Matronics Email Lists :: View topic

Matronics Email Lists
Web Forum Interface to the Matronics Email Lists

Get Email Distribution Too! FAQ Search Memberlist Usergroups Register

Profile Log in to check your private messages Log in

ARINC Wiring Shield?

Matronics Email Lists Forum Index -> AeroElectric-List

View previous topic :: View next topic

Author Message

art(at)zemon.name
Guest

Posted: Mon Aug 26, 2019 3:47 am Post subject: ARINC Wiring Shield?

Folks,

I am about to integrate one of the new Garmin 175 GPS receivers into my airplane and that involves a couple of ARINC wire pairs. I am surprised to see that the MGL-to-Garmin integration guide suggests shielded twisted pair wiring. From my read of the signaling, ARINC uses a voltage differential between the wires so it should be immune to noise and not require shielding (much like the CAN bus).

Is shielding actually required (or even commonly suggested) for ARINC wire pairs?

Â  Â  -- Art Z.

-- 

https://CheerfulCurmudgeon.com/Love the stranger for you yourselves were strangers in Egypt. Deut. 10:19

dj_theis

Joined: 28 Aug 2017
Posts: 56
Location: Minnesota

Posted: Tue Aug 27, 2019 5:34 am Post subject: Re: ARINC Wiring Shield?

Hi Art,



I think it might be a case of â€œbelt and suspendersâ€ but in the industrial arena, it is not uncommon for vendors implementing CAN networks to recommend twisted + shielded cabling.



https://www.belden.com/hubfs/resources/technical/technical-data/english/773105T.pdf?hsLang=en



Granted, an airplane is not as electrically â€noisyâ€ as a modern manufacturing site but as I expect the more learned contibutors to the forum will advocate, following the vendorâ€™s recommended installation practices is usually good advice.



Best regards,



Dan.

_________________
Dan Theis
Scratch building Sonex #1362
Still working on the Revmaster Alternator improvement

nuckolls.bob(at)aeroelect
Guest

Posted: Tue Aug 27, 2019 2:29 pm Post subject: ARINC Wiring Shield?

At 08:34 AM 8/27/2019, you wrote:

  	  Quote: 	
		  --> AeroElectric-List message posted by: "dj_theis" <djtheis58(at)gmail.com>



 Hi Art,



 I think it might be a case of â€œbelt and suspendersâ€ but in the industrial arena, it is not uncommon for vendors implementing CAN networks to recommend twisted + shielded cabling. 	


   Yeah . . . probably . . . but about the only way

   you can buy twisted pair is under a shield. I've

   probably got a few thousand feed of 2x22S . . .

   how many feet do you need . . . send me a mailing

   address



 

   Bob . . .

henador_titzoff(at)yahoo.
Guest

Posted: Wed Aug 28, 2019 4:34 am Post subject: ARINC Wiring Shield?

"Yeah . . . probably . . . but about the only wayÂ  you can buy twisted pair is under a shield."

This is not quite true, Bob.Â  There are several examples of twisted wire cables that do not have shields.Â  The most ubiquitous example is Ethernet, also known as 10BASE-T, 100BASE-T and 1000BASE-T. These three media are basically the same but are modified slightly to accommodate faster and faster data transmission.Â  They are defined in the IEEE 802 standard.Â These 3 cables are hardly distinguishable by the naked eye except for color coded plastic jackets, and they contain no shielding.Â They only contain 4 twisted wire pairs that are held in close proximity to each other by the jacket.

It was determined by defense contractors back in the 1980s that twisted wires will accomplish two things.Â  First, each pair keeps its magnetic field to a bare minimum by ensuring that the twisted wires are in close proximity to each other.Â  As current flow in one direction on one wire, its mate carries the return current back to its source.Â  The resulting opposite magnetic fields from each wire cancel each other out due to their close proximity. This means that their twin magnetic fields do not interfere with the other twisted wire pairs' magnetic fields.Â  Second, if the twisted wire pair transmitters and receivers are differential instead of single-ended, they can reject offending magnetic fields, because the injected voltages are the same in the two wires, due to their proximity to each other. The differential receiver sees the offending voltages but rejects them due to its design.

The major defense contractors have known about this technique for a while now and have implemented them in both wire cables and printed circuit boards to reduce the chances of electromagnetic interference.Â  Minor defense contractors and commercial companies have taken a much longer time to discover how to transmit signals from point A to B and have made painstaking mistakes that have affected their schedules and reliability tremendously.Â  In many cases, they still haven't figured it out.

All of this came about because of increasing signal speeds.Â  Most people think of signal speeds in terms of clock speed.Â  The correct way of looking at it is the signal's rising and edge times. The true electromagnetic spectra of these rising and falling edges can be studied by use of the Fourier transform.Â  This transform will reconstruct the edges using discrete frequency spectra, whose frequencies extend harmonically beyond the clock speed. In other words, the frequencies generated by a simple clock or signal pulse go way beyond what you will see on the signal's oscilloscope waveform.Â  Then by using Maxwell's equations, one can study how the fields radiate out and come up with solutions.

Today's technology is much different than 1980s technology, with spectra going into the microwave region. In the commercial world, major companies like Intel have had to develop and rely on tools that study both motherboard and IC layouts to minimize interference from closely placed components.Â  They have developed techniques like microstrip and stripline to minimize electromagnetic interference, much like what twisted wire pairs do.Â  They do not rely on shielding, because they're keeping each signal's offending magnetic fields to a bare minimum, even when located microns from each other.Â  The two methods mentioned above were borrowed from the radar world. In other words, it took RF engineers to solve computer engineers' problems as speeds increased.

In my opinion, the EFIS manufacturers are behind the 8-ball, mostly because their engineers have not been exposed to what I say above.Â  Furthermore, these companies are really run by software engineers who are living in the ideal world and have no clue about signals and their spectra and how to control them. I've seen several examples of EFIS manufacturers solve "glitches" with software.Â  These software patches are only band-aids and continue to proliferate in future designs, because they don't truly understand the problem.Â  As band-aids accumulate, the EFIS functionalities become even more quirky.Â  I see this in many other products, and people explain them away as "just a glitch."

Shields are used by the "experts" for two reasons.Â  One is to provide a more uniform controlled impedance to transmit RF power and receive weak RF energy.Â  Two is for physical integrity such as the wire telling the ignition mag to turn on and off. That wire could easily be replaced with a twisted wire pair, but shielding provides better physical protection.Â  You could run it in a PVC pipe and accomplish the same thing.

I know this all sounds mysterious to most people, but I know for a fact that 3 defense manufacturers started looking at signal integrity problems starting in the 1980s when things started getting faster, and glitches started popping up.Â  These 3 companies wrote design manuals for their engineers to follow.Â  These books were cook books designed to avoid previously made mistakes. They actually have review boards that ensure the guidelines were followed.Â  If you don't follow them, you better have a damn good reason why you didn't. I know of several cases where design engineers actually came up with better ideas, and the design manuals were revised.Â  In all cases, reliability and problem free operation overrode other factors

nuckolls.bob(at)aeroelect
Guest

Posted: Wed Aug 28, 2019 10:40 am Post subject: ARINC Wiring Shield?

>This is not quite true, Bob.  There are several examples of twisted wire cables that do not have shields.  The most ubiquitous example is Ethernet, also >known as 10BASE-T, 100BASE-T and 1000BASE-T.



    But where would we find Cat5 cable

    on an airplane? I was speaking of wire

    I might pull from inventory at Beech . . .



 

   Bob . . .

Kellym

Joined: 10 Jan 2006
Posts: 1705
Location: Sun Lakes AZ

Posted: Wed Aug 28, 2019 12:18 pm Post subject: ARINC Wiring Shield?

Well, my Dynon Skyview screens use Cat5 wire between the 2 screens to 

sync data updates and common data like altimeter setting, etc. Not 

required but eases syncing of things. Otherwise, no, I don't think of 

any uses for ethernet cabling in aircraft.



On 8/28/2019 11:37 AM, Robert L. Nuckolls, III wrote:

 	  Quote: 	
		   

  >This is not quite true, Bob.  There are several examples of twisted 

 wire cables that do not have shields.  The most ubiquitous example is 

 Ethernet, also >known as 10BASE-T, 100BASE-T and 1000BASE-T.

 

     But where would we find Cat5 cable

     on an airplane? I was speaking of wire

     I might pull from inventory at Beech . . .

 

    Bob . . .

_________________
Kelly McMullen
A&P/IA, EAA Tech Counselor # 5286
KCHD

henador_titzoff(at)yahoo.
Guest

Posted: Wed Aug 28, 2019 4:11 pm Post subject: ARINC Wiring Shield?

I used Ethernet cable as an example of twisted wire without a shield being in use.

For aircraft use, twisted wire pairs without a shield are easy to find - they're in in Beech inventory.  All you do is twist 2 same length Tefzel wires by holding one end in a vise and putting the other 2 ends in a drill.  The drill will twist them together.  Remember that the wires do not have to be tightly wound, which makes them awkward to work with and adds weight.  They only have to be wound enough to keep them in close proximity.  Also, you don't really need a drill.  You can hand twist them, because they don't have to be tightly wound.

The trick to a well designed, highly reliable system is to run as many signals as possible using twisted wire pairs.  This minimizes offenders that may induce voltages in signal wires that aren't twisted (stragglers).  Even power wires to black boxes should be twisted, because power surges can also create magnetic fields.

Henador Titzoff



         



                  

                                                                    On Wednesday, August 28, 2019, 03:39:55 PM EDT, Robert L. Nuckolls, III <nuckolls.bob(at)aeroelectric.com> wrote:                 

                 



                 



                   

 >This is not quite true, Bob.  There are several examples of twisted wire cables that do not have shields.  The most ubiquitous example is Ethernet, also >known as 10BASE-T, 100BASE-T and 1000BASE-T.



    But where would we find Cat5 cable

    on an airplane? I was speaking of wire

    I might pull from inventory at Beech . . .



 

    Bob . . .

nuckolls.bob(at)aeroelect
Guest

Posted: Wed Aug 28, 2019 6:54 pm Post subject: ARINC Wiring Shield?

At 03:15 PM 8/28/2019, you wrote:

  	  Quote: 	
		  --> AeroElectric-List message posted by: Kelly McMullen <kellym(at)aviating.com>



 Well, my Dynon Skyview screens use Cat5 wire between the 2 screens to sync data updates and common data like altimeter setting, etc. Not required but eases syncing of things. Otherwise, no, I don't think of any uses for ethernet cabling in aircraft. 	


    . . . but of course. I've seen some

    products exploit the value of crimp

    on RJ series connectors. I've got

    some transceivers that have RJ45

    connectors on the microphone.



    The logical extension of using

    those connectors is to exploit the

    constellation of off-the-shelf

    connectors, wire and tooling.



 

 

   Bob . . .

jsajpf

Joined: 21 Feb 2019
Posts: 8

Posted: Thu Aug 29, 2019 1:10 pm Post subject: Re: ARINC Wiring Shield?

Do a web search for FAA Report DOT/FAA/CT-86/40, â€œAircraft Electromagnetic Compatibility" and review discussions on shielding. Section 3.3.1 deals with shielding ARINC 429 signal wires.

The doc is somewhat dated now, having come out after the first generation of airplanes making heavy use of the ARINC 429 protocol were certified. Its been a foundational document in developing design and analysis assumptions (at least in large transport category airplanes).  The principles remain valid even in the age of AFDX/ARINC 664 ethernet based protocols.



I personally wouldn't want to spend the energy justifying (if even only to myself) deviating from installation recommendations in this matter.



John

matronics(at)rtist.nl
Guest

Posted: Fri Aug 30, 2019 2:57 am Post subject: ARINC Wiring Shield?

In addition, the shielding may help to gain mechanical strength and 

chafing protection.



Rob



On 8/27/2019 3:34 PM, dj_theis wrote:

 	  Quote: 	
		   



 Hi Art,



 I think it might be a case of Ã¢â‚¬Å“belt and suspendersÃ¢â‚¬Â but in the industrial arena, it is not uncommon for vendors implementing CAN networks to recommend twisted + shielded cabling.



 https://www.belden.com/hubfs/resources/technical/technical-data/english/773105T.pdf?hsLang=en



 Granted, an airplane is not as electrically Ã¢â‚¬ÂnoisyÃ¢â‚¬Â as a modern manufacturing site but as I expect the more learned contibutors to the forum will advocate, following the vendorÃ¢â‚¬â„¢s recommended installation practices is usually good advice.



 Best regards,



 Dan.

Display posts from previous:

	Matronics Email Lists Forum Index -> AeroElectric-List	All times are GMT - 8 Hours
Page 1 of 1

Jump to:

You cannot post new topics in this forum
You cannot reply to topics in this forum
You cannot edit your posts in this forum
You cannot delete your posts in this forum
You cannot vote in polls in this forum
You cannot attach files in this forum
You can download files in this forum