Brushless/ESC radio interference.

[Simon Chaddock]

I have not got to the bottom of it yet but the Thumdrum 2813/14 and its 30a Mystery ESC on the Giant Dragon creates significant interference on 35MHz. when it is running but none when it is off.

Placing the aerial wire directly away from the motor restores full ground range even with the motor running.

Run the aerial close (3") to the motor  and all control is lost immediately it starts running.

It has quite long wires (14") from the ESC to the motor but short (4") ESC to battery.The ESC itself is quite close to the radio.

 

Its almost as if the motor and its wires are generating a radio signal.

 

Almost identical layouts in two other planes do not exhibit this problem.

 

I read a post not long back where someone was complaining that a conversion to brushless created radio interference.

The consensus of opinion in the replies suggested that as brushless were much less likely to create radio interference so the problem lay elsewhere.

 

I hope I can find out where! 

[Lee Smalley]

all electronic equipment emits rf of some form some vastly higher than other esc can generate large amount of RF noise some makes are better than others traditionally brushed moters create more noise than brushless but again this is dependent on build and component quality, this is why 2.4 is a god send to electric fliers as the rf signal is high enough in the band to be pretty much immune to most rf noise created by electronic equipment, try a different esc (better make) and you may see a dramatic improvement

[Martin Harris - Moderator]

Do you have a switching BEC in the new setup?  These can generate a lot of high frequency "noise". A lot of ESCs come with these incorporated nowadays as they are more efficient than the old linear devices but look identical.

Edited By Martin Harris on 03/11/2010 12:36:40

[Colin Naylar]

Incorporating a ferrite ring on the ESC to Rx lead often almost eliminates this interference - I got mine from Giant Cod. Wind about 4 or 5 turns of the cable onto the ring. It enabled me to finally fly my Seagull Decathlon EP after horrendous servo jittering.

[Erfolg]

If you know the problem source and you seem to know the solution. That is by re-routing the aerial, why not make it a permanent solution?

[Simon Chaddock]

Thanks for the replies.

What surprises me is the interference only appears with the motor actually turning  When the motor is off the ground range is equal to my other set ups with no servo jitter even at extreme range, they just stop responding.

 

It is of course possible the ESC itself is to blame but  the same make/type have given no trouble before. A bit of bind to change but I do have another.

 

In the meantime I have moved everything as far apart as possible and re-routed the aerial. 

[Biggles' Elder Brother - Moderator]

If you have a noisy bearing in the motor this can generate rf through moving metal-to-metal contact. That might explain why you only get the interferance when the motor is running.

 

BEB

[Lee Smalley]

the ferrite ring only stops rf noise coupling onto the wire running to the rx if the noise is being emitted and being picked up by the board in the rx or the servos  and their leads then this will be of no help but it is worth a try 

[Tim Mackey]

Posted by Biggles' Elder Brother on 03/11/2010 14:05:16:

If you have a noisy bearing in the motor this can generate rf through moving metal-to-metal contact. That might explain why you only get the interferance when the motor is running.

 

BEB

 

Or that its simply radiated noise caused by the brushes /commutator. It doesnt surprise me at all TBH , and either the motor or the ESC or indeed both could be causing sufficient "noise" to upset the Rx. I tried all sorts of things in an early electric model of mine many years ago, and could never completely stop glitching - even the best DC receiver I could afford at the time suffered. Years later, on one of my brushless EDFs I had simiular issues, no amount of ferrite rings, top class ESCs and schulze receivers etc cured it.

My only salvation was the day I fitted 2.4Ghz radio.

[Tim Mackey]

Posted by Lee Smalley on 03/11/2010 14:22:54:
the ferrite ring only stops rf noise coupling onto the wire running to the rx if the noise is being emitted and being picked up by the board in the rx or the servos  and their leads then this will be of no help but it is worth a try 

 I never found much joy using these things TBH - if it is tried, then its important to both wind as many turns of the lead as possible through the ring ( probably need to remove the plug to achieve this ) and fit the rings as close as poss to the Rx itself. That way, if noise is being transmitted along servo wires, it will be damped out ( hopefully ) just before it enters the rx. Fitting them further up the wire could allow the interference to reappear lower down, and still enter the rf board.

[Erfolg]

If rerouting the aerial prevents the glitching and provides a full range, why not do this?

 

Surely putting ferrite rings onto leads is no more a solution than re-routing?

 

I can understand the desire to eliminate the source of the issue, yet it seems identification of the precise source and eliminating the arisings is improbable, particularly without scientific instruments.

 

My own experiences, as a manager, rather than a engineer/technician, is that in industry, it is not unusual for interference or potential for interfereancefor of systems to be addressed, by shielding, routing cables to crossing each other at 90 degrees (power relative to instrument routes), by physical separation etc. Often elimination of the cause not being practical.

 

That is my two pennies worth.

[Lee Smalley]

rf noise can couple on to almost anything, if  the copper tracks on a pcb board are of the right length (often a harmonic of the rf noise frequency) then they can couple onto the pcb directly (thats why good pcbs have as short a track runs as possible) so often ferrites don't solve the problem i did solve one jittery servo by putting one as close to the servo as possible and re-routing will not always solve the issue either if the rf freqency is of just the right freq then no matter where you put the aerial in the aircraft it can cause problems contray to what people believe 2.4 does not make you immune to rf noise, it just means your radio link from ground to aircraft is pretty bomb proof but if the rf noise is effecting the rx directly or coupling on to wires then it does not matter if you are 2.4 gig or not

2.4 gigidy just gives us one less thing to worry about

[Plummet]

The amount of energy picked up from the radio waves by an aerial is ridiculously tiny.  This has to be amplified. and the bits at the right frequency extracted.

 

Any motor will generate possible interference.  The switching by the commutator in a brushed motor will make sparks, which create radio waves.  A brushless motor works by generating a changing magnetic field.  The wires to and from a motor will also generate fields.  Any changing magnetic field will induce a voltage in a conductor it encounters.

So if the aerial goes anywhere near an electric motor - and remember, servos contain an electric motor - there is the risk that it will pick up interference.  If the amplitude of the interference is greater than the signal then the signal may be swamped.  There is the risk that the interference will totally overload the RXs amplifiers, in which case all signal may well be lost.

 

Keep aerials away from other stuff.

 

Plummet

[Erfolg]

Lee

 

I would have thought that these days of autorouters and design criteria driven CAD, design of  PCB's, that many of the issues you mentioned would have been eliminated?

 

I do remember as a youth working (for a period in training) on Heavy Industrial Transformer testing, that the world was full of signals, with fine Fuse wire type hangings in addition to the permanent caging, to try and eliminate spurious signals. As you say the world was a noisy place in the late 60's, I would guess noise has not diminished.

[Simon Chaddock]

I finally solved the interference problem.

 

It was the combination of the RX crystal and  the TX.

 

Using the same channel (66), the same RX and crystal but a different TX and crystal gave no range  problem at all, whether or not the motor was running.

 

Although the old TX works fine with other RXs and other channel 66 crystals I guess the manufacturing tolerances on this one were just over the limit.

 

Moral - use matched crystals or synthesised RXs or 2.4G!

 

 

 

[Lee Smalley]

Erfolg, no mate we still have our product tested for suceptability, emmissions all the time and you would be amazed at how many times they fall down at certain frequencies all our stuff is designed on CAD and the guys know what to look for and things to avoid when designing, but they still fall down and every time they do it means loads more testing for me !! if anything rf noise has increased soo much over the years due to the expansion of electronics in every day life and products that are made in the far east are the worst culprits, they do not always pay the same attention to emmissions as we have to in this country