What's new
Van's Air Force

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

A different relay question

svyolo

Well Known Member
I have a different relay question, maybe a bit esoteric. I am going to run EFI/EI and will have some switches to control pumps/ignitions/etc. They will only turn relays on and off, no power switches.

I was thinking of using SPDT relays. They have an "87" and "87a" terminal. With the coil unpowered, the power goes to 87a. With the coil energized, the power switches to 87.

I was thinking in "normal" ops it would be nice to not need any trons to hold relays closed. I would like most of my switches and relays unpowered, when powering the EFI. Any problem with using 87a instead of 87? Is this terminal less robust, or not designed for continuous use?

Thanks in advance for any response.
 
For most relays, the 87 and 87a terminals are rated the same. To turn the 87a output off, the relay needs voltage applied across 85/86. Which means the relay will be drawing power to hold it 'off'.
 
And if it was me - I wouldn't use relays in a EI/EFI system in the air (automotive applications are a different story). A properly rated switch without a relay is going to be more reliable. Adding another component such as a relay where it isn't really needed is only going to increase potential failure chances.
 
Consider what happens if the relay coil or its circuit fails. Will the load turn on with undesired results?
The normally closed contacts of some relays have a lower ampere rating than the normally open contacts do.
 
And if it was me - I wouldn't use relays in a EI/EFI system in the air (automotive applications are a different story). A properly rated switch without a relay is going to be more reliable. Adding another component such as a relay where it isn't really needed is only going to increase potential failure chances.

I would extend this statement to say 'a high quality properly rated switch'. Many of the switches used in E-AB Aircraft and sold by reputable suppliers have a high failure rate.

In my rocket, I eliminated most load carrying switches and used relays instead.
For the electronic and P-lead ignition, I used $$ Honeywell switches and these are modest loads.
 
I have had very bad luck with switches of all types in lots of applications for a very long time. Like decades. I have had very little problem with relays. I think I have only replaced a couple of relays in my life. I think I will put my trust in relays, and will probably make sure I use Bosch/Tyco types used in cars.

I like the idea of everything running with no current running through any switches, hence my question about 87a. If the load path powering the switches fails, the wire fails, etc, all the relays are still powering their loads. I only turn something off, by using amps to power the coil and close the really.
 
And if it was me - I wouldn't use relays in a EI/EFI system in the air (automotive applications are a different story). A properly rated switch without a relay is going to be more reliable. Adding another component such as a relay where it isn't really needed is only going to increase potential failure chances.

Are you aware that most every GA airplane flying today has it's entire electrical system controlled by a relay (A contactor is nothing more than a relay designed for high amperage loads by using a plunger inside the winding instead of the winding actuating a lever).
 
Are you aware that most every GA airplane flying today has it's entire electrical system controlled by a relay (A contactor is nothing more than a relay designed for high amperage loads by using a plunger inside the winding instead of the winding actuating a lever).

Yes, but they still fail. Which is why critical systems run direct from a battery bus, not through the contactor.

The OP was asking about using relays to run specific critical systems. I'm not saying relays/contactors don't have a place in aircraft, I'm saying that they should only be used if they are actually required.

Using a relay to control a fuel pump/EI/EFI system just isn't required.
 
Last edited:
My thoughts on using relays is my experience with failed switches. Here is the way I want to use relays in my EFI/EI.

There will be a bus dedicated to running the engine. That is the only bus that needs to remain powered to keep the airplane airborne. The relay panel will be set up so the power comes in, goes through a fuse before each relay, and the relay is a SPDT, and the normal configuration is for the relay coils to be normally unpowered, and the current flows through the 87a terminal. In other words the relay doesn't have to function. It just has to sit there. For the 2 fuel pumps, 2 ignitions, and 2 alternators, there will be miniature switches that energize the relay coils, to turn the associated circuit "off", to test the function of its backup. Or in an emergency. But if those switches, or the power to them fail, there is no change in the normal function of the engine bus. They are only for emergency or diagnostic use.

I know common sense says a simple switch should be more reliable than a relay. By the same logic a Model T should be more reliable than a Honda Accord. But in both cases, that has not been my experience.

I am trying to design the normal mode of my EFI/EI bus to be as simple, automatic, and bulletproof as possible. The only switch that will be critical, is the one that turns power on to that bus.
 
Are you going to use one or two relays to power pumps 1 and 2? Either way, if you use a ?normally closed? relay, power is going to pass through it (87a) and supply power to ?pump 1? when your toggle switch is in the pump 1 position and only be ?energized? when you put your toggle switch in the ?pump 2? position and that?s when the ?call wire? calls for the relay to be activated which will then supply power to pump 2 (87). So, essentially a standard ?normally closed? relay is doing exactly what you?re wanting it to do.

I?m actually designing mine with 4 relays along with an oil pressure safely switch that will automatically turn the fuel pumps off in the event of an emergency and sudden engine stoppage. Hopefully I can describe this in words where it makes sense. For starters, I?m planning on using the ?bus manager? in my setup. I know there?s been some discussion about the bus manager and some of it?s possible failsafe shortcomings. I won?t go into that discussion but I have a few ideas that I?m working on that will alleviate any concerns regarding ?emergency power?. What I really like about it is it?s ability to manage automatic switching of fuel pumps 1 and 2 in the event of a fuel pump failure and its ability to manage two batteries. So, enough about the bus manager and back to the fuel pump relay(s). First off I?m incorporating two separate circuit breakers (one for each fuel pump) in the essential bus. Each fuel pump will have its own relay but they will be wired so that either relay will be able to run either fuel pump. So basically, the (NC) position (87a) on each relay will be wired to pump 1 and the (NO) position (87) will be wired to pump 2. This way, if either one of the relays were to fail for any reason the other relay would be able to handle the job of running either one of the pumps. Now, the second two relays will power the first two relays. I know this sounds crazy, but bear with me. As I mentioned above, I?ll have a C.B. for each fuel pump, so each relay will have its own separate power which will feed the two relays that are actually operating the fuel pumps as discussed above. So these second two relays will just be on/off relays with no need for the ability to switch between (87a) and (87). These will both be normally open (NO) relays and will close when the ?call wire? calls for them to be closed. So, as mentioned above, I?ll also be incorporating an oil pressure switch into the system. The ?call wire? to activate pump 1 will be routed through this simple automotive oil pressure switch which will cut power to that relay in the event of an accident and sudden engine stoppage which will shut the fuel pumps off immediately and automatically. All modern day cars have this safe feature built in to them. Remember, pump1/auto is the ?normal? operating position. I will have the ?call signal wire? for the pump 2 toggle switch position wired so that when the toggle switch is in the pump 2 position, it will be able to energize the call wire and bypass the oil pressure sender. So basically, when the toggle switch is in the pump 2 position, it will allow for the priming/pressurizing of the fuel rail and allow staring of the engine when there?s no oil pressure present yet. Once the engine had stared and there?s oil pressure present, then both fuel pumps will work as normal with the ability to switch between them both manually and automatically. Remember though, this safety feature is only valid when the toggle switch is in the pump1/auto position, which is the normal operating position and its bypassed when the switch is in the pump 2 position. I hope all of that made sense and if anyone can think of why I shouldn?t design it like that, please let me know.

Mark
 
Mark you lost me 3 sentences into the second paragraph. I would need to see a diagram. But in general I agree with your logic.

I am going to use a separate circuit, fuse, and relay for each FP, IGN, and and ECU. All relays will be normally "off", which will power the 87a circuit. The only time an 87 circuit will be energized, is if FP2 is being used, or I am checking the functionality of another circuit and associated equipment.

I will wire it so that if current is lost to FP1, that FP2 automatically turns on, and also have a switch to manually turn FP2 on. I will also have a small backup battery on a diode and a parallel switch to connect that battery to the engine bus. I think that will be the limit of my "automation".

I thought about the VP-X and the Bus Manager. I believe it is overkill for my airplane which will be very simple other than the EFI/EI. Part of me is also a little uncomfortable with the technology. If I was building a hard IFR airplane I would definitely want more bus automation.

Part of me likes the logic of the oil pressure shutting off the FP's. Part of me would worry that would cause the FP to shut off at an inadvertent time. Newer motorcycles and scooters have a kickstand kill switch. If you put the kickstand down, the engine stops. It won't start if you have it down. I can't tell you how many times I have cycled the kickstand 8 times to try to get the engine to start.

But, I don't like fire either.
 
Yeah, I thought my explanation might get a little murky without a diagram. LOL!! I?m really not sure how your going to wire your relays up so that both 87a and 87 will be activated if power is lost to the relay?? In a (NC) relay 87a will have its 12v passing through it to pump 1 without the coil being energized....but that coil will have to be energized for 87 to become energized, so I?m really not sure how you?re going to accomplish both tasks without power being applied to activate the coil??

I agree, the VPX is a nice unit, but I think it?s overkill. I?m planning on C.B.?s for the essential bus and a simple automotive fuse panel for the main bus. However, take another look at the bus manager. It?s a pretty nice unit and it really simplifies things. I know it?s not for everyone, but it may be worth a second look. I think you said that you?re planning on a two alternator setup along with a small backup battery?? IMO I think just a good solid backup battery is all you?ll need. It?ll be simple to install and maintain and it?ll buy you plenty of time to get on the ground. Personally, I?m going with one 680 on the firewall and one EarthX remotely mounted behind the baggage bulkhead (I?m building a 7). The EarthX will be there for emergency power only and will not be part of the staring power....so I won?t have to run large gauge wiring toward the back of the airplane. You?re not suppose to mix and match led-acid and lithium batteries together in your electrical system, but the bus manager is able to accommodate this configuration.

Regarding the oil pressure switch, I understand your concerns. However, if wired the way I?m planning on, when you flip the toggle switch into the pump 2 position it bypasses that oil pressure sender altogether and the relays are energized as normal. I?d be glad to talk to you on the phone if you?d like for me to clean up any confusion on how I?m planning my installation. You can end a PM to me if you?d like to chat.

Mark
 
I’m planning on using the “bus manager” in my setup.....What I really like about it is it’s ability to manage automatic switching of fuel pumps 1 and 2 in the event of a fuel pump failure...

Someone really needs to settle the "Cannot run two electric pumps at the same time" question. If there is no issue with two pumps running, automatic pump switching (via Bus Manager or otherwise) is an unnecessary complication. The pilot simply switches the aux pump to ON for takeoff and landing.

First off I’m incorporating two separate circuit breakers (one for each fuel pump) in the essential bus. Each fuel pump will have its own relay but they will be wired so that either relay will be able to run either fuel pump. So basically, the (NC) position (87a) on each relay will be wired to pump 1 and the (NO) position (87) will be wired to pump 2. This way, if either one of the relays were to fail for any reason the other relay would be able to handle the job of running either one of the pumps. Now, the second two relays will power the first two relays. I know this sounds crazy, but bear with me. As I mentioned above, I’ll have a C.B. for each fuel pump, so each relay will have its own separate power which will feed the two relays that are actually operating the fuel pumps as discussed above. So these second two relays will just be on/off relays with no need for the ability to switch between (87a) and (87). These will both be normally open (NO) relays and will close when the “call wire” calls for them to be closed. So, as mentioned above, I’ll also be incorporating an oil pressure switch into the system. The “call wire” to activate pump 1 will be routed through this simple automotive oil pressure switch which will cut power to that relay in the event of an accident and sudden engine stoppage which will shut the fuel pumps off immediately and automatically. All modern day cars have this safe feature built in to them. Remember, pump1/auto is the “normal” operating position. I will have the “call signal wire” for the pump 2 toggle switch position wired so that when the toggle switch is in the pump 2 position, it will be able to energize the call wire and bypass the oil pressure sender. So basically, when the toggle switch is in the pump 2 position, it will allow for the priming/pressurizing of the fuel rail and allow staring of the engine when there’s no oil pressure present yet. Once the engine had stared and there’s oil pressure present, then both fuel pumps will work as normal with the ability to switch between them both manually and automatically. Remember though, this safety feature is only valid when the toggle switch is in the pump1/auto position, which is the normal operating position and its bypassed when the switch is in the pump 2 position. I hope all of that made sense and if anyone can think of why I shouldn’t design it like that, please let me know.

Ok.

Big picture...it's a complicated system to incorporate a "safety feature" (the oil pressure switch) which is largely useless in the practical view, and brings its own failure modes to the party, and adds operational complication. Let's start with "useless".

In general, there are only two conditions with zero oil pressure. You're either (1) out of oil, or (2) the prop (thus the pump) has stopped turning.

Assume the out-of-oil case...a popped hose, cooler, whatever. Why would you want to surrender control of when and where the engine shuts down? If it is running, it extends the glide. It won't seize immediately.

Assume any other reason why the engine has stopped making power. As long as the prop is windmilling, you'll have oil pressure. The fuel pump will run all the way to impact, unless you shut it off manually. Given that MASTER OFF, FUEL SELECTOR OFF, and/or PUMPS OFF are all part of a reasonable forced landing checklist, why an auto system?

Crash with power? The fuel pump will be running until prop strike; the system will be pressurized...just like any other fuel delivery system.

At impact, it's worth remembering you're not in a car. Even with the pumps stopped, the unfortunate reality is that we sit with 20 gallons or so at each elbow, and the tanks are not particularly impact resistant. Impact with 40 psi (or 25, or 3, or 0) doesn't change the overall fire odds very much, if at all.
 
Yeah Dan, my only goal with the oil pressure sender/switch is to instantly shut the fuel pumps off and to stop fuel from being sprayed all over a hot engine in the event of a crash and subsequent fuel rail or fuel line rupture. I know there?s an infant number of scenarios that would have a pilot swimming in gas, but having the fuel pumps shut off instantly without any input from the pilot seems like a logical safety feature to incorporate if it can be done without introducing too much complexity.

Mark
 
Mark;
I was going to run the two FP's on two separate circuits for redundancy. For the most part, the primary functions of all the EFI/EI will be run through a dedicated 87a circuit. 87 will only be used in these cases to turn 87a off, to turn that circuit off to test its' backup (FP2 and IGN2)

For the most part, insert the key, turn to start, and you are ready to go flying. No switches to flip unless I want to start with the avionics master off, or I want to check both ignitions.

For electrical power redundancy I will only use 1 alternator, with one or the other of the following:

1. 1 start battery, 1 very small secondary battery as a 'load" for the alternator. I will have two VR's wired, and switchable. The VR's fail a lot more than the alternators (B&C). This should keep me flying.

2. 2 full size batteries. 1 VR wired. 1 identical carried as a spare. 2 batteries should run my EFI for 70-90 minutes.

I am building a day VFR bush plane. I don't need fail operational redundancy. It would be nice to get home under my own power.

DanH? Your opinion is always welcome.
 
Mark;
I was going to run the two FP's on two separate circuits for redundancy. For the most part, the primary functions of all the EFI/EI will be run through a dedicated 87a circuit. 87 will only be used in these cases to turn 87a off, to turn that circuit off to test its' backup (FP2 and IGN2). For the most part, insert the key, turn to start, and you are ready to go flying. No switches to flip unless I want to start with the avionics master off, or I want to check both ignitions.

The operational simplicity is admirable, and the use of NC is a reasonable approach to runup pad functions. The catch is that one key. It requires powering the EFI/EI via the master contactor, which means (1) you can't kill the master in flight, and (2) it's only as reliable as the master system.

I'm an advocate of separating airframe electrical demands from engine electrical demands. Dual engine demands get dual battery-direct power.

So...

For electrical power redundancy I will only use 1 alternator, with one or the other of the following:

1. 1 start battery, 1 very small secondary battery as a 'load" for the alternator. I will have two VR's wired, and switchable. The VR's fail a lot more than the alternators (B&C). This should keep me flying.

2. 2 full size batteries. 1 VR wired. 1 identical carried as a spare. 2 batteries should run my EFI for 70-90 minutes.

...I vote for option #2, a pair of batteries, each with a hot bus. Go with EarthX if you like, as the B&C VR includes OV protection.

Remember, if each battery will run its EFI/EI for 70 to 90 minutes, you actually have something like 70 + 60 = 130 minutes of flight time. You'll run master ON and both engine systems for departure and initial climb, then kill the master and one EFI/EI for cruise, holding BATT2 in reserve.
 
Last edited:
Someone really needs to settle the "Cannot run two electric pumps at the same time" question. If there is no issue with two pumps running, automatic pump switching (via Bus Manager or otherwise) is an unnecessary complication. The pilot simply switches the aux pump to ON for takeoff and landing.

Not an issue, *if* the regulator & the bypass/return line is sized correctly. Easy to do if you purchase Walbro or other quality pumps, and select the right regulator. This appears to be the one pictured on the SDS web site:
http://www.borlainduction.com/fuel-pressure-regulators.html
Claims to be good for over 100gph, using -6 fittings. The Walbro GSL393 pumps 155L (40gallons) per hour. So two Walbros would be well under the capacity of the Borla regulator. A lot of aftermarket regulators will handle a lot more volume than the Bora (though it shouldn't be needed).

I have no idea whether the hi-$ pre-assembled pump/regulator combos will handle both pumps at once.

Charlie
 
Claims to be good for over 100gph, using -6 fittings. The Walbro GSL393 pumps 155L (40gallons) per hour. So two Walbros would be well under the capacity of the Borla regulator.

IIRC, Mr. Paisley's kits were delivered with an Aeromotive regulator. A quick look at Aeromotive's FAQ's says that if the regulator/return won't handle the volume (from one pump, or two), the system just needs a larger regulator, not automatic pump switching.
 
pumps

You guys are WAY overthinking this.
If you want to have some kind of backup power available to a pump, just have a switch that can put the power to the pump directly. The standard Bus Manager pump configuration works very well and has never failed to keep a pump running.

It's best if you don't run two pumps at the same time - this is 70gph running through your fuel system. This is more than a -6 system should be flowing.

Robert
 
Thanks for all the input to everyone that responded.

I like the idea of the Bus Manager and VPX, but am just not knowledgeable enough about them to feel comfortable using them. I think I am most uncomfortable have multiple layers of software driven devices interacting with each other. I want at least one analog layer between each software driven device. The idea of 2 or 3 computers doing the binary equivalent of "Dueling Banjoes" scares me.
 
It's best if you don't run two pumps at the same time - this is 70gph running through your fuel system. This is more than a -6 system should be flowing.

I assume you're concerned about suction feed to the pumps. A little over a gallon per minute shouldn't be a problem on the pressure side.

Anyone have a reference for line size allowables in aircraft design?
 
I assume you're concerned about suction feed to the pumps. A little over a gallon per minute shouldn't be a problem on the pressure side.

Anyone have a reference for line size allowables in aircraft design?

Here are the Aeropquip tech specs including Flow Capacities of Hose Assemblies. I do not have a reference for any standards, tolerances, excess capacity, etc.... Perhaps there is something in the FAR's.

http://aeroquipperformance.com/rt-2899-technical-specs.html
 
Back
Top