5
\$\begingroup\$

I am looking to clean up the power supplied to a video transmitter on the UAV I am building. The video transmitter consumes about 200mA of current at 16V DC. The 16V line has noise from motors and other electronics in it. I am using a capacitor to smooth out the current. In the past I would also use a ferrite ring and loop the power cable through it 10 times, and that would really clean up the signal. Now, since I am making a custom PCB for it, I wanted to use a surface mount component that would do the same thing.

Would something like this work?

\$\endgroup\$
  • \$\begingroup\$ Add lots of ceramic capacitors really close to both the motor drivers (if possible) and near the camera and video transmitter. Digikey do have some small smd common mode chokes somewhere, I think in the "coupled inductors" section, one of those would help a lot as well \$\endgroup\$ – Sam May 1 '16 at 0:14
  • \$\begingroup\$ Tom, are ceramic capacitors better than standard electrolytic capacitors? The standard once are available in larger capacitance values. \$\endgroup\$ – Bogdan May 1 '16 at 16:39
  • \$\begingroup\$ For blocking high frequency noise, oh yeah. at a MHz or so, a 1000uF electrolytic could have an impedance of several ohms (their spiral wound shape gives them high series inductance) while even a tiny 0603 1uF ceramic capacitor could be a few milliohms. Electros are good for bulk storage, i.e. supplying big chunks of power for things like sudden acceleration (relativley slow, but high energy events). Ceramics are for when you need to absorb really high frequency (but small energy) events, some samsung ceramics still only have an ohm of impedance at a GHz or so. \$\endgroup\$ – Sam May 1 '16 at 22:48
1
\$\begingroup\$

A ferrite ring with maybe ten loops (both send and return) will be a pretty good common mode choke i.e. any higher frequency AC noise that occurs on both wires will not easily pass through: -

enter image description here

As you can see in the above attenuation graph, common-mode signals will get much more heavily attenuated than differential noise.

That's the back story to what I believe you have previously used.

The device you propose is a different kettle of fish - it is a single coil and realistically it falls into the category of a ferrite bead. It does nothing to change the common mode signals associated with your power feed because it is a single series device that acts to block HF stuff on one wire only. It's probably not going to be a great deal of use either because it's attenuation impedance at (say) 1 MHz is a few ohms, maybe 10 ohms maximum.

It isn't a common mode choke and it isn't really useful for "blocking" until maybe 10 MHz - see the graphs in the data sheet.

So basically there's no comparison and no great advice to be given - the two devices do different things BUT if the combination of common-mode and differential noise with no protection causes problems then adding this ferrite bead may be enough but it's difficult to tell.

Of course you can buy PCB fittable CM chokes so maybe I'd consider that as the best option.

\$\endgroup\$
1
\$\begingroup\$

It could, potentially, depending on the nature of your noise. You will want to select a ferrite bead which has a high impedance at the frequency that you want to reduce. Read the datasheet(s) and make sure that the impedance curve looks appropriate.

Without knowing more about the noise sources present, the linked device will probably help, but it may not be optimal. Motors produce noise which is very wide in bandwidth, so that bead will certainly knock out some of it, but it may not be quite as effective as a large ferrite ring with multiple loops.

By appropriately decoupling all of your ICs, and isolating the motors are much as possible from the rest of the supply voltages, you should be able to cut down on noise quite well. If you have long supply runs on your board, you may want multiple ferrite beads near each load to reduce motor noise that couples into the supply rails.

\$\endgroup\$
0
\$\begingroup\$

Probably you should employ two inductances with higher value but lower current in series with two capacitors in a p filter to ground because probably you have a ground plane and noise should be referenced to it near to the video transmitter. (http://www.digikey.com/product-detail/en/taiyo-yuden/FBMH1608HL601-T/587-1737-1-ND/1147062)

\$\endgroup\$

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.