Shielding servo leads? Suggestions...

[Foxy]

A fellow flier is having great troubles with glitching in a brushless powered Crazy 8. Specs are: mega 16/15/3 3:1 propped to around 30amps, Hacker Master ESC, Schulze 835w rx on 36mhz, 8 CBP1800 cells, Medusa optoisolator. The glitching has been eliminated from the throttle channel, but the servos are still occasionally glitching very badly, often resulting in the aircraft rolling 180deg. I have a suspicion that the interference is entering the servos via the leads rather than being passed on via the RX, as its a filterered RX and the throttle channel is fine. I'm considering shielding the servo leads for him to try and stop the glitching, do any of the "learned ones" out there think this will be effective? Any other suggestions? I may also put some ferrite rings on the servo leads.
Any help much appreciated.

Adam

[Comatose]

Well, I don't think it'll be effective, but its easy enough to try. Strip the ground lead of the servo a bit without severing it. Maybe a quarter of an inch. Get a strip of aluminum foil, an inch long and the length of your servo lead. Cut a tab a quarter of an inch wide into the strip and wrap that around the exposed ground lead, then wrap the entire strip around the servo lead. Cover that in heat shrink tubing and shrink.

Aluminum foil is great for testing whether shielding is necessary/effective.

I'm pretty sure that that isn't the problem, though. The throttle will be a bit more glitch resistant than the servos, because the ESC has software to ignore errant pulses. You don't say whether your servos are analog or digital, but unless they're digital your servos do not. First thing I'd try is a separate RX pack for one flight. You've probably got one that came with a radio at some point, and never used. I know I do anyway. (=

Or, if you have a scope, look at your servo signals on your scope. That'd clear it up really fast.

[KillerWatt]

probably just adequate red and black wire twisting the servos power leads does enough stray radiated signal canellation......A misused or misapplied wire harness "ground"shield can actually cause more problems depending on it's real RF impedance to the commom system neutral .....The swisted power conductors just want to cancell the noise out......... kv

[Foxy]

Thanks for the quick response guys. The servos are hitec analogs. The reason I suspect the servos/leads is that the RX is a DSP type, it supposedly inspects and discards/corrects bad signals sent to the servos. He is already using a seperate RX pack and an optoisolator, so we are running out of options! I have waved my rf scanner around the setup and had a listen on the rx freq, there is a LOT of noise being generated by the motor at around 3/4 throttle, most of it eminating from RX and the end of the batt pack.
Good suggestion re the cro, it has been tucked away for a long time, almost forgot I had one. I'll see what it shows up. Apparently a lot of fliers on 36mhz are having similar issues, Aircraft-world.com was offering a sum of cash to anyone with solution.
I too was a little concerned about the shielding making things even worse, I know RF can be an odd one to deal with. If so i'll go with the twisting and also try a toriod with as many turns as I can get through it.

Thanks again
Adam

[tve]

I know, probably a dumb question, but I find it hard to believe the problem is with a brushless motor emitting RF. I would suspect a power supply problem first....
Are you seeing noise on your spectrum analyzer when the motor is not running? In other words, is the problem the transmitter?

[Foxy]

There are no dumb questions tve, only judgemental people ;) The motor is definately emitting rf. The tests with the scanner were done with a direct connection to the rx via a trainer lead, ie no tx running. The noise comes from the coils in the motor being switched on and off rapidly.

Thanks
Adam

[vintage1]

I'd say inspect the installation, and move the ESC well away from the receiver.

ESC's emit a lot of RF a does the motor. Short motor and battery wires help, and keep ALL that lot away from the RX.

Not sure what a medusa optoisolator is, but it sounds suspicious - can you describe the way things are hooked up a little more explicitly?

Are you not running of the normal BEC?

[Foxy]

The optoisolator is basically that, a circuit made by Medusa Products that allows you to turn a BEC ESC into a truely opto isolated non bec unit. If you check out the website it may be clearer. We are using a seperate rx pack. This has been an ongoing problem, I have posted about it in the past, we thought the opto had cleared it up but we were wrong. The ESC is as far away from the rx as possible. If the donuts on the servo wires don't work I'm not sure what we will do.
Thanks for all the suggestions.

Adam

[olmod]

this is a long shot but i had some experience building a large faraday cage ,i wonder if something like that in minature would snuff it out at the scource.cheers. :)

[DangerBird]

How about a few ferrite beads?

[Foxy]

I'm going to give the ferrite rings a go soon.

Thanks
Adam

[Comatose]

I had a thought. Are you sure that the glitching is related to the servos and not something in the radio signal? For example, it could be a cracked crystal, a broken antenna wire, something along those lines. Have you tried the same setup with a different reciever and/or transmitter?

[Mr.RC-CAM]

Who's DSP Rx is it? FWIW, just because it is "DSP" it does not mean it won't glitch. Unless it is a recent Berg design. :)

Besides all the fun with the chokes, beads, twisted pairs, faraday cages, shields, supply filters, and such, I suggest you try another Rx brand too. You never know...

RC-CAM