of.

Robert Reed
Posts: 2276
Joined: Wed Nov 24, 2004 1:01 am
Location: ASHTABULA,OHIO
Contact:

Re: Ripple Cancellation

Post by Robert Reed » Wed Nov 04, 2009 5:34 pm

First thing is to determine where the ripple is coming from - the supply or feedback from an active circuit component. Then devise a plan of attack.

User avatar
MrAl
Posts: 3862
Joined: Fri Jan 11, 2002 1:01 am
Location: NewJersey
Contact:

Re: Ripple Cancellation

Post by MrAl » Wed Nov 04, 2009 9:09 pm

Hi again,


I have to agree with Robert where it would be a good idea to be sure where the problem is coming from.
Sometimes the load itself causes the problem.

Yes, modern regulators go higher in frequency than the old ones, but 2.5MHz does sound kinda high.

I's also have to ask if the addition of the cap(s) did ANYTHING at all, as in taking the ripple down
any amount at all.


I also agree that if caps alone do not solve the problem than a PI filter could be employed.
These kinds of filters knock the heck out of ripple and even a 2uH coil would probably suffice.
A 2uH air core coil could be wound by hand too, if you need construction details i can post some.
A cap before the coil and a cap after the coil can knock the ripple down 100 times. I've done this
on various supplies just to see what would happen in some cases and even with a switching
frequency of 300kHz (far below the 2.5MHz being discussed here) a small ferrite bead wrapped
with a few turns of thin wire (dont need too heavy for a short piece) creates just enough inductance
to be used with two caps and even at 300kHz it knocks the offending ripple down pretty low.
Maybe 0.1uf before the choke and 0.1uf after the choke.
LEDs vs Bulbs, LEDs are winning.

User avatar
Bob Scott
Posts: 1192
Joined: Wed Nov 20, 2002 1:01 am
Location: Vancouver, BC
Contact:

Re: Ripple Cancellation

Post by Bob Scott » Thu Nov 05, 2009 7:40 am

sofaspud wrote:2.5MHz ripple from a power supply? I don't recall ever running across anything like that. Are modern switchers running that fast, or is this some "non-standard" app? The aforementioned LC filters and an LDO regulator with good ripple rejection are the only options I'm aware of.
Sometimes the AC fuzz is not real. Sometimes the 'scope picks up HF AC if the measuring equipment grounding is wonky. Haha.. There's a word you don't see often, "wonky". :mrgreen: Make sure the clip lead on the 'scope probe is connected.
-=VA7KOR=- My solar system includes Pluto.

Robert Reed
Posts: 2276
Joined: Wed Nov 24, 2004 1:01 am
Location: ASHTABULA,OHIO
Contact:

Re: Ripple Cancellation

Post by Robert Reed » Thu Nov 05, 2009 9:26 am

The pi filters will work well, however if the ripple is being introduced by an active circuit- it is only that circuit that should be isolated with the filter from the main supply line, otherwise it would serve little purpose merely connected to the supply's output. In that instance,unwanted supply line coupling from stage to stage will still occur.

User avatar
haklesup
Posts: 3046
Joined: Thu Aug 01, 2002 1:01 am
Location: San Jose CA
Contact:

Re: Ripple Cancellation

Post by haklesup » Thu Nov 05, 2009 10:06 am

The most common values of caps are .1uF, .01uF and 1uF.
Those values are great for bypassing noise at the point of load either at the ICs or at the point of entry onto a board but at the output of a power supply you probably need a larger value cap. Start with a combination of a 470uF electrolytic and a 47uF Tantalum across the supply and ground. Both should be low ESR types. You may als need the smaller values for the higher frequency, they don't store much energy but they do have low impedance at higher frequencies.

2Mhz you see may be the superposition of several frequencies so that's why the range of cap values.

If you could figure the ripple current rather than the ripple voltage, its easier to select the optimal cap value. The charts in the Digikey catalog list ripple I as a parameter.

User avatar
Externet
Posts: 1862
Joined: Fri Feb 01, 2002 1:01 am
Location: Mideast USA
Contact:

Re: Ripple Cancellation

Post by Externet » Thu Nov 05, 2009 12:43 pm

As Bob says, I believe you are getting a false reading at the oscilloscope, from airborne signals, parasitic oscillation, leaky clocks somewhere around. Try another oscilloscope or probe.
15mV ripple should not affect anything powered by such supply anyway.
Miguel
- Abolish the deciBel ! -

User avatar
CeaSaR
Posts: 1760
Joined: Sat Nov 08, 2003 1:01 am
Location: Phoenixville, PA USA
Contact:

Re: Ripple Cancellation

Post by CeaSaR » Thu Nov 05, 2009 4:10 pm

Bob,

I just had "wonky" twice in 1 day. 1 from you and 1 from my niece!

Sambuchi,

2.5 MHz? Not power supply as far as I can tell. That would be some sort of distant harmonic, but would be
so much lower than the 2nd, 3rd or primary frequency. Look at something else in or out of circuit.

CeaSaR
Hey, what do I know?

Dean Huster
Posts: 1263
Joined: Wed Dec 05, 2001 1:01 am
Location: Harviell, MO (Poplar Bluff area)
Contact:

Re: Ripple Cancellation

Post by Dean Huster » Thu Nov 05, 2009 4:45 pm

Is the load running at that frequency such that what you're seeing is not supply ripple per se but load current variations?
Dean Huster, Electronics Curmudgeon
Contributing Editor emeritus, "Q & A", of the former "Poptronics" magazine (formerly "Popular Electronics" and "Electronics Now" magazines).

R.I.P.

User avatar
MrAl
Posts: 3862
Joined: Fri Jan 11, 2002 1:01 am
Location: NewJersey
Contact:

Re: Ripple Cancellation

Post by MrAl » Thu Nov 05, 2009 11:07 pm

Hi again,


Yes i have to agree that tracking down the *cause* of the ripple is the first thing to do.
That will lead to the best solution, if any is really needed at all.

If it is the scope probe then the solution is simply:

"Dont ship your circuit off with a scope probe attached to the output"

hee hee :smile:
LEDs vs Bulbs, LEDs are winning.

Post Reply

Who is online

Users browsing this forum: No registered users and 43 guests