EMI from DLE 30cc and DX7 with AR7000 RX

[gazzazh]

Today after the last problem with my Sparrowhawk ( the cursed Plane) i went to th filed to test it out and found another problem ... THe servos are twitihing like never before..

 

The Ignition, ignition battery and RX battery is all located at the cowl. The RX and Satellite rx is located at the rudder servo.. which is far from the ignition.. I have not idea what to do.. i am totally lost here .. any suggestions.. I tried putting ferret rings on the throttle and but still nothing.. also when i start the engine the plane the rx goes into burnout.. as soon as i start the engine and everything starts twitching..

 

Please help.. ...

 

 

[Lee Smalley]

wrap the ignition unit tightly in tin foil and try again, on the ground, shiny side in

[gazzazh]

HI Lee,

 

Do you mean having the shinny side facing the Ignition.... and should i then wrap with other material over the tin foil..

 

Thanks a million lee, you are always a great help.. Cheers

 

[Peter Beeney]

gazzazh,

A mate had this exact problem a while back with 2.4 GHz radio and an ignition system on a petrol engine, it’s also been reported by others elsewhere. We found that shielding, and that include a common wire connection from all the shielding and components to the negative batteries had little or no effect. However, this is of course not to say that it won’t be a perfect cure on your model. We concluded that the interference was being induced on the power supply wires from the receiver battery by the ignition pulses on the ignition battery wires. This seemed to prove there is very little decoupling on the 2.4 rx. Decoupling is the provision of capacitors across the negative and positive supply lines to ‘soak up’ any sudden voltage excursions. We found that spacing the batteries apart was the answer, in fact as you moved the batteries and wires apart you could watch the servos lose their jittering. As you moved them back again so it returned.
I know what you mean about twitching like never before, the servos go absolutely berserk! Just flicking the prop starting caused all the surfaces to flick!

He fixed the ignition battery under the cowl, very close to the ignition unit and shortened the wiring, and left the rx battery in the fuz. He’s never had so much as the suspicion of even a quiver on the servos since.

Hopefully Lee’s suggestion will work ok, but in the event that it doesn’t this may be worth a try.

PB

[gazzazh]

Thanks Peter, I will try both and let you know, i also heard that the stock plug has been causing issues and excessive EMI , so will replace with NGK and see what happens... Hopefully this will get fixed and i can finally enjoy this model.. Cheers..

 

 

Posted by Peter Beeney on 06/11/2011 16:46:36

He fixed the ignition battery under the cowl, very close to the ignition unit and shortened the wiring, and left the rx battery in the fuz. He’s never had so much as the suspicion of even a quiver on the servos since.

Hopefully Lee’s suggestion will work ok, but in the event that it doesn’t this may be worth a try.

[Lee Smalley]

My job involves emc and rf testing and interference is a problem hard to solve shielding is just one thing to try rf interference can latch or couple for the technical term, onto any cable and this can be passed onto the rx or sevos you have to find the route cause bit by bit, I suggest you unplug all servos and test 1 by 1

[gazzazh]

Thanks all.. got it fixed.. all i had to do is remove the C-sub Nimh Battery and replace with an A123.. and no more problems. all went well.. Flies great.. had to make some modifications to the landing gear.. but other than that.. it a nice scale plane.. a bit on the heavy side.. but great to watch in the sky..

 

 

Cheers.

and thanks a million again..

 

[Lee Smalley]

well it was not rf interference then??

[gazzazh]

It is interference.. I had to change the location of the battery to get the interference to stop... Also had to change the stock Plug .. originally the battery was below the ignition.. Had to move the battery to the center of the fuse to get things working properly..

 

Thanks a million lee..

 

Cheers,

 

 

[Martin Harris - Moderator]

Posted by gazzazh on 06/11/2011 14:54:48:

THe servos are twitihing like never before..

 

I tried putting ferret rings on the throttle and but still nothing..

 

 

I'll bet there was considerable twitching in the local ferret community!

[Lee Smalley]

so the wires from the rx battery were picking up the rf noise and passing it back down to the RX??? interesting

[gazzazh]

Posted by Lee Smalley on 21/11/2011 12:50:50:

so the wires from the rx battery were picking up the rf noise and passing it back down to the RX??? interesting

Yep......

[gazzazh]

Posted by Martin Harris on 18/11/2011 12:30:02:

Posted by gazzazh on 06/11/2011 14:54:48:

THe servos are twitihing like never before..

 

I tried putting ferret rings on the throttle and but still nothing..

 

 

I'll bet there was considerable twitching in the local ferret community!

Hi Martin.. LOL.. So funny.. you made me laugh.. I meant ferrite rings.. but i am sure everyone is happy in fairy land with my little ferret..

 

Cheers,

Ps: i write most of my comments on my iphone.. so sometimes i make silly mistakes.. but hey if it made you laugh.. then great...

 

 

 

[Peter Beeney]

gazzazh - May I suggest, that as in my previous post above, the interference has little to do with radio frequencies, but rather more to do with the fact that pulses can be induced on the receiver power wires by the operation of the capacitor discharge ignition system on the petrol engine. The CD ignition unit is not that juicy, the one’s I’ve seen run for quite a while on a smallish battery, so it’s possible the pulses can be quite heavy but of short duration. When the wires (the ign. battery and receiver) run parallel and close together for a significant distance, a rapidly changing current on one will induce similar on the other. I think, exactly as in the case that we had, and some other, this is what was happening.
This does happen to power supply wires in other situations, too, so there is invariably a minimum separation distance between lines. Separating the wires so that they are not close obviously worked for you. Interestingly you mentioned the A123, I’m sure it’s quite possible the internal distance of these is lower than the hydrides, this would also reduce the magnitude of the pulses. If you had a source (battery) of infinitely low resistance there would very little pulsing.

I’m sure this will crop up again, and when it does we shall be ready for it!

Glad you got it sorted ok.

PB

Edited By Peter Beeney on 21/11/2011 13:53:23

[Solly]

Hi - glad to hear you've sorted the problem. Can I ask what make of servos you are using? I had the same problem and was surprised to discover that it disappeared when I changed from Futaba to Spektrum digi servos! I swapped from one to the other several times just to make sure.

[Lee Smalley]

PB is correct this can happen but the wires would have to be very close together for this to happen, if they are not that close then its prob RF noise from ig unit

[gazzazh]

Posted by Solly on 21/11/2011 15:24:31:

Hi - glad to hear you've sorted the problem. Can I ask what make of servos you are using? I had the same problem and was surprised to discover that it disappeared when I changed from Futaba to Spektrum digi servos! I swapped from one to the other several times just to make sure.

HI Solly ,, I am using Hitec 645MG...

 

Cheers,

 

[gazzazh]

Thanks all, this has been an interesting thread...

 

Cheers,

 

[Peter Beeney]

Interesting stuff, Solly, we’ve also done something similar but in different circumstances to which gazzazh had. This seems to happen always with 2.4 and standard servos and usually with twin elevator servos mounted right at the back of the fuselage; and always when they are on a Y lead. Significantly, you certainly do not have to run the engine to get it to happen. It starts when the radio is switched on. Not always with petrol engined models, it’s occurred on glow models too. In this case, I don’t think the engine is significant at all. We’ve found that changing the standard servos for digital immediately cures the problem 100%. So now this is the route we take straightaway now, we don’t play around trying to come up with an answer.

 

May I ask if your case is similar? And with petrol or glow? Also are your Futaba servos standard or digital?
If we get a few reports of this happening, with the details, we may be able pick up on what is going on.

Lee, With the greatest respect, I think I’m not entirely convinced about the r/f on the power lines. For a start, I don’t think I’ve ever heard any reports that any type of interference has ever been able to affect the servos in such a way. The same applies to r/f entering the receiver via the aerial. Also would the ignition system generate radio frequencies in the 2.4 GHz region in sufficient quantities and with sufficient strength to cause the servos to act in this manner? I personally think the voltage excursions are affecting the servos directly, not the receiver; although the receiver has to be working. In the case that we were dealing with there were quite a number or servos and it wasn’t practicable to change them all for digitals; and I’m not even sure that they weren’t digital types to start with anyway; although I think that’s unlikely. We were definitely able to prove there was interaction between the two sets of wires, you could see the servo jittering action come and go as you moved the wires closer or further apart. That was about 3 or 4 inches, at a guess.

So it would seem that not even 2.4 is entirely impervious to a spot of jiggery-pokery at times, I’m confident we solved the petrol ignition problem ok, but so far I’ve been unable to explain the Y lead phenomenon. There were some very odd angles to this. But by the same token, so far I’ve not seen any other rational explanation for it, either.
Perhaps I ought to just sling some servos together and lay them out in the in a model configuration to see if I can duplicate it. This worked for one person that suffered from this, whatever he did it would not go away. Until he used two separate extension leads and two channels, that is.
Again, I’m convinced it’s a power supply problem, I just can’t figure out exactly what!

PB

[Solly]

Peter, - my incident was different from your own experience. The engine used was a Brison 2.4 and interference only occurred when the engine was running. I initially used standard Futaba servos but experimented using Hitec, Spektrum (digital) and even an old Skyleader (non digital)servo. I found that the Spektrum and Skyleader servos were solid at all engine speeds, but the Futaba and Hitec servos twitched significantly as soon as the engine was running. The Futaba and Hitec servos are now in glo-powered airframes and work perfectly. I put it down to the fact that some servo circuits are simply more sensitive to ignition interference than others? Incidently, when I tried using ferrite rings it actually made the problem worse!

[Lee Smalley]

peter you are welcome to come our Anechoic chamber where we do all our rf testing if you don't believe it, i test out wireless mbus/zigbee comms protocol (2.4gig LPR) in there and i have put some radio gear in there, it does get affected, in the real world the effect is rare and slight, rf can latch onto any wires if the conditions are right and pass up these wires to the rx or servos, causing all sorts of weird activity and it does depend on the manufacture of A. what is causing the RF noise and B. the bits that might be affected!

 

example for you

I had a large petrol powered aircraft, when engine running one of the HS645s in the tail would twitch, not massively but enough to effect flight none of the others were effected, i connected the servo without its extension lead and had no problems, connect a shorter lead and no problems, another lead of the same length and it was back, it was clear that a certain freq being generated by the ignition unit was coupling onto the cable due to its length and effecting the servo, its rare but it does happen and 2.4 gig radio has nothing to do with it

never mind PB everyone get enlightened some times

[Former Member]

[This posting has been removed]

[Peter Beeney]

Solly - Many apologies, I forgot one simple question, do you use 35 MHz or 2.4 GHz? I’m not sure if this is significant even, both seem to be susceptible to interference, but your tale seemed to be slightly reminiscent of 35 MHz type interference. Certainly an instant cure does seem to be changing to digital servos.
As it happens, I’ve still got some Skyleader gear too, looking back on it the servos were by today's standards pretty grim, the little tapered servo arm locking arrangement particularly so. But the rest of it was brilliant. I have to admit, I certainly got a lot of use out of mine for many years.

Lee - Many apologies to you too, I’m afraid my post wasn’t very clear. We have two types of interference, the first is related to petrol engines and separating the batteries, that’s the ignition and receiver packs, totally removed it. If that happens to be related to some sort of r.f. problems then I’m quite happy with that, I’m not that knowledgeable about it anyway, and we don’t have any sort of test gear to be able to prove it. If it happens in the future we shall just try the battery trick first.
The second version is not related to any sort of engine in any way, I did try and point that out in my previous post. One of the lads had a problem with the double elevator servo setup on a Y lead, 2.4 radio, and it was very pronounced, it did it all the time. Glow engine, but that never ran, couldn’t even get to that point. We tried everything we could think of to fix it, many times over, but without success. I related this story in a thread a while back back, maybe a year ago. There were no suggestions of any reasons, or answers, as I remember. One of the things we tried was reverting back briefly to a 35 MHz receiver, that effected an instant total cure for the jittering, he was able to fly the model. However, he didn’t have the facilities that the 2.4 gave him, and he wanted to make a clean break from 35 anyhow, so he simply used two separate channels on the 2.4 rx; which he’s used up to now. But he did mention recently there were some drawbacks to this, with trimming and so on I believe, and he wondered if there was any chance of going back to trying a Y lead. As it so happens in between I’d now seen this happening to others and we quickly found that substituting a pair of digitals was the answer. I happened to have two digital servos with me at that time, he slung them in, coupled up with a Y lead and he thought it was instantly Christmas already! He just bought me two new replacements, he just refused to take them out again. Talking of Christmas, later on he was given another receiver, and that was exactly the same.

So, the reason for the engine related interference I thought was power related, for the reasons I’ve explained. Perhaps it’s not that important, at least we know how to cure it.
The reason I concluded that 2.4 seemed to be partly responsible for the second case was simply because 35 was not in any way affected. So a perfectly reasonable assumption, I’d have thought. And, as I said, at the moment I don’t really know what is going on, but I have a feeling this is power related, too. Maybe next year, when the weathers better, we will cobble something together and have a little tinker with it, to see if we can get some sort of inkling, anyway.

I can say I’m suitably enlightened, though, Lee… … at least for half an hour! Only kidding!

PB

[Solly]

Peter, I should have said that I initially used 35 MHz and changed to 2.4GHz. in an effort to solve the problem. When this did not work I then started to look at the servos and the wiring. I'm fairly convinced that the interference was not coming through the Rx but from the servos/leads themselves. I should also add that the batteries were placed 20 inches from the engine. The main thing is that by using other servos the problem has disappeared, - there's usually an answer if you persevere!

[Lee Smalley]

It’s a weird blackart Peter we have our PCBs made and see them fall over even at low (relative) Mgz and the solution can take many, many days to find some times its the length of a certain copper track on the PCB sometimes its a certain component, sometimes we never find it and have to just wrap the PCB in a metal can, rf susceptibility is very strange and can vary from one product, thankfully for us its very, very rare, I have only been doing RF testing for a few years but I was told to “never be surprised” I have even heard of rf noise affecting some certain batteries people think because 2.4 gig is a lot higher than the rf noise generated by say an electric motor that it does not get effected, well its partially true, the transmitted signal does not get effected, but that does not preclude the rf noise coupling onto any wires or directly onto any PCB (most Rxs are made in such a way that this cannot happen) and thus noise is transmitted up those wires to anything else they are connected to, as I said its very rare but it can effect 35meg or 2.4 it depends on the design, or installation within the airframe!