Lowering cockpit RF interference

[[email protected]]

Hello all EEs.

What is standard operating procedure for when and where to use shielded wire in cockpit wiring?

What is standard operating procedure for when and where to use ferrite sleeves?

Do both the positive and negative lead need to pass through the ferrite sleeve?

Thanks for the education.

[Tom Gardner[_2_]]

wrote:
> Hello all EEs.
>
> What is standard operating procedure for when and where to use shielded wire in cockpit wiring?
>
> What is standard operating procedure for when and where to use ferrite sleeves?
>
> Do both the positive and negative lead need to pass through the ferrite sleeve?
>
> Thanks for the education.

It can all be a bit trial and error, but
http://audiosystemsgroup.com/SAC0305Ferrites.pdf
gives some of the theory and practical aspects of it.

Key points:
- mount "close" to interference source
(close = less than 1/20th of the wavelength)
- mount "close" to whatever is being affected by
the interference
- multiple loops are better, usually
- resonance can cause surprises; be prepared to experiment
- any piece of wire can be an antenna, including the
cable shields

[Richard[_9_]]

On Friday, December 14, 2012 11:04:39 AM UTC-8, wrote:
> Hello all EEs. What is standard operating procedure for when and where to use shielded wire in cockpit wiring? What is standard operating procedure for when and where to use ferrite sleeves? Do both the positive and negative lead need to pass through the ferrite sleeve? Thanks for the education.

What equipment do you have in the cockpit,specify all equipment, Make and Model?

What kind on interference are you getting?

Richard
www.craggyaero.com

[Dave Nadler]

http://www.zug.com/img/ThreadImg/A/85510/1054165754_5715.jpg?ts=1355523304

[Eric Greenwell[_4_]]

On 12/14/2012 11:04 AM, wrote:
> Hello all EEs.
>
> What is standard operating procedure for when and where to use shielded wire in cockpit wiring?
>
> What is standard operating procedure for when and where to use ferrite sleeves?
>
> Do both the positive and negative lead need to pass through the ferrite sleeve?
>
> Thanks for the education.

The simple answer: whenever the equipment installation manual tells you
to use it.

Typically use shielded wire for antenna cables (ie, coax) and microphone
cables.

Newer equipment usually comes with sleeves on the cables they supply.

If the equipment installations done properly, nothing additional is
needed in an unpowered glider. Powered gliders can have interference
generated by the engine running, but again, if the engine electrics are
installed properly and in good condition, there usually isn't a problem.

Generally, I'm amazed at what people can get away with doing!

Most electrical noise problems come from specific problems in the
installation, and can be tough to track down. Ferrites can be an easy
fix sometimes; shielded wiring isn't so easy to install, and it's often
hard for even an experienced EE or technician to know exactly which
cable(s) is the problem and should be replaced.

--
Eric Greenwell - Washington State, USA (change ".netto" to ".us" to
email me)

[Martin Gregorie[_5_]]

On Fri, 14 Dec 2012 14:24:28 -0800, Eric Greenwell wrote:

> Most electrical noise problems come from specific problems in the
> installation, and can be tough to track down. Ferrites can be an easy
> fix sometimes; shielded wiring isn't so easy to install, and it's often
> hard for even an experienced EE or technician to know exactly which
> cable(s) is the problem and should be replaced.

The only electrical noise problem I've had was when I installed a 28v T&B
which was powered of the glider's 12v supply with a solid state 12v->28v
DC-DC converter. The solution was to put an electrolytic capacitor across
the 12V supply to the converter. This was a specialized noise cancelling
capacitor and IIRC it is 150 uF with a 1.5A transient current capacity.


--
martin@ | Martin Gregorie
gregorie. | Essex, UK
org |

[Dan Marotta]

I had the same problem, though with a 12v T&B, and used a ferrit in the
power lead to solve it.

>
> The only electrical noise problem I've had was when I installed a 28v T&B
> which was powered of the glider's 12v supply with a solid state 12v->28v
> DC-DC converter. The solution was to put an electrolytic capacitor across
> the 12V supply to the converter. This was a specialized noise cancelling
> capacitor and IIRC it is 150 uF with a 1.5A transient current capacity.
>
>
> --
> martin@ | Martin Gregorie
> gregorie. | Essex, UK
> org |

[Martin Gregorie[_5_]]

On Sat, 15 Dec 2012 11:49:06 -0700, Dan Marotta wrote:

> I had the same problem, though with a 12v T&B, and used a ferrit in the
> power lead to solve it.
>
I think I tried a ferrite core on the 12v side of the voltage converter
without success and then remembered that Maplins (the UK equivalent of
RadioShack) had these capacitors at a reasonable price. Better yet it was
easy to fit: I just soldered it across the T&B power switch.

>
>> The only electrical noise problem I've had was when I installed a 28v
>> T&B which was powered of the glider's 12v supply with a solid state
>> 12v->28v DC-DC converter. The solution was to put an electrolytic
>> capacitor across the 12V supply to the converter. This was a
>> specialized noise cancelling capacitor and IIRC it is 150 uF with a
>> 1.5A transient current capacity.


--
martin@ | Martin Gregorie
gregorie. | Essex, UK
org |

[bumper[_4_]]

On Saturday, December 15, 2012 2:32:26 PM UTC-8, Martin Gregorie wrote:
> On Sat, 15 Dec 2012 11:49:06 -0700, Dan Marotta wrote:
>
> I think I tried a ferrite core on the 12v side of the voltage converter
>
> without success and then remembered that Maplins (the UK equivalent of
>
> RadioShack) had these capacitors at a reasonable price. Better yet it was
>
> easy to fit: I just soldered it across the T&B power switch.
>
> martin@ | Martin Gregorie
>
> gregorie. | Essex, UK
>
> org


A couple of thoughts here . . .

When you say you soldered it "across" the power switch, that might lead one to assume you soldered it to the two terminals of your switch, i.e. in parallel with the switch contacts). Normally a filter capacitor would be connected across the power leads, i.e. from + to -. While a filter inductor (such as a ferrite) is connected in series.

When trouble shooting RFI, it helps to be able to quantify results. If it's interferance on comm freqencies, a hand held radio can be useful - squelch off, and distance away from suspected RFI source as appropriate one can use the handheld as a field strength meter for radiated interferance.

This is not so much use when dealing with closely spaced instruments on the panel where one is causing problems with another (LNAV did that to Comm on one of previous glider). There, just slapping in clamp-ferrites did the job.

If you have enough slack in wiring leads, and the size of the ferrite allows, you get almost double the iductance for loopiing the wire/s through the ferrite twice, (or even more for 3 times etc). Sometimes adding a capacitor in concert with the ferrite helps even more, especially when noise is coupling via power leads (ferrite or inductor in series and cap across the wires in parallel to form a classic pi filter).

Ferrites are not all the same, they have a frequency band at which they will be most effective. Same thing with the size of capacitors used in filtering apps. (do a search on inductive and capacitive reactance and resonant frequecy). In the U.S., Digi-Key is one good source for ferrites etc.

bumper
zz
Minden, NV

bumper

[Martin Gregorie[_5_]]

On Sun, 16 Dec 2012 11:02:57 -0800, bumper wrote:

>
> When you say you soldered it "across" the power switch, that might lead
> one to assume you soldered it to the two terminals of your switch, i.e.
> in parallel with the switch contacts). Normally a filter capacitor would
> be connected across the power leads, i.e. from + to -. While a filter
> inductor (such as a ferrite) is connected in series.
>
The switch is a double pole push-on push-off type, so one pole each for
ground and 12v I was able to use it as a two position tag strip, which
made wiring easier. The capacitor is across the +12 and ground connectors
on the T&B side of the switch.

> When trouble shooting RFI, it helps to be able to quantify results. If
> it's interferance on comm freqencies, a hand held radio can be useful -
> squelch off, and distance away from suspected RFI source as appropriate
> one can use the handheld as a field strength meter for radiated
> interferance.
>
It was purely electrical noise that cut in when the T&B was nearly up to
speed. I don't think it was RF because:

(a)ferrites didn't have any effect

(b) if it was coming from the DC-DC 12-28v solid state adapter I'd have
expected to hear it loudest when the T&B started because thats when you
see the biggest current drain from an electric motor.

(c) an old, mechanical T&B shouldn't be an RF source unless its sparking
like hell. This is a Mil-spec R C Allen type MD-4A, so possibly ex-USAF,
that has been modified to suit a glider's rate of turn rather than a
power plane. I'm told that these devices have a centrifugal switch that
sets the RPM and has the side effect of switching in and out at several
tens of cycles a second when its up to speed. That is what the racket
sounded like too.

> This is not so much use when dealing with closely spaced instruments on
> the panel where one is causing problems with another (LNAV did that to
> Comm on one of previous glider). There, just slapping in clamp-ferrites
> did the job.
>
This was very easily localized: T&B off: no problem. T&B turned on: nasty
noise on the radio once it had spun up. No effect from turning any other
stuff on or off.

....other good stuff snipped.


--
martin@ | Martin Gregorie
gregorie. | Essex, UK
org |

[Dan Marotta]

Been a long time since EE school, but, IIRC, "switching on and off" creates
a LOT of RF noise.


"Martin Gregorie" > wrote in message
...
> On Sun, 16 Dec 2012 11:02:57 -0800, bumper wrote:
>
>>
>> When you say you soldered it "across" the power switch, that might lead
>> one to assume you soldered it to the two terminals of your switch, i.e.
>> in parallel with the switch contacts). Normally a filter capacitor would
>> be connected across the power leads, i.e. from + to -. While a filter
>> inductor (such as a ferrite) is connected in series.
>>
> The switch is a double pole push-on push-off type, so one pole each for
> ground and 12v I was able to use it as a two position tag strip, which
> made wiring easier. The capacitor is across the +12 and ground connectors
> on the T&B side of the switch.
>
>> When trouble shooting RFI, it helps to be able to quantify results. If
>> it's interferance on comm freqencies, a hand held radio can be useful -
>> squelch off, and distance away from suspected RFI source as appropriate
>> one can use the handheld as a field strength meter for radiated
>> interferance.
>>
> It was purely electrical noise that cut in when the T&B was nearly up to
> speed. I don't think it was RF because:
>
> (a)ferrites didn't have any effect
>
> (b) if it was coming from the DC-DC 12-28v solid state adapter I'd have
> expected to hear it loudest when the T&B started because thats when you
> see the biggest current drain from an electric motor.
>
> (c) an old, mechanical T&B shouldn't be an RF source unless its sparking
> like hell. This is a Mil-spec R C Allen type MD-4A, so possibly ex-USAF,
> that has been modified to suit a glider's rate of turn rather than a
> power plane. I'm told that these devices have a centrifugal switch that
> sets the RPM and has the side effect of switching in and out at several
> tens of cycles a second when its up to speed. That is what the racket
> sounded like too.
>
>> This is not so much use when dealing with closely spaced instruments on
>> the panel where one is causing problems with another (LNAV did that to
>> Comm on one of previous glider). There, just slapping in clamp-ferrites
>> did the job.
>>
> This was very easily localized: T&B off: no problem. T&B turned on: nasty
> noise on the radio once it had spun up. No effect from turning any other
> stuff on or off.
>
> ...other good stuff snipped.
>
>
> --
> martin@ | Martin Gregorie
> gregorie. | Essex, UK
> org |

[Martin Gregorie[_5_]]

On Sun, 16 Dec 2012 17:12:24 -0700, Dan Marotta wrote:

> Been a long time since EE school, but, IIRC, "switching on and off"
> creates a LOT of RF noise.
>
Even at those low voltages and currents?

But what does solving the problem by fitting a simple capacitor across
the supply say about the relative amounts of RV vs. electrical noise?
Bear in mind that there's around 50cm of unshielded cable between the
capacitor on the switch and the T&B - cable that was supplied with it.


--
martin@ | Martin Gregorie
gregorie. | Essex, UK
org |

[Dan Marotta]

You're making my brain hurt thinking about these things! Switching DC on
and off amounts to a square wave which, as we all know, is made up of an
infinite series of sine waves, i.e., DC to light, and somewhere in there is
the RF spectrum.

From my days as an Air Force radio tecnhician, before becoming an edumacated
AF ossifer and pilot, my experience with receivers indicated that just a
couple of micro volts at the antenna input would break squelch. Now that
was from a 5 watt transmitter many, many miles away (line of sight); and
your source is merely inches away. A capacitor of the correct value acts as
a short circuit to RF at the tuned frequency whereas an inductor (RF choke)
acts like a high resistance (at the tuned frequency). Either or both of
these of the proper values and connected in the proper way will greatly
reduce or eliminate RF interference.

In my case, with the turn indicator causing the interference, I rummaged
through an electronics parts bin (actually a jelly jar) and found a
ready-made RF choke which consisted of a torroid with two wraps of wire
giving 4 loose ends. I connected one wrap (about 4 or 5 turns) in series
with +12v and the other in series with the ground to the instrument.
Problem solved.


"Martin Gregorie" > wrote in message
...
> On Sun, 16 Dec 2012 17:12:24 -0700, Dan Marotta wrote:
>
>> Been a long time since EE school, but, IIRC, "switching on and off"
>> creates a LOT of RF noise.
>>
> Even at those low voltages and currents?
>
> But what does solving the problem by fitting a simple capacitor across
> the supply say about the relative amounts of RV vs. electrical noise?
> Bear in mind that there's around 50cm of unshielded cable between the
> capacitor on the switch and the T&B - cable that was supplied with it.
>
>
> --
> martin@ | Martin Gregorie
> gregorie. | Essex, UK
> org |

[Martin Gregorie[_5_]]

On Mon, 17 Dec 2012 09:42:00 -0700, Dan Marotta wrote:

> You're making my brain hurt thinking about these things! Switching DC
> on and off amounts to a square wave which, as we all know, is made up of
> an infinite series of sine waves, i.e., DC to light, and somewhere in
> there is the RF spectrum.
>
> From my days as an Air Force radio tecnhician, before becoming an
> edumacated AF ossifer and pilot, my experience with receivers indicated
> that just a couple of micro volts at the antenna input would break
> squelch. Now that was from a 5 watt transmitter many, many miles away
> (line of sight); and your source is merely inches away. A capacitor of
> the correct value acts as a short circuit to RF at the tuned frequency
> whereas an inductor (RF choke) acts like a high resistance (at the tuned
> frequency). Either or both of these of the proper values and connected
> in the proper way will greatly reduce or eliminate RF interference.
>
Thanks for the clear explanation. Much appreciated.

> In my case, with the turn indicator causing the interference, I rummaged
> through an electronics parts bin (actually a jelly jar) and found a
> ready-made RF choke which consisted of a torroid with two wraps of wire
> giving 4 loose ends. I connected one wrap (about 4 or 5 turns) in
> series with +12v and the other in series with the ground to the
> instrument. Problem solved.
>
Good one, and another good explanation. I've not previously understood
exactly what an RF choke did.

BTW, about 10 years ago I remember meeting a fellow glider pilot, name of
Dan, at the Sierra Cup free flight bash, but didn't catch his last name.
Was that you by any chance?


--
martin@ | Martin Gregorie
gregorie. | Essex, UK
org |

[[email protected]]

On Monday, December 17, 2012 11:42:00 AM UTC-5, Dan Marotta wrote:

> Either or both of these of the proper values and connected in the proper way will greatly reduce or eliminate RF interference.
>
>
Thanks Dan for for your input. Since exercise is good for the brain, inquiry minds want to know how to determine proper capacitor and RF choke values and how to connect correctly.

[Dan Marotta]

> BTW, about 10 years ago I remember meeting a fellow glider pilot, name of
> Dan, at the Sierra Cup free flight bash, but didn't catch his last name.
> Was that you by any chance?

> martin@ | Martin Gregorie
> gregorie. | Essex, UK
> org |


Nope! Wasn't me...

[Dan Marotta]

Frankly, I've been out of school far too long to do a decent job of
designing a filter. I've included a link on filter design below if you want
to slap a few components together. Or you could simply go to your local
electronics supplier and tell them that you've got a buzzing in your radio
caused by something switching DC and you want to insert a low pass filter or
an RF choke in the DC line.

This will help you design a lowpass filter:

http://www.radio-electronics.com/info/rf-technology-design/rf-filters/simple-lc-lowpass-filter-design.php

However, simply wrapping your +12v line through a ferrit ring will serve the
purpose.
> wrote in message
...
> On Monday, December 17, 2012 11:42:00 AM UTC-5, Dan Marotta wrote:
>
>> Either or both of these of the proper values and connected in the proper
>> way will greatly reduce or eliminate RF interference.
>>
>>
> Thanks Dan for for your input. Since exercise is good for the brain,
> inquiry minds want to know how to determine proper capacitor and RF choke
> values and how to connect correctly.

[JohnDeRosa]

Your question is less about "I have a problem and want to fix it." versus, "I don't want any problems in the first place. What are the best practices to follow?"

Literally the only interference problem I have every had was a vario that deflected when I keyed the transceiver. Solution: replace the vario. I must live a blessed life. This has also been in pure gliders only so I don't have to worry about ignition noise which is by far the worse culprit for interference. However, new electronics run at some high frequencies (1Mhz processors are sloooow these days), so interference can happen.

So back to what I do to prevent problems in the first place.

1) Start with QUALITY - Components, wiring, terminals, switches, etc. Buy from an aviation source like Wicks or Aircraft Spruce. Radio Shack and ACE Hardware DO NOT have aviation aisles. Modern avionics seldom seem to create issues as they have to pass some pretty stringent testing to be certified. Step away from anything still using crystals (are they still legal?).
2) Coax for microphones and antennas. But every microphone and antenna already comes with coax. The more important bit is making sure that the shield has a good ground at the avionics end. A good BNC connection for the antenna (have someone knowledgeable help). Many transceivers have a separate ground pin for the microphone - use it!
3) Ferrite beads - While my EE specialty is not in the RF spectrum (get it?), putting these on doesn't hurt and might help ... so I use them. As someone in this thread said, run both the +/- wires through the bead twice if you have the slack. Non-split beads are generally better than split beads. The trick is to prevent high frequency noise from one device getting into another device through your power bus.
4) Air gaps - Wires that might have data flowing through them, like a GPS to a PDA, can induce noise into another wire running right along side of it in a bundle. So separate any suspect wires from one another.
5) Grounding - Ground all your devices to a single common location as close to the electronics as possible, including the negative side of the battery.. Even ground the metal stick, push tubes, etc, if possible.

Good luck, John

[[email protected]]

Thank you Dan for your article link

John DeRosa wrote:
> Your question is less about "I have a problem and want to fix it." versus, "I don't want any problems in the first place.

Exactly.

Do late model sailplanes actually have a "factory" grounding point?

Am I correct to assume the best grounding point would be the heaviest metal frame work closest to the instrument panel such as the control yoke or the adjustable rudder pedal track?

So in the perfect world all things metal (especially long lengths like push rods) and your negative instrument buss should be grounded to a common ground point?

[Eric Greenwell[_4_]]

On 12/20/2012 7:58 AM, wrote:
> Thank you Dan for your article link
>
> John DeRosa wrote:
>> Your question is less about "I have a problem and want to fix it."
>> versus, "I don't want any problems in the first place.
>
> Exactly.
>
> Do late model sailplanes actually have a "factory" grounding point?
>
> Am I correct to assume the best grounding point would be the heaviest
> metal frame work closest to the instrument panel such as the control
> yoke or the adjustable rudder pedal track?
>
> So in the perfect world all things metal (especially long lengths
> like push rods) and your negative instrument buss should be grounded
> to a common ground point?

I don't know how many adhere to the "single grounding point"; however,
note that the point is for the instruments, not necessarily everything.
The other metal bits, which are not carrying signals, can be grounded
elsewhere, such as the near the battery. A powered sailplane will likely
also have an "engine" ground, so the starter currents, ignition system,
generator, etc. are kept from the instrument ground.

--
Eric Greenwell - Washington State, USA (change ".netto" to ".us" to
email me)

[Tom Gardner[_2_]]

wrote:
> Thank you Dan for your article link
>
> John DeRosa wrote:
>> Your question is less about "I have a problem and want to fix it." versus, "I don't want any problems in the first place.
>
> Exactly.
>
> Do late model sailplanes actually have a "factory" grounding point?

Certainly not when they are in the air!

OK, that's a joke, but it does bring into sharp
relief that the "ground" concept is a fiction
(except under limited circumstances that are not
relevant here). It is also a fiction on the ground
as well, particularly where RF is concerned (see
any book on antenna design).


> Am I correct to assume the best grounding point
> would be the heaviest metal frame work closest
> to the instrument panel such as the control yoke
> or the adjustable rudder pedal track?

I doubt it, but read the service manual and contact
a competent engineer/technician. I would be
very reluctant to allow electricity to be conducted
through any part of the airframe.

[Eric Greenwell[_4_]]

On 12/20/2012 9:17 AM, Tom Gardner wrote:
> wrote:
>> Thank you Dan for your article link
>>
>> John DeRosa wrote:
>>> Your question is less about "I have a problem and want to fix it."
>>> versus, "I don't want any problems in the first place.
>>
>> Exactly.
>>
>> Do late model sailplanes actually have a "factory" grounding point?
>
> Certainly not when they are in the air!
>
> OK, that's a joke, but it does bring into sharp
> relief that the "ground" concept is a fiction
> (except under limited circumstances that are not
> relevant here). It is also a fiction on the ground
> as well, particularly where RF is concerned (see
> any book on antenna design).

How do you think we should be discussing the issue?

--
Eric Greenwell - Washington State, USA (change ".netto" to ".us" to
email me)

[Tom Gardner[_2_]]

Eric Greenwell wrote:
> On 12/20/2012 9:17 AM, Tom Gardner wrote:
>> wrote:
>>> Thank you Dan for your article link
>>>
>>> John DeRosa wrote:
>>>> Your question is less about "I have a problem and want to fix it."
>>>> versus, "I don't want any problems in the first place.
>>>
>>> Exactly.
>>>
>>> Do late model sailplanes actually have a "factory" grounding point?
>>
>> Certainly not when they are in the air!
>>
>> OK, that's a joke, but it does bring into sharp
>> relief that the "ground" concept is a fiction
>> (except under limited circumstances that are not
>> relevant here). It is also a fiction on the ground
>> as well, particularly where RF is concerned (see
>> any book on antenna design).
>
> How do you think we should be discussing the issue?

I don't really understand the question, but maybe
we shouldn't be discussing it here. Interference
arises due to many causes, is transmitted by many
means, and is received in different ways [1]. A specific
solution to one combination of source/route/destination
will not be effective for other combinations. I'm
sorry, but there aren't any generic solutions.

Probably the best general advice is to have good quality
equipment, cables and connectors, and to ensure the
connectors are properly tightened.

And even that won't prevent the "rusty bolt effect" :)


[1] as a single _simple_ example, given signals coupling
from one wire to another leading to interference, there
are two types of interference, near-end crosstalk and
far-end crosstalk. The causes are different and the
solutions are different.

RF introduces a whole slew of more subtle effects.
There are _many_ books on the topic, which is a good
indication that it isn't a simple problem with a simple
solution.

But of course one solution to one problem may well
be simple.

[Chip Bearden[_2_]]

This is like posing a question to a lawyer: "It depends."

What might be helpful to non-experts is a typical step-by-step
approach to identifying the type and source of interference and then
addressing it. For you AI types (is that term still used?), an expert
system flow chart. EX: If it's a noise in the radio speaker, go to
step 2a. If it's a noise in the audio vario speaker, go to step 2b. If
it's a disturbance in the needle or indicator of an instrument, go to
step 2c. Etc. "It depends" covers too much ground.

I've chased interference before, the most maddening case being hearing
a French African channel on Voice of America non-stop thru the
aircraft radio years ago when flying at Caesar Creek, OH. It was my
brother's glider, actually, but we worked on it together. From memory:
capacitors, homemade high-pass, low-pass, and band-pass filters,
shieded cables, shielding around various components, etc. Eventually
we drove the trailer over to the VOA facility, parked on the road
outside, pulled the fuselage out, and started working. That got action
pretty fast. I didn't realize VOA had security guards. Once they
believed our story, and that the Komet trailer wasn't full of
explosives, we were actually invited in to meet the manager and get
info on freqs, antennas, power, etc. Pretty awesome numbers, as I
recall. I honestly can't remember what finally did the trick. Maybe it
was my brother moving to VA. :)

Chip Bearden
ASW 24 "JB"
U.S.A.