[rjcthree]

I'm laying out electrical, and realized that I'm back into heavy/automotive frame of wiring - that is, with the exception of under the dash, means common ground returns(to source). This would essentially double the length of wire the current would have to travel, resulting in having to step up in wire size to minimize losses(think in terms of a landing light). This is bad.

But how conductive is alodined surfaces in contact? I have to imagine bonding wing to fues will be required, but if I use the tip rib as the ground for position lights, taxi/langing lights, and strobes . . .

I've read that resistance along an airframe used as a common ground is 'negligable' if bonded correctly, but I'm not sure you can say that across the board. I'm also a little worried about corrosion induced by crrent flow.

Per the Vans' website, 5555 other builders have dealt with this one - but my search came up a bit dry, and I have yet to get into contact with a tech advisor . . .

Thoughts, please.

Rick 90432 - final tank fitting and conduit running (grr) prior to closure.

 

[jonbakerok]

Use the airframe for your ground return

There is a far greater chance that your single giant ground return wire will corrode or break than there is for all the thousands of connection points in your airframe to simulaneously corrode to the point that the electrons won't be able to find their way back to the battery. As far as alondizing, it is only a few atoms thick, is not an insulator, and the act of installing a fastener instantly removes it anyway.

 

[L.Adamson]

I used a larger ground wire (#8 if I remember) from the main firewall ground, to a junction box under the seats. From there, it splits to a grounding connection on the first rib of each wing. Lights at the wing tips are grounded to the wing. A heavier load such as my heated pitot, carries the ground wire back to the wing ground mentioned above. Many other grounds such as the auto-pilot servo behind the baggage bulkhead and in the wing, have ground wires running back to the seat junction box.

With aircraft wire being as many stranded as it is, I don't worry about corrosion or breaking. Since I've been doing electrical work in one form or another for many years, I'm quite satisfied with the way I've done it.

L.Adamson

 

[az_gila]

FAA grounding specs.

....
I've read that resistance along an airframe used as a common ground is 'negligable' if bonded correctly, but I'm not sure you can say that across the board.......

In this case, "bonded correctly" is defined by the "bible".

AC 43.13 has some nice pictures and hardware detail on how to bond ternimals to the airframe to prevent corrosion effects

It's in Chapter 11, Table 11-15 and 11-15.

This long link might get you to the right chapter on theFAA web site...

http://rgl.faa.gov/Regulatory_and_G...9baac81b86256b4500596c4e/$FILE/Chapter 11.pdf

These methods have been well tested over time and do work...
There is no advantage to re-inventing the wheel here, which will drive up your $$$ and complexity...

Note that the exception to airframe ground is for the starter current (100+ amps?) - and this is handled with ground wires (or usually ground bonding braids) that connect the engine, motor mount, airframe and battery negative together as a low resistance return for the starter current.

gil A

 

[rv9av8tr]

Preventing ground loops

An electrical / avionics engineer in our EAA Chapter recently told us that it is important to not use the airframe directly as a grounding point, in order to prevent ground current loops. Ground loops lead to RF noise generation which degrades communications and antenna performance. It's best to use a single point Star ground or Modified Star (multiple isolated grounds in various locations of the airplane, which all feed back to your single point ground via a wire).

He said especially your strobe power supply case, which can be physically grounded directly via the case to the airframe, should not be. Mount the power supply on a sheet of lexan and ground the case via a wire to your star ground.

I know that lots of folks use their airframes directly as multipoint grounds, but it's clear how that can lead to RF noise generation and degraded signals.

 

[Brantel]

An electrical / avionics engineer in our EAA Chapter recently told us that it is important to not use the airframe directly as a grounding point, in order to prevent ground current loops. Ground loops lead to RF noise generation which degrades communications and antenna performance. It's best to use a single point Star ground or Modified Star (multiple isolated grounds in various locations of the airplane, which all feed back to your single point ground via a wire).

He said especially your strobe power supply case, which can be physically grounded directly via the case to the airframe, should not be. Mount the power supply on a sheet of lexan and ground the case via a wire to your star ground.

I know that lots of folks use their airframes directly as multipoint grounds, but it's clear how that can lead to RF noise generation and degraded signals.

I second this statement, As an Electrical/Automation engineer, I have delt with controls that have thousands of I/O and analog signals. Usually most noise related issues can be traced back to poor grounding practices which lead to ground loops.

I firmly believe in the star grounding method outlined above and any device that uses its metal case to provide shielding should be isolated to prevent a ground loop opportunity. If the device has a metal case and its ground terminal are internally connected, I would recommend case isolation and take the ground termninal back to the common star point.

 

[az_gila]

Avionics are separate...

An electrical / avionics engineer in our EAA Chapter recently told us that it is important to not use the airframe directly as a grounding point, in order to prevent ground current loops. Ground loops lead to RF noise generation which degrades communications and antenna performance. It's best to use a single point Star ground or Modified Star (multiple isolated grounds in various locations of the airplane, which all feed back to your single point ground via a wire).

He said especially your strobe power supply case, which can be physically grounded directly via the case to the airframe, should not be. Mount the power supply on a sheet of lexan and ground the case via a wire to your star ground.

I know that lots of folks use their airframes directly as multipoint grounds, but it's clear how that can lead to RF noise generation and degraded signals.

Moderation is needed here...

As the original poster said, he excluded the "stuff" behind the panel.

The Nuckell's approach of a single point ground behind the panel for avionics (the sensitive stuff) and airframe ground for the rest is a good approach... and is well proven.

Noise examples I've seen have been bad connections, sensitive ANR headsets (yes, found a case on a Piper yesterday) and multiple avionic grounds.

Keep the avionics separate, and you will be OK.... lots of history on this one... Anything else will add $$$, weight (small, in our case) and complexity.

gil A - and yes... I'm an Electronics Engineer too... dealt with sensors far more sensitive than our comm recievers working next to MegaWatt lasers...

PS .. all above comments above null and void for a glass plane....

 

[CESSNADON]

Airframes make a fine ground...It would take an eternity for the ground currents to cause troublesome corrosion on an airframe. I concur with the previous poster, Keep the avionics stack grounds connected to a single point and the other equipment, grounded at its mounting location.

 

[Ironflight]

There tens of thousands of metal airplanes flying around using the airframe as the ground, and it works just fine. I also agree that using a single point common ground for the avionics is a very good idea! Another point to make is that when you are wiring up audio stuff (headset jacks, comm radios, audio panels, entertainment sources), wire them exactly as shown in the documentation - probably the audio panel drawings will be the best - but I am willing to bet that in most cases, you'll be told to ground all shields at the audio panel end ONLY, and let the shields, jacks, and radios "float" at the other end. DO it that way, and the potential for hum and buzz is greatly reduced!

Just my experience with several aircraft...

Paul

 

[Rivethead]

An electrical / avionics engineer in our EAA Chapter recently told us that it is important to not use the airframe directly as a grounding point, in order to prevent ground current loops. Ground loops lead to RF noise generation which degrades communications and antenna performance. It's best to use a single point Star ground or Modified Star (multiple isolated grounds in various locations of the airplane, which all feed back to your single point ground via a wire).

He said especially your strobe power supply case, which can be physically grounded directly via the case to the airframe, should not be. Mount the power supply on a sheet of lexan and ground the case via a wire to your star ground.

I know that lots of folks use their airframes directly as multipoint grounds, but it's clear how that can lead to RF noise generation and degraded signals.

It's exactly for this reason that I'm running the extra wire and taking everything back to the battery.

 

[CESSNADON]

Woe be it for me to challenge an avionics engineer, speaking from personal experience, his reccomendations are gross overkill. The major contributors of noise in airplanes are the alternator and the ignition system. If both are installed properly you should not have any noise. What is the logic behind mounting the strobe power supply on a piece of plastic?? The proper sized powersupply wire which is connected at a low impedence point on the buss is the best way to avoid the ticking noise caused by the strobes. There are two main reasons that intercoms hum...the first is a poor connection on the low side of the microphone connections and tthe second is ....that they don't know the words.....

 

[FrankK90989]

And then

Before you install your (add name of electrical equipment here) take out your handy dandy tester and check the ground terminal to the chasis. Most all the stuff I checked the chasis and ground is common, I did not check if these items would power up just grounding the chasis. I don’t know if I an a believer even though I did the common ground thing.

edit--Icom radio\garmin xponder will power up with no ground wire hooked up-----

 

[Brantel]

There are several variations that can work and several that can cause an opportunity for ground loops. I have listed some of them below and what could be concidered as good and bad ways to wire loads. In some cases, you can't eliminate the opportunity for a ground loop. In this case, do the best you can and move on. The less exposure your system has to ground loops, the less risk you will have of a problem.

 

[Brantel]

 

[az_gila]

Numbers

The figures above rely on resistance to make ground loops occur, or not occur.

The point is that the resistance of the metal skin and structure from a grounded box back to the battery is much smaller than the resistance of a wire. Hence most of the current will flow in the skin (like water, path least resistance) and not in the wire, if given a choice.

This works as long as the initial contact point is low resistance, so the AC43.13 ground point attachment techniques I linked to earlier should be followed.

A large chunk of aluminum sheet is less resistance than your #16 strobe return wire....

It's been done this way for many years, why re-invent the wheel?

NOTE... taking sensitive avionics to a single point ground behind the panel is a good idea, and this short line will isolate audio from such things as a possible strobe return ground loop.

Remember, grounded boxes on strobe supplies, etc. also act to suppress radiated (in and out) RF as a shield.

gil A

 

[Dave Cole]

You make a good point, Gil.

You make a good point, Gil, when you say that "...the AC43.13 ground point attachment techniques...should be followed."

But, at the same time, your emphasis on proper ground point attachment techniques also illustrates a potential weakness of airframe grounds. Improperly terminated ground connections or those that suffer degradation due to corrosion and/or loosening of the fasteners from vibration can result in increased resistance at the ground point, which can lead to a variety of problems.

Some people are comfortable with the "way it has always been done", and others choose to use a separate ground wire to minimize potential problems.

Neither way is right or wrong.

 

[fl-mike]

Grounding wars......
Shielding and grounding are a never ending debate among engineers.
I think the term "ground loop" is confusing folks. The ground current is not flowing in a loop through the airframe and shield or return, the "loop" is the total area formed by the electrons going to and from the load. Minimizing the area of the loop is the thing you want to achieve. This is most important for signals with high frequency content, as a loop makes a good antenna (UHF TV antenna). A strobe power supply would not typically need much special treatment unless the supply is drawing large spikes of current due to a crappy input filter design. (Squarish signals have a lot of high frequency content). At high frequencies, the electrons will try to follow a path back to the source that mirrors the route out. This is called the image plane. If that path is broken, the electrons must take a detour and a bigger loop is created. Bad things can also happen when low frequency return currents from one piece of gear decide to hitch a ride on the shield of someone else's cable. This is the main cause of hum in audio circuits and is why it's a good idea to isolate the mic jack from the airframe (the mic signals are very low level and susceptable).

One thing that trips up a lot of folks is the inductance of a wire. That is why using pigtail shield terminations destroys shield effectiveness at higher frequencies. At audio freqencies, they are okay, but anything near RF is a no-no. That's why RF connectors use a 360 deg termination.

Position lights, landing lights, flap motors, etc. are all okay using the airframe as the return. Avionics, low level audio, etc. it's best to provide a dedicated return so you have some control over it.

That's my take anyway. Here's a link to some guys I've spent some time (and money) with.
www.emiguru.com

Nuckolls has good advice in his book too.

 

[gmcjetpilot]

Audio noise

Ditto Gil. This is about being out at the wing tips or tail, with items that don't care about gnd loops. Local ground's
are fine, if you use the "std practices AC" for good gnd bonding to air frame.

Audio & RF receivers are a different story. Noise can come from:

-RFI
-ground loops
-power supply (charging system)

If we did not have an intercom and radio and headsets we would NEVER hear the noise. Noise is an audio thing.
This is why your audio systems grounds must be isolated from air frame ground. Ground loops are a big deal with
audio, having a greater affect, because audio works on milliamps. For other items the (theoretical) ground loop, is not
going to make noise, well noise you'll hear. Brian Chesteen is right and his pics are worth a 1000 words, but in some
cases a (theoretical) ground loops will be moot, a non-issue. It's like the tree that falls in the forest, that no one is
around to hear.

Yes, noise can be picked up "wirelessly" through RFI. That is why good radios and intercoms which use shielding
sheldied wire (RFI solved). A bad local gnd on some device can make RFI noise, basically sparks at the right Freq,
which act as a transmitter. Good local ground bond solves this. There are issues of voltage drop, but an all metal
RV petty much assures an OK gnd, as long as you charging system and battery are grounded to the airframe properly.

Yes, noise picked up through power supply line is lower with modern alternators and voltage regs, but the biggest
culpritin the power line is from the charging system, in my experience. You can put a band-aid on some capacitor
or power supply filters. Make sure your charging system is installed properly (with good grounds).

Personally I like Bob Aeroelectrics central ground bus, but for local grounds out in the tail or wing tip, its fine. On
the other hand if you want to isloate a device from the air frame (gnd), run dedicated ground wires out there, more
power to you (pun intended). It adds weight, but not too much. I'm just not sure if it's value added.

 

[az_gila]

Uncertfied electronics - LED nav. lights

An interesting problem was just solved (after many man-hours) on a local new RV - audio noise depended on ground position (near hangars) and intermittent in the air.

It was traced back to one of the "experimental" LED nav. lights. Apparently the constant current source in the LED electronics was oscillating.
One of the fault finders looked up the component data sheet for the Integrated Circuit driver, and it called out for decoupling capacitors if the supply line was longer than 18 inches... needless to say, the 12 volt power line to the wing tip is longer than 18 inches...
No decoupling existed on the circuit card.

So, consider any source of non-proven electronics in your RV a potential source of interference...

Including things that you might think of as passive, but actually contain electronics.

gil A

 

[Brantel]

Thats why the documentation that comes with these constant current drivers tells you to install a cap if the incoming power leads are longer than 18".

Guess the installer of that system threw out the directions with the trash.

If installed properly, these work just fine.

 

[az_gila]

parts?

Thats why the documentation that comes with these constant current drivers tells you to install a cap if the incoming power leads are longer than 18".

Guess the installer of that system threw out the directions with the trash.

If installed properly, these work just fine.

Isn't the driver part of the electronics circuit card?

The "installer" in this case just put them in the wing tips... he did not make the electronics card. He bought one of the Experimental vendor's RV wing tip LED nav sets.

Or is this a "floating part" in the power leads, and no electronics circuit card exists?

gil A

 

[RV8RIVETER]

Gil

Do you have a link for the info, I probably have the same LED lights and will need to modify.

Thanks,

 

[Brantel]

Isn't the driver part of the electronics circuit card?

The "installer" in this case just put them in the wing tips... he did not make the electronics card. He bought one of the Experimental vendor's RV wing tip LED nav sets.

Or is this a "floating part" in the power leads, and no electronics circuit card exists?

gil A

Not sure about the one you are refering to, mine was homebrew using a common constant current driver which goes inline with the power wiring. Directions clearly state to use the cap if longer than 18".

 

[jdeas]

Osc regs

An interesting problem was just solved (after many man-hours) on a local new RV - audio noise depended on ground position (near hangars) and intermittent in the air.

It was traced back to one of the "experimental" LED nav. lights. Apparently the constant current source in the LED electronics was oscillating.
One of the fault finders looked up the component data sheet for the Integrated Circuit driver, and it called out for decoupling capacitors if the supply line was longer than 18 inches... needless to say, the 12 volt power line to the wing tip is longer than 18 inches...
No decoupling existed on the circuit card.

So, consider any source of non-proven electronics in your RV a potential source of interference...

Including things that you might think of as passive, but actually contain electronics.

gil A

This is something I see a lot of. Analog voltage regulators will Osc if not properly used. The 78xx series we all use has a taste for 70Mhz sqaure waves! This can cause all sorts of phantoms in the rest of your gear due to the harmonics and the noise created on your grounds. If you use 78xx series make sure you have the recommended capacitance connected the recomenneded distance or closer to that regulator.

With the exception of audio gear, I would be comfortable with using the airframe as a return for all solid state devices. This is any system that does not create large current impulses at a regular rate (I.E. lights, flap motor, etc).

BTW: It may be easier for some to look at what we call ground loop as signal induction. Current will always create a voltage drop across a conductor. Skin or wire it will exist. While it is important in audiofile work to keep this at very low levels the same is not true for most gear. Only low signal level or high impedance circuits need special care. These include engine sensors, mikes, etc. On all these items a sheilded line connected on one side will protect the signal integraty without creating 'ground loops'.

** It has been argued for 20 years what side of a signal line is best to tie to ground. The only thing everyone agrees on (but not everyone does) is that only one side of the shield may be tied. This ikeeps current from running through the cable's sheiled and 'inducting' noise into low level or high impedance circuits.

 

[macrafic]

An interesting problem was just solved (after many man-hours) on a local new RV - audio noise depended on ground position (near hangars) and intermittent in the air.

It was traced back to one of the "experimental" LED nav. lights. Apparently the constant current source in the LED electronics was oscillating.
One of the fault finders looked up the component data sheet for the Integrated Circuit driver, and it called out for decoupling capacitors if the supply line was longer than 18 inches... needless to say, the 12 volt power line to the wing tip is longer than 18 inches...
No decoupling existed on the circuit card.

gil A

Gil, would you be willing to email me the vendor's name (not post it)? I have already installed my LED lights and would like to know if I have a problem I have to deal with up front. Also, do you know if the vendor is aware of this problem and if they have fixed it?

 

[Stabird]

Noise of a Different Sort

We had a noise interference problem with the SL30 radio installed in an RV9A. The noise was a strong hash with little high frequency content. The noise comes on with the engine running and turning on the LED Nav lights. What is interesting is the engine must be running for the noise to occur. It does not come on when operating on the battery alone. The NAV lights were individually tested and the noise was isolated to the left NAV light. Note that the wing tip lights were LED type and the tail incandesant.

The LED lights have a constant current 700 ma power supply made by LEDdynamics model # 2008B-700. It is a switching power supply and switchers are known noise generators. The power supply data sheet notes that a capacitor should be placed across the power input leads if they are 18" or longer. The RV leads are at least 13 feet long. Installing a 225-microfarad electrolytic in parallel with a .01-microfarad ceramic has eliminated the radio noise.

The bottom line. The noise signal strength is a function of input voltage to the current regulator. At a battery voltage of 11.4 volts the noise signal was not strong enough to cause problems with the radios. With the engine running system voltage increased to 14.5 volts changing the operating point of the current regulator and increased the noise signal strength.

The noted power input shunt capacitors have corrected the problem. It has been suggested to substitute an 8-ohm, 20-watt power resistor for the current regulator to eliminate the current regulator.

Here?s the link for the current regulator data sheet. www.leddynamics.com/LuxDrive/powerpuck.php

 

[Brantel]

It has been suggested to substitute an 8-ohm, 20-watt power resistor for the current regulator to eliminate the current regulator.

Be carefull designing passive voltage droping circuits for driving LED's. One size does not fit all in this case. Depending on the number of and how you wire your LED's and their forward voltage, this circuit can be very different across different installations. This method also creates alot of wasted energy (heat) from the resistor.

Nearly every modern electronic device made these days has some sort of switching power supply or DC to DC converter in it. They can work fine if properly integrated.

 

[FrankK90989]

do da math

Here is da calculator.
http://www.luxeonstar.com/resistor-calculator.php