What's new
Van's Air Force

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

Failing/Intermittent toggle switch

bruceh

Well Known Member
My Fuel Pump switch is a DPST toggle switch. The one circuit on there runs the idiot light, and the other pulls the VP-X circuit for the pump down to ground to turn it ON. Over the last couple of years I have had the switch occasionally not work to make the pump turn on. The idiot light on the panel will light up, but no sound from the pump. Using the VP-X menu on the Dynon EFIS you can alternatively get the pump turned on. If I get my hand behind the panel and wiggle the connectors back there, it will usually come back to life and work fine. I had every intention of swapping out the toggle switch at this annual, but when I was under the panel I realized that I only had a spare SPST switch, so I figured I'd do the swap at some other point later on.

Today, I went out for a quick flight and for the first time, the wiggling of the connector did pretty much nothing. I guess the switch finally is kaput! Thankfully, the VP-X menu can run the pump ON/OFF (but no idiot light when it is ON). Looks like an order to SteinAir is needed!

I'm just curious as to what the MTBF might be on this toggle switch. These are the good quality Carling switches from Stein, so I would have thought these would last forever.
SA-820-SWITCH-L-DSC01637-2-600x600.jpg

Anyone else had a toggle switch go flaky at 660 hours?
 
https://en.wikipedia.org/wiki/Mean_time_between_failures

Remember, if some don't fail, there would be no meaning to MTBF.

They're good switches; you should be able to look up the MTBF specs on Carling's web site. Their biggest weakness seems to be using them with the old strobe systems, which apparently have surge currents *much* higher than their rated current.

Sounds like you're using one with virtually zero load current; sometimes, as strange as it may sound, no load can actually encourage corrosion on the contacts. But regardless, you having a failure doesn't mean that the MTBF is bad.

One last thing; have you verified that it's truly the switch, & not the crimp or the Faston?
 
I used this same exact switch on my propane truck to activate the PTO. It was switched 60 times a day, 5 days a week for 18 years without a single failure.

If you got your hand behind the panel and wiggled the connectors back there and made it work, it sounds like you have a bad connection on your wire terminal, I don't expect the switch to be bad.
 
Can you put a multimeter on the switch and to the ground block and see if the switch is going to ground? Or swap the connectors from one side of the switch to the other - if the pump then works, but the light does not, then one side of the switch is bad. If the pump still doesn’t work, then it may be your ground or the wire from the vpx to the switch?
 
Like Charlie said, switches need a heavier load to burn away corrosion.
Switches that only carry low level signal current, should have gold plated contacts.
Switches that carry heavier current should have silver plated contacts.
Like people, switches need exercise. Exercising switches wears away corrosion.
Not using switches to make them last longer is counter productive.
Disuse will actually promote corrosion and failure.
Sometimes a failed switch can be revived by connecting a heavy load and
turning the switch on and off several times.
 
About 12 years ago, I posted a forensic analysis of Carling switch failures on VAF and Aeroelectric.

Details are lost in the public forums, but here is the summary:

<deranged rant mode on>

Do not use switches whose terminals use hollow attachment rivets! Stress and vibration will loosen the terminals, leading to switch failure and/or thermal runaway. I had 3 in-flight Carling switch failures: Strobes, Landing/Taxi lights and a Master switch. I replaced all cockpit switches with Honeywell 11TS or 12TS series solid rivet versions, with no further problems.

Use a small coiled service loop on the wires to decouple vibration, but secure the wires firmly in the bundle

Low current (signalling) switches should use Gold contacts. Power switching use Silver contacts.

In my Rocket, I used low current switches everywhere and a bank of relays for power switching, much like automotive systems. These days, vendors are offering electronic load centers that perform the same function and remove many switch failure modes, but add new failure modes.

I am surprised/dismayed/concerned how we are offered 'consumer grade' devices from reputable sources and willingly accept them because they are cheap. At some point, every switch becomes mission critical... a bad Master switch can lead to an alternator overvoltage, which can fry your lithium battery, which can cause a bad day.

Some of my switches I paid almost $40 for, but I don't have to worry.

<mode off>

V
 
Last edited:
Teardown?

Bruce,

Any chance of you doing a teardown to find the specific cause?
If you are not interested in doing that, I would be more than happy to.

I'm curious if it is corrosion, a broken lead, or something else.
Let me know.
 
Do not use switches whose current-carrying rivets go through plastic. That plastic will flow over time, loosening the rivets. That results in high resistance which makes heat and more resistance and more heat.
 
Bruce,

Any chance of you doing a teardown to find the specific cause?
If you are not interested in doing that, I would be more than happy to.

I'm curious if it is corrosion, a broken lead, or something else.
Let me know.

When I tore down my failed Carling switches I found the contact rocker had become banana-shaped from overheating. Loose rivets--overheating--looser rivets--more overheating--heat, smoke, failure.
 
VDC Rated Switch

Your switch is rated for alternating current (AC) only. Because the voltage level never crosses zero volts (sinusoidal wave) with DC current, arcing occurs each time the switch is thrown. This creates heat and eventually burns out the contacts. Find a switch that is rated for DC current. It will have much heavier contacts.
 
While theoretically correct, that principle has little application at our voltage levels. If you look at the ratings for switches with both AC & DC current ratings, the 12V DC rating typically is the same as the 120VAC rating.
edit: Should add, the DC rating has more to do with how fast the contacts separate (to quickly extinguish the arc), than the size of the contacts.

But more to the point, if you look at the OP's application, there is virtually zero current being handled by the switch.
 
Last edited:
I agree with Charlie. All switches are capable of carrying DC current. We just do not know how much current unless the switch manufacture publishes the DC rating. As a rule of thumb, a snap action switch will carry the same current at 12 VDC as it can at 120 VAC. What about de-rating? Reducing the voltage by a factor of 10 (120 down to 12) automatically de-rates the switch.
Look at the switches that B&C sells. The AC current rating is published, but not the DC rating. Many of those AC switches have been installed in aircraft.
READ what Bob Nuckolls has to say about switch ratings.
 
This is a fun hangar discussion, but to the OP, dont worry about it. Buy another switch and move on. Nothing is fool proof; sh!t happens. Take apart the old one and see what the deal is. Educational at the least; reparable and reusable at the best. Its not rocket science.

FWIW, I always buy an extra or two of whatever switch I'm using where-ever so I have it on hand to replace a bad one, tho in my panel design, I tried to stick to one on/off type toggle so if I have to make a dark and stormy night field repair, I can at worst rob one from peter to make paul work again.
 
I am using all relays. For some EFI functions will have a small protected panel with 6 mini toggles that I already bought. They have silver contacts. Anybody know a good source for miniature toggles with gold contacts?
 
I got an idea, why not use a high quality switch to start with and avoid this whole mess?
Carry spare switches... you got to be kidding!
Relays.. they depend on a switches to control them, and now you just added a bunch of additional failure points.
 
Last edited:
Carry spare switches... you got to be kidding!

Not sure if that was meant in response to my comment, tho to clarify my spare switches are in the parts bin not the plane.

That said, carry whatever you think you need as a spare....spare oil, spare inner tube, spare food, spare spark plug, spare fuel, spare pilot...as long as its well thought out, each to his own.
 
svyolo,

Here's one source:
https://www.newark.com/w/search/prl/results?brand=nkk-switches&st=miniature%20toggle%20switch
The NKK switches can be ordered with either gold or silver contacts.

FWIW, the builder of my RV-4 used miniature toggles of unknown pedigree for mag switches. Never a problem since 1991. I've used them in all sorts of stuff for all my adult life, never paying any attention to pedigree, and can't recall one failing when used within its ratings.

Having said that... while there are several legitimate reasons for using relays (shorten the hi-current wire run, panel space for switches, etc), it does tend to drive reliability in the wrong direction.
 
One thing I am a proponent of is putting an extra PTT switch in the panel on the pilot's side. There's almost no cost in terms of complexity. I had one fail on me not too long ago while flying with a student and it caused some headaches for a few days until it could get sorted out.
 
One thing I am a proponent of is putting an extra PTT switch in the panel on the pilot's side. There's almost no cost in terms of complexity. I had one fail on me not too long ago while flying with a student and it caused some headaches for a few days until it could get sorted out.

Otto, Otto, Otto for PTT (P1 rolling sleeve goodness). As Walt said, cry once and never regret spending a couple extra $. Do not use those crappy butt contact Radio Shack type push buttons.

Trust your ordinance...I mean radio...to nothing else!

https://www.otto-controls.com/p1-sealed-mil-prf-8805-3-momentary-snap-action-pushbutton
 
follow up

I just wanted to follow up on this post. I finally got around to swapping out the old switch with an identical new one and all is good now. The original switch was flaky around 10% of the time and recently started being flaky more like 25% of the time.

I did open up the old switch and it looked fine inside. The one side that drove the fuel pump signal to ground for the VP-X had just a tiny dot on the otherwise clean metal blade inside the switch. That might have been the issue? It rubbed right off, so I don't think it was any sort of hard corrosion. Sorry, I didn't take any pictures, but I still have the switch laying around.
 
My money is on the connector to the tab (Faston, probably)...either loose crimp or, as suggested, build-up of corrosion. I'll bet if you had remove it and re-connected it a time or three, it would have fixed the problem. :)
 
So there may be an issue with fast-ons in low current applications. There may not be enough current to break through surface corrosion. I don't know the design of the VP-X, but what I do when I use switches to control my electronic gizmos is provide 'wetting current'.

This usually means a 3K to 10K pullup resistor. If driving a relay, the coil provides the required current.

Perhaps Vertical Power can comment on this.
V
 
Back
Top