ferrite bead resonance

[Internauta]

hello,
is there a formula to calculate the Q  of an LC filter made with a series ferrite bead + parallel capacitor to ground?

I am trying to add some RF filtering to an AC/AC 15V wall wart output before hitting the rectifier diodes in my linear power supply, but I fear to create more troubles than improvements if I miss the right inductance..

here is attached a schematic from Rane with a very similar circuit. How can I select the proper ferrite ?
Could be more effective the use of a common mode choke instead of the ferrites in this case?

thanks!!

 

[PRR]

Ferrite beads sold for RFI reduction "usually" have so much internal resistance that any resonance is weak.

Why do you think you need RFI reduction?

 

[Internauta]

HI PRR, do you think it's not needed with this type of power supply?

 

[abbey road d enfer]

here is attached a schematic from Rane with a very similar circuit. How can I select the proper ferrite ?
Could be more effective the use of a common mode choke instead of the ferrites in this case?

thanks!!

This is weird. I would think the leakage inductance of the transformer would completely dominate over the ferrite beads. Now the cable between the transformer and the PSU acts like an antenna, but again its impedance would dominate.
Seems to me like obsessive precaution...

 

[Internauta]

OK, thanks for the answer!! 
Is C22 helpful in some way or is it weird/not needed as well?
cheers

 

[abbey road d enfer]

OK, thanks for the answer!! 
Is C22 helpful in some way or is it weird/not needed as well?
cheers

You may leave it here, can do no harm a,d cost next to nothing. just make sure it has the proper voltage rating for worst case (+20% mains).

 

[Internauta]

OK for C22 !
thanks again

 

[PRR]

> I would think the leakage inductance of the transformer would completely dominate over the ferrite beads.

I suspect it is a wall-wart, 3 or 6 feet away, and that RANE is obsessing about the FCC digital-device rules. Any rapid switching inside the box will leak onto the several-foot cable to the wall-wart and be radiated. Crap-up our over-the-air TV signals.

I'm re-reading Popular Electronics around 1980. The rise of video games and then computers forced a lot of bureaucracy around incidentally radiated signals. 

 

[Internauta]

> Any rapid switching inside the box will leak onto the several-foot cable to the wall-wart and be radiated. Crap-up our over-the-air TV signals.

Seems a good reason to have a filter in there isn't it??

In order to make some test, would make sense if I connect a signal generator in place of the wall wart and check with an oscilloscope how much of the input signal (from a few hundred Hz to several Mhz) finds its way to the rectifier diodes to check "immunity" ?
and then,  in order to check for unintentional radiated signals, insert the above signal in the OUT of the voltage regulators and see with the scope how much signal leaks out of the 15VAC connector of my box?
 

 

[abbey road d enfer]

> Any rapid switching inside the box will leak onto the several-foot cable to the wall-wart and be radiated. Crap-up our over-the-air TV signals.

Seems a good reason to have a filter in there isn't it??

In order to make some test, would make sense if I connect a signal generator in place of the wall wart and check with an oscilloscope how much of the input signal (from a few hundred Hz to several Mhz) finds its way to the rectifier diodes to check "immunity" ?

You would need to insert a complex impedance emulating that of of the transformer. Then you would have to distinguish between Common-Mode and differential-mode noise, and drive the cable acordingly.

and then,  in order to check for unintentional radiated signals, insert the above signal in the OUT of the voltage regulators and see with the scope how much signal leaks out of the 15VAC connector of my box?

The issue would be the same. In doing that, you would generate only diff-mode noise, which is almost a non-issue. Common-mode noise would be dominant.

I don't know what this Rane product is and what your application is, but if it's linear audio, I wouldn't pull my hair over this.

 

[Internauta]

thanks  Abbey,
the Rane application is a DJ mixer, without internal digital FX, while mine is a guitar preamp with just analog stuff as well.
It does feature an LFO, which is powered with its own + - and ground, which return to the power supply regulators without sharing the path with other circuits.

It would be great to insulate my preamp from cell phones noise for example, so I was trying to find out some way to mitigate this kind of problems.
what I did at the moment is filtering the guitar input with a ferrite bead in series with guitar ground (insulated from chassis) and an RC filter on the signal conductor.
Out (which is impedance balanced) has ceramic caps across the output 2 conductors to ground,
soldered directly on the TRS connector, which is also used to bring the circuit board ground to the metal chassis.

the third and last connection to the outside world is the AC input jack, which is the one I really don't know how to make immune from RF ...
thanks again for your help

 

[abbey road d enfer]

the third and last connection to the outside world is the AC input jack, which is the one I really don't know how to make immune from RF ...
thanks again for your help

If I noticed issues and I could trace them to RFI/EMI getting into the box via the AC cable, I would put caps to chassis and perhaps inductors, but not ferrite beads, I would rather use RF chokes. But it seems overkill to me.

 

[Speedskater]

Jim Brown, past AES committee chair on EMI/RFI has several papers & Power Points on ferrites:

"RFI, Ferrites, and Common Mode Chokes For Hams"
http://www.audiosystemsgroup.com/RFI-Ham.pdf

"Measured Data For HF Ferrite Chokes"
http://www.audiosystemsgroup.com/FerriteDataHF.pdf

"Understanding How Ferrites Can Prevent and Eliminate RF Interference to Audio Systems"
http://www.audiosystemsgroup.com/SAC0305Ferrites.pdf

About 50 more  Jim Brown papers & Power Points:
http://www.audiosystemsgroup.com/publish.htm

 

[abbey road d enfer]

One must remember that ferrite bead inductors are series elements, and as such provide rejection only if there is a complementary shunt element. That shunt element can be the input impedance of the receiver (typical of RF applications), but for audio systems the input impedance does not provide the shunt effect.
Typically the ferrite bead inductor should be complemented by a capacitor, but this capacitor may be too large for medium and high-impedance applications.

 

[ricardo]

If I noticed issues and I could trace them to RFI/EMI getting into the box via the AC cable, I would put caps to chassis and perhaps inductors, but not ferrite beads, I would rather use RF chokes. But it seems overkill to me.

The problem with RF chokes is they are much higher Q than beads so depending on the rest of your circuit, you may need to have Zobels & othercomplicated  stuff to damp the high Q resonances.

Beads normally have LoQ at the RF frequencies of interest so are much easier to use than RF chokes.

But EMI protection is a LOT more than adding chokes & caps.  The problem is a 3D mechanical one too.  Your EXACT earthing, grounding and your box and how its constructed have major influence.  Some of it is covered by Jim Brown but probably not all. eg the beads & caps have to be AT where the wires enter the box via the SHORTEST POSSIBLE LEADS.  Two inches of cable make the beads, chokes & caps useless for EMI rejection 

Today, a mobile phone is probably the worst offender and easiest to test.  There's loadsa VERY expensive pro (?!!) gear which will announce to all & sundry that you've received a text message.

If your stuff doesn't misbehave when a phone receives a message when place on top of it, you are probably OK.  Some studios ban all mobile phones cos of this.

It's worse in areas of poor reception cos the phone will up its power to compensate.

In da old days, it was thyristor theatre lighting.