EEVblog Electronics Community Forum
Electronics => Projects, Designs, and Technical Stuff => Topic started by: gascencio on November 10, 2022, 10:13:29 am
-
Hi everybody,
I'm doing conducted emissions test on a product.
We are failing on the 20-40 MHz band when doing LF emission. We used a CM - DM splitter to determine that the source problem is CM noise.
When trying different filters/ configurations i realized that putting a SMD ferrite (782853152) in series with the DC + input makes no difference. That seems normal as this a DM filter and we have a CM problem
What I find strange is that when putting a ferrite bead ( 2661021801) on the DC + input cable, i see significant attenuation.
Why is so? SMD ferrite should be way more efficient as it's impedance is higher than the ferrite bead at that frequency band
[attachimg=1]
Thank you very much
-
The cable ferrite bead is physically large. It contains a large amount of ferrite material, so logically, it takes a lot more magnetizing force (i.e. current) to saturate it. The SMD ferrite bead contains a lot less ferrite material, so it generally takes a lot less current to saturate it. I would say that's the main reason, although depends on the current you're actually passing through it.
-
Sometimes it's not enough to have a ferrite bead to block the conducted emissions. You also need a path you want the emissions to take. For example, a cable usually has a shield or housing that is grounded to a chassis, so that the conducted emissions are both blocked by the ferrite bead and shunted to the chassis.
If you have a ferrite bead at one end of a common-mode path, and the other end open, what has been created is a dipole antenna and it will radiate anyways. Usually this is indicative of some unbalance of currents, for example unbalance driving a differential pair, or a RF signal being conducted without a return current closely routed to minimize the loop area. This is especially possible if there is a break in a ground plane.
-
although depends on the current you're actually passing through it.
I'm passing less than 100mA through the cable, the smd ferrite is given for 700mA, so I'm still far from saturation
-
although depends on the current you're actually passing through it.
I'm passing less than 100mA through the cable, the smd ferrite is given for 700mA, so I'm still far from saturation
Rating is thermal maximum. Saturation is almost never rated in datasheet. Classic catch for young players. Good initial assumption for saturation is one tenth of rating, but you really have to measure the part to be sure.
Small SMD ferrite beads are good for supplying ADCs, voltage references etc. where current is in range of a few mA.
-
For the power line, the WE-MPSB is a better choice. Also, maybe you need common mode suppression, not differential.
Since you are already talking with Würth, drop them an email, they have very good support.