What's new
Van's Air Force

Don't miss anything! Register now for full access to the definitive RV support community.

Redundancy, things to think about

cajunwings

Well Known Member
Good friend was out flying his plane last week and his recently installed Garmin panel upgrade (G3x, 175, 507 ect) went dark. G5 stayed up as intended on its own backup battery. Wasn’t a big deal as the Wx was good and was close to home, however this plane does IFR fairly often. This setup has a avionics master switch with a associated relay and the relay is what failed. My vote to simplify and beef up this arrangement is to replace switch & relay he has with just a heavy duty switch with locking toggle or a guard. What say ye forum wizards? I’m closing in on my own Garmin panel install.

Don Broussard
RV9 Rebuild in Progress
57 Pacer
 
Either that or drop the avionics bus completely. Electronics made in the last decade or more are very tolerant of voltage spikes during engine start.

Being an IFR pilot, he should also consider installing the optional backup battery, as mention in the G3X install manual. (TCW IBBS-12V)
 
Last edited:
If a person wants an avionics master, that is up to them - but I put all EFIS components (displays, AHRS, EIS, etc) on breakers directly off the main buss. In an airplane with a glass cockpit, you’re not going flying if the panel is blank, and all of the new electronics can take starting transients without damage (they might reboot, but you can fix that with an Aux battery to the alternate power feeds.

I sometimes build with an avionics master, but that is for radios, transponders, and the like. And there really isn’t a great reason for doing that anymore, other than “we used to do it that way.... ;)
 
Yes, the avionics relay is a failure point and we have seen a few of those. Most were workmanship issues.
I don't like using the automotive "headlamp" relay for the avionics. I've always used another high quality continuous duty solenoid.
I've also placed an avionics relay bypass switch, just in case.

Some thoughts...

It's nice to be able to listen to the ATIS and call for clearance prior to engine start. However, most avionics won't stay powered up during an engine start. Yes, they are probably more resilient to spikes these days, as someone mentioned, but to me, it seems hard on them to cycle during an engine start. Plus, I've seen some of them that don't necessarily come up properly when they have been power failed in that way.

IF you are thinking of replacing the avionics with a switch. make sure it is an appropriately-rated DC switch. Internally they are a little different. AC power will cleanly break at the switch due to the sine wave. DC current will arc across the contacts, wearing them until they fail.

Some avionics circuits are still quite large in amp requirements for IFR panels, needing 10-15 amps, so plan accordingly.

Vic
 
That's what I have, heavy duty switch that goes to an avionics fuse panel (spade fuses).
If all goes dark I have a Garmin handheld with its own battery, wired into an independent ship-mounted antenna and keepalive power, and a handheld radio that can be plugged into the ships antenna via a cutout plug.
 
There is no such thing as a voltage spike during engine starting. That is an old
wives tale. The voltage SAGS during engine starting. That voltage sag did
cause transistors to fail back in the 1950s. That problem was fixed 50 years
ago. It is not necessary to shut off avionics while cranking the engine. I have
started my engine hundreds of times without shutting off avionics.
 
How much easier/lighter/heavier would be putting in a few “classic” instruments for backup vs batteries? Blasphemy?
 
I offer that replacing the obvious Avionics Master relay/switch single point failure risk with a Master Solenoid/switch single point failure risk is equal to rearranging the deck chairs on the Titanic.

Avionics redundancy and continued IFR flight after any power component failure are not exclusively the same animal.

Carl
 
It's nice to be able to listen to the ATIS and call for clearance prior to engine start. However, most avionics won't stay powered up during an engine start.

One of the benefits of using a dual battery setup is the magic glass and such do not brown out when I start the engine.

Just a data point for those thinking of using a second battery.
 
No avionics relay for me. I used an MS35058-22 switch from Aircraft Spruce. It's rated at 25A resistive, 15A inductive, and 7A lamp load. I used an 18V MOV across the contacts, as I did for all switches.
 
There is no such thing as a voltage spike during engine starting. That is an old
wives tale. The voltage SAGS during engine starting. That voltage sag did
cause transistors to fail back in the 1950s. That problem was fixed 50 years
ago. It is not necessary to shut off avionics while cranking the engine. I have
started my engine hundreds of times without shutting off avionics.

And the only "spike" to worry about these days is from the collapsing field when power to an inductive load is shut off -- Thus the starter relay, starter solenoid, starter motor, master relay, electric fuel pumps, etc. all should have a diode or MOV to move this energy to ground - thus eliminating the spike.
 
There's no easy answer to this question although there are plenty of options. As discussed avionics relays/switches introduce single point failures. Doubling up seems inelegant. Multiple batteries can be a maintenance headache. I went for a switched connection between battery and half the bus, with a large diode to provide power normally. But that has limitations as the bus voltage is always lower than the alternator voltage. Everything has its own fuse, those items that don't have an on/off switch have a breaker. Decide what is important and build your system for that, no ideal solution that suits everyone.
Pete
 
As we saw in a thread last week, even simple switches can fail. I believe that redundancy is the key to safety, followed by use of components with lower MTBF rates. While I have no data, I would speculate that switches have a lower failure rate than relay/switch combinations.

In my 10, I have one relay/contactor for each of two batteries, feeding each buss. They have a L/R/Both switch. Still a common failure point with the switch, but have redundancy on the relays and batteries. I have a separate engine buss, to allow buss shutdowns without taking out the EI.
 
Last edited:
My installs and my own aircraft:

1) Back up (TCW battery or your choice) to keep the PFD/AHRS/ENG monitor all running normal in case of electron loss.

2) I only use Honeywell MIL spec switches for avionics/radio bus power, no relays needed or required.

3) G5 with its own battery

(the MS35058-22 someone posted above is not a high quality sw IMO)
 
No avionics relay for me. I used an MS35058-22 switch from Aircraft Spruce. It's rated at 25A resistive, 15A inductive, and 7A lamp load. I used an 18V MOV across the contacts, as I did for all switches.

What is a MOV?

I am strictly vfr but I will have my eris on the backup battery for startup, then will switch it to the main bus. No avionics master, as per nuckolds. But if it all goes dark I will look out the window. I am quite comfortable flying without an ASI. I had a great instructor.
 
I like avionics masters, thats me. my RV-4 has one, only one. again, thats me, i don't fly it IFR. but it does have two separate supply systems for the EIs. that is a must have. there is no right or wrong answer, except that if one thing fails in IFR and takes out needed systems thats wrong. If you like avionics masters great, if your going to fly IFR, put in a backup. its not that complicated, one more switch, one more relay. I also do not like the cheap auto relays, a good ice cube is not that much more.

bob burns
RV-4 N82RB
 
I'm not a fan of MOVs for spike suppression.
I prefer a diode or TVS.
MOVs degrade with each spike and will eventually fail to a shorted condition. For situations where a spike is a rare occasion, then MOVs are not bad, but not in a repetitive spike application.
 
No avionics buss for me. Straight off the main buss. One less point of failure.
 
(the MS35058-22 someone posted above is not a high quality sw IMO)

Agreed. I'd say those are "entry level" mil spec in that they have a mil spec, but an MS24523 (Honeywell 1TL commercial type) are much higher quality.
 
Here's a copy of the MS24523 specification. The electrical loads that the switches are rated for are on page 3 in Table I.

View attachment MS24523J.pdf

One thing interesting is that the life or reliability of the switches does not seem to be specified.

Dave
 
I'm not a fan of MOVs for spike suppression.
I prefer a diode or TVS.
MOVs degrade with each spike and will eventually fail to a shorted condition. For situations where a spike is a rare occasion, then MOVs are not bad, but not in a repetitive spike application.

Which TVS diode would you recommend for a DC switch in a 14V system?
 
One other approach not yet mentioned is to use the normally closed (NC) contacts on the avionics bus relay, with a Honeywell switch that powers the relay up in order to open the circuit to the avionics bus. That way a failure of the avionics switch will result in the relay contacts closing when power is lost to the relay coil.

Since this is still not a perfect solution (wires can break or come loose with any installation), I have the IBBS battery providing power to my critical avionics, have a second standby alternator (driven off the vacuum pad), and also have the separate G5 battery to keep that instrument alive in any full power system outage.

Belts and suspenders for the avionics, plus an old fashion mag to keep the engine running.

Just another approach in our wonderful world of Experimental Aircraft.
 
Which TVS diode would you recommend for a DC switch in a 14V system?

I'd just use a regular diode, but if you want to use a TVS, anything between a 1.5KE18CA or 24CA should be fine for protecting the switch contacts. Only saying a 1.5KE because the leads are big and easier to work with, and a CA (bidirectional) so you don't have to worry about orientation.

Again, a plain old diode is proven and works well. 1N4003 or 1N5402 will serve you well and are cheap!
 
FWIW... the STC for G3X Touch in certificated aircraft requires that PFD 1, the ADAHRS/magnetometer/OAT probe, and EIS are wired to the Essential (hot) Bus. They were designed to function just fine on there despite engine starts, and it cuts out the opportunity for failure that an avionics bus introduces. If you have a GAD 27, it can function as a keep-alive for your EIS during engine start. For Experimental/LSA I'd recommend using a TCW backup battery, just from prior favorable experience.

For the OP... Glad to hear the G5 did its job! :eek:
 
I'd just use a regular diode, but if you want to use a TVS, anything between a 1.5KE18CA or 24CA should be fine for protecting the switch contacts. Only saying a 1.5KE because the leads are big and easier to work with, and a CA (bidirectional) so you don't have to worry about orientation.

Again, a plain old diode is proven and works well. 1N4003 or 1N5402 will serve you well and are cheap!

Thanks you!
 
I scratched my head quite a bit while pondering this scenario. I ended up with an Essential bus, powered by its own battery. The battery is charged via a circuit breaker and Schottky diode from the main bus. Output of the battery is switched by a DPST switch thus I've halved the load through each switch in the package, or perhaps doubled the reliability of the switch, however you wish to consider it.

The Essential Bus powers the GRT Mini-X (which also has an internal battery), the GRT HX, the GRT remote EIS and the GNS480. All of these units are brought on line as the first items on the pre-start checklist thus they are all up and ready by the time I want to crank the engine. With their power being provided provided by the Essential Bus Battery none of these devices even blink during engine start.

I also have an avionics master with power being controlled by a heavy-duty 40A-rated relay. The relay conducts power in the "normally closed" position thus provides fail-safe operation. Turning the avionics master switch to the "off" position energizes the relay, moving its contacts from the Normally Closed to the Normally Open position, interrupting power to the avionics bus. This is about as failure tolerant a design as I could come up with and in fact mirrors similar fail-safe power designs in transport-category aircraft.

The Avionics bus powers the audio panel, SL30, transponder and ADSB-In receiver as well as the autopilot and its servos.

While I can't believe this design is bullet-proof I have to believe it's pretty bullet-tolerant. So far in a couple of hundred hours of operation it hasn't given me any cause to believe a redesign is necessary.
 
Good friend was out flying his plane last week and his recently installed Garmin panel upgrade (G3x, 175, 507 ect) went dark. G5 stayed up as intended on its own backup battery. Wasn’t a big deal as the Wx was good and was close to home, however this plane does IFR fairly often. This setup has a avionics master switch with a associated relay and the relay is what failed. My vote to simplify and beef up this arrangement is to replace switch & relay he has with just a heavy duty switch with locking toggle or a guard. What say ye forum wizards? I’m closing in on my own Garmin panel install.

Don Broussard
RV9 Rebuild in Progress
57 Pacer


Years ago, I did an avionics panel upgrade for the flying club. This included an IFR KLN-89B (the first in the PNW according to the FSDO) and dual KX-155s. The consensus was to include an "avionics master switch".

This single point failure mode was a great concern then. So, we installed a 20amp "breaker switch". We installed something like a Potter & Brumfield W31-X2M1G-20 as I recall.
 
I scratched my head quite a bit while pondering this scenario. I ended up with an Essential bus, powered by its own battery. The battery is charged via a circuit breaker and Schottky diode from the main bus. Output of the battery is switched by a DPST switch thus I've halved the load through each switch in the package, or perhaps doubled the reliability of the switch, however you wish to consider it.

The Essential Bus powers the GRT Mini-X (which also has an internal battery), the GRT HX, the GRT remote EIS and the GNS480. All of these units are brought on line as the first items on the pre-start checklist thus they are all up and ready by the time I want to crank the engine. With their power being provided provided by the Essential Bus Battery none of these devices even blink during engine start.

Virtually identical to how my 10 was setup, except I also had a SD8 feeding the second battery direct. Very similar panel also.

Worked fine for the 6 years I had the plane, and the new project is getting the same setup.

Either battery powers the panel and engine, either battery is charged from the main alt, backup alt is dedicated to backup battery.
 
Good friend was out flying his plane last week and his recently installed Garmin panel upgrade (G3x, 175, 507 ect) went dark. G5 stayed up as intended on its own backup battery. Wasn’t a big deal as the Wx was good and was close to home, however this plane does IFR fairly often. This setup has a avionics master switch with a associated relay and the relay is what failed. My vote to simplify and beef up this arrangement is to replace switch & relay he has with just a heavy duty switch with locking toggle or a guard. What say ye forum wizards? I’m closing in on my own Garmin panel install.

Don Broussard
RV9 Rebuild in Progress
57 Pacer

Almost all of the G3X LRUs have a Power 2. The easiest/safest way to fix this is to install a backup battery that powers the equipment though Power 2, and make sure the backup battery is charged from the main bus. The GPS 175 doesn't have a power 2 so you need to power that off the main bus or pass-through power from your backup battery.

I highly recommend TCW IBBS! I just installed the 3ah version and it powers the following for 1 hour before reaching the 9.5V cutoff, with screens on full bright...

  • GDU 370
  • GDU 375
  • GSU 25 #1/GMU11
  • GEA 24
  • GAD 29
  • GTX 335
  • GPS 175
  • GMC 307

I didn't put my COM on the backup battery because transmissions are a big load, and I carry a handheld nav/com for backup.

With wiring, this unit is only 1.3 lbs, and uses stable LiFEPO4 cells. The added benefit is the ability to boot up your avionics prior to engine start and not have them reboot from the voltage drop during cranking (brownout).

I also have an Essential bus that is normally powered off the main bus, but can be isolated and powered directly off the battery with up to a 10A load. I did this instead of a separate generator that added weight, because I can fly around for about 4 hours on a healthy battery alone.
 
Last edited:
Back
Top