wire twisting and ferrite rings.

[danhampson]

I've been looking into increasing range and quality of transmitted video signal and came across an article that talks about the use of ferrite rings and wire twisting to reduce noise on the aircraft.

My questions are which is best ferrite ring or cable twisting?

My second question is even though I have the pos and neg wire disconnected on 3 of the esc ' s would the cables still need to be twisted or a ring added?

[Graham Lawrie]

I am no expert Dan, and not aware of the cable twisting effects:) but the simplest filter for high frequency interference is the ferrite ring, which is also available in a clamp version to go around cables without wrapping cable around it:)

[pvolcko]

Twisting helps eliminate common mode noise. The noise will hit one side of the twist creating a certain voltage and then the other side of the twist creating an opposing voltage, cancelling out.

So the theory goes.

There is a downside, the twisting also couples the twisted lines together. This can result in voltage biases on the signal line and the signal line fluctuating voltage can create what appears to be noise on the DC lines. Normally twisted lines are done only in pairs and only with differential signals (one line has the positive version of a signal and the other has the negative version, both based on a common reference voltage), this way signals do not induce noise on the twisted partner line and simple common mode rejection opamp circuits on each end have a strong chance of eliminating any picked up noise that is not part of the transmitted signal.

Ferrite rings will help eliminate high frequency noise, regardless of mode.

The two methods are not mutually exclusive.

Personally, I don't like twisting or braiding ESC and servo and A/V lines. The rejection is highly dependent on a proper twist, and because the lines are carrying different signals (power and actual signal) the coupling effects are more pronounced than the noise rejection benefits. I'd stick with ferrite rings and chokes for these kind of multi-signal lines. Try to keep high voltage/high current circuits away from low level signal voltages. You can wrap lines in conductive braiding and tie to negative/ground to help eliminate stray signal interference too, if you have a problem with noise on video/audio transmissions (this is where coaxial wiring schemes can really help).

[danhampson]

I think as it's on a mini quad and space is at a premium I will try the braid method first and see how the video quality and range are to start with.

Do the none connected wires from the other 3 esc's need to be isolated.

Also would it be worth insulating the wires with copper tape?

[pvolcko]

Copper tape is the method I mentioned last (conductive braiding, maybe should have said sheathing). You need to ground the copper in order for it to have any benefit (tie to negative terminal). It will keep out outside interference (and step signals penetrating out), but it will do nothing if the interference is due to a signal line inside the sheathing.

Are you actually having a problem you're trying to fix? Or prevent a problem from showing up? If the later, I'd suggest trying without anything more than careful routing of wires. No reason to fix something that isn't a problem.

As for the unconnected wires, you can twist/braid them, it won't hurt, but you likely will get little benefit from it.

[Lee]

I unsoldered the pos/neg wires from the 3 escs and removed them. Problem solved

[danhampson]

Are you actually having a problem you're trying to fix? Or prevent a problem from showing up? If the later, I'd suggest trying without anything more than careful routing of wires. No reason to fix something that isn't a problem.

I haven't got the video gear you it is more of a preventative measure to be honest.

I unsoldered the pos/neg wires from the 3 escs and removed them. Problem solved

I'm not that confident at soldering small stuff to be honest lee. So the shielding and braiding is the way in going to have to go.

[pvolcko]

If you don't have vtx or cameras mounted then I doubt you'll need any of this.

[danhampson]

I'm ordering the fpv gear on Friday so it should be with me on Monday and installed pretty much same day.

I've found this stuff do you think it would do the trick?
http://www.easycomposites.co.uk/carbon- ... e-5mm.aspx

[Stambo]

Is this for your Miniquad
Just cut the pos and neg wires from the ESC servo leads nice and close to the ESC and maybe put heatshrink on the small bit that's left.
You are limited in range on a mini by the size of battery you can carry.

[danhampson]

Yeah its for my mini quad. And it's a bit of an experiment. I'm planning another larger quad build (more of a long range machine) and looking into ways to get the best out of the equipment.

[danhampson]

Cheers for the advice guys I've done a bit of work on it this morning and done as Stambo suggested and trimmed the pos and neg on 3 of the esc's and then braided the third. I've had to change the orientation of the cc3d as after I braided the cable it no longer reached the pins. I'm going to do the computer bit later here's a shot of how it looks now.

20150108_065229.jpg (1.94 MiB) Viewed 3075 times

[pvolcko]

Hmmm... why did you cut the negatives to the three ESCs? It may not matter, but generally you only cut the positive power lines to the extra ESCs and leave the negatives attached to provide solid common grounding to all the electronics. You may find you need to reattach those if things behave squirrely.

[danhampson]

I removed the negative as lee said he desolders his and Stambo trims his down, I thought that it makes a much neater, it left enough so that they can be reattached if need.

[Derek]

No doubt, it is cleaner that way, but I've not heard about anyone removing or cutting the black wires either. I have a few planes that run an external bec and even on those I've just pulled the red wire from the servo connector and simply folded it back. On the other hand, both Lee and Stambo are much more experienced with multi rotors than I am so their suggestions may just work. I'd be interested to hear your results.

[danhampson]

im going to try to redo the programming on the cc3d in the hour or so and if there is enough light ill give it a test hover in the garden and let you know the results

[Derek]

Sounds good buddy! Good luck!!!

[danhampson]

I've done the computer bit, well as much as can do testing in the dark lol. Everything went fine no unexpected behaviour, i have just got to set the accelerometer bias (a bit of front right drift) which will have to wait until tomorrow when theres some daylight and i can see more.