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

Fuel System Manager
Goto page Previous 1, 2, 3 Next

Matronics Email Lists Forum Index -> AeroElectric-List

View previous topic :: View next topic

Author Message

Eric Page

Joined: 15 Feb 2017
Posts: 269

Posted: Sat Feb 21, 2026 7:49 am Post subject: Re: Fuel System Manager

Here's a quick conceptual doodle showing how the switch wiring might be done to handle OFF-MANUAL-AUTO switching, using a 2TL1-1 (or, for a much cheaper alternative, the B&C S700-2-1).



Switched LPFS ground connection to be shared with HPFS using diode steering.



AFC logic remains the hurdle...

Switch_Concept.jpg

Description:

Filesize:

101.51 KB

Viewed:

678 Time(s)

911pete

Joined: 09 Aug 2018
Posts: 18

Posted: Sat Feb 21, 2026 8:33 am Post subject: Re: Fuel System Manager

Not being an electronics guy my question is would latching relays utilizing the fuel sensors be a path to control the logic?

user9253

Joined: 28 Mar 2008
Posts: 1961
Location: Riley TWP Michigan

Posted: Sat Feb 21, 2026 8:40 am Post subject: Re: Fuel System Manager

Pete, you wrote in a previous post, "The goal is to always keep the main tank 

full until the wings are empty".   That seems reasonable.  It is hard for me to 

understand how the fuel computer knows when to open and close the wing tank

valves when there are only two level sensors in the main tank, high and low.  It 

would not be desirable to let the main tank fuel level go down to the low level 

sensor before opening the wing tank valves.

  Suppose the computer closes the wing tank valves when the main tank is full.  

As fuel is used, the high level sensor tells the computer that the main tank is no 

longer full.  The computer could then wait for a period of time, say 5 minutes,  

before opening the wing tank valves again.  That time delay period needs to be 

short enough to keep the main tank relatively full.  The time delay also needs to 

be long enough so as not to wear out the solenoid valves due to short cycling.

_________________
Joe Gores

911pete

Joined: 09 Aug 2018
Posts: 18

Posted: Sat Feb 21, 2026 9:18 am Post subject: Re: Fuel System Manager

I think you are mistaking where the low level sensor is. The main tank is approximately 75 gallons. The low level sensor is set at around the 60 gallon level. The plane burns 1.5 to 2 gallons per minute at non takeoff settings, so the valves would typically cycle every 7 to 10 minutes. If the fuel gets below the low level sensor for longer than 20 seconds the low level light would come on giving you around 30 minutes to land.

user9253

Joined: 28 Mar 2008
Posts: 1961
Location: Riley TWP Michigan

Posted: Sat Feb 21, 2026 11:28 am Post subject: Re: Fuel System Manager

OK, I modified my drawing in post number 5 by moving the lower fuel level 

sensor from the bottom of the tank to a higher position.

_________________
Joe Gores

Eric Page

Joined: 15 Feb 2017
Posts: 269

Posted: Sun Feb 22, 2026 6:37 am Post subject: Re: Fuel System Manager

Pete, do you have a datasheet link for your LPFS and HPFS?  I'd like to see their electrical requirements and what position feedback, if any, they offer.



Yes, latching relays could definitely be part of the solution, but they can't be driven directly by the fuel level sensors.  This type of relay requires a brief pulse (a few tens of milliseconds) to set or reset its coil(s) (some have two), not a continuous current like a normal relay.  This would require some electronics to turn the long-duration on/off cycles of the level sensors into these short pulses.



Also, we want the absence of fuel at the Low Level Sensor to open the solenoids, but we don't want them to close when that sensor sees fuel again; they have to wait until the High Level Sensor also sees fuel.  That's going to require some electronics too.

user9253

Joined: 28 Mar 2008
Posts: 1961
Location: Riley TWP Michigan

Posted: Sun Feb 22, 2026 9:07 am Post subject: Re: Fuel System Manager

Eric,

What do you think of the attached circuit?  It uses discrete components.  No computer necessary.  Integrate it with your switch circuit.

Valve Control.png

Description:

Filesize:

6 KB

Viewed:

559 Time(s)

_________________
Joe Gores

user9253

Joined: 28 Mar 2008
Posts: 1961
Location: Riley TWP Michigan

Posted: Sun Feb 22, 2026 9:29 am Post subject: Re: Fuel System Manager

After thinking about it, why use an AND Gate?

The upper liquid level switch could energize the relay.

The relay can hold through the lower liquid level switch.

_________________
Joe Gores

Last edited by user9253 on Sun Feb 22, 2026 10:04 am; edited 1 time in total

user9253

Joined: 28 Mar 2008
Posts: 1961
Location: Riley TWP Michigan

Posted: Sun Feb 22, 2026 9:48 am Post subject: Re: Fuel System Manager

Simple relay circuit attached.

Valve Control_.png

Description:

Filesize:

5.55 KB

Viewed:

543 Time(s)

_________________
Joe Gores

911pete

Joined: 09 Aug 2018
Posts: 18

Posted: Sun Feb 22, 2026 5:07 pm Post subject: Re: Fuel System Manager

Eric



Here is a copy of the specifications for what I will use for the low pressure solenoid. The original spec’d solenoid is not available. The high pressure one is a Duke’s aerospace part, they are the original supplier of the low pressure one as well, but they are no longer in business as well. All the solenoids just have 2 wires so I assume they are simple on open or closed.



As far as the logic I started investigating Arduino controllers and that might be the path of least resistance for this whole project.



Pete

IMG_1335.png

Description:

Filesize:

390.17 KB

Viewed:

496 Time(s)

Eric Page

Joined: 15 Feb 2017
Posts: 269

Posted: Mon Feb 23, 2026 6:30 pm Post subject: Re: Fuel System Manager

Joe:  Yep, I was over-thinking it.  Well done.  Your second diagram does exactly what's needed with nothing but one relay per side.  You have to wonder why they used a PLC in the first place.



Pete:  An Arduino isn't necessary.  In fact, the original PLC wasn't necessary.  The diagram in Joe's last post does exactly what you want with just a relay.



I've started work on fleshing out Joe's design to a full schematic with specific part numbers.  I have some other stuff going on so it may take a few days...

911pete

Joined: 09 Aug 2018
Posts: 18

Posted: Mon Feb 23, 2026 6:58 pm Post subject: Re: Fuel System Manager

Great news. No rush.

user9253

Joined: 28 Mar 2008
Posts: 1961
Location: Riley TWP Michigan

Posted: Tue Feb 24, 2026 10:10 am Post subject: Re: Fuel System Manager

Below is a schematic of the Viper Jet mail fuel tank level control.

It was confusing to draw because the valves are normally open.

Power is applied to close the valves.

The low fuel bingo light feature can be added with a delay-off relay.

Suggestions welcome.

Valve Relay and SW.png

Description:

Filesize:

15.88 KB

Viewed:

305 Time(s)

_________________
Joe Gores

user9253

Joined: 28 Mar 2008
Posts: 1961
Location: Riley TWP Michigan

Posted: Wed Feb 25, 2026 6:08 am Post subject: Re: Fuel System Manager

The OP wants a warning light to come on if the main fuel tank low level sensor

has not reset after 20 seconds.  Amazon has a timer relay that should do it.

It has both high and low input triggers.

Search Amazon for "drok timer relay".  Their OVERALL PICK costs $11.99.

Reading the reviews, it seems the instructions are written in Chinglish.

But there are Youtube movies with instructions.  Search Youtube for DROK XY-LJ02.

_________________
Joe Gores

wsimpso1

Joined: 04 Nov 2018
Posts: 36
Location: Saline MI

Posted: Thu Feb 26, 2026 7:10 am Post subject: Re: Fuel System Manager

Sorry I did not see this sooner.



No lack of ambition in this project. Operational risk in this bird is higher than I am comfortable with, but the OP and I are different folks...



Before we get into details, I have run through many schemes for a similar system (and only running fuel for a 200 hp engine) so I appreciate the issues you face. We had quite a conversation on the topic here:

https://homebuiltairplanes.com/threads/transfer-pumps-and-fuel-system-configuration.30820/ 

There were other discussions of what makes sense and how to manage failures that will happen.



Failure Modes and Effects Analysis is a good idea for review of each design option. Perhaps it is in the background and we have just not seen it here. I emphasize that every failure mode of every element of the system should be tracked down, then consequences (effects) at system mode must be chased through. To really do it right, you will need severity of consequences and some idea of the frequency of each failure mode. This gives you a look at the relative riskiness of each set of options, and usually will drive the designer towards managing failure modes. 



Now to details shown in the thread. 



The shown systems are all complicated, and complicated makes for more failure modes all competing to see which fails first. A drive for less failures will drive simpler systems.



That being said, I love jet pumps. As long as your engine pump runs, you have transfer pump power. You must then have plumbing runs to each tank and those runs do have their failure modes and frequencies too...



Solenoid valves take power. If they take power to shut off, that means loss of power to the valve OR stuck open will immediately attempt to transfer all fuel in that wing to the main tank, likely with much of it going out the main vent. If you lose ship's power entirely, both wing tanks will dump to the main, with most of it likely going out the vent. That may quickly leave you with just the main tank, so a runway had better be well within reach.



Your solenoid valves are 3000 psi lines with 1/2" pipe connections, but the jet pumps will provide far far lower pressures to shove fuel to the main. Your flows for this thirsty engine may be inadequate with 1/2" plumbing and a jet pump. Total kinetic energy of the output is unavoidably less than the kinetic energy of the jet. Know what your jet pump output pressure vs flow curve look like, then figure out how much head loss you will have between wing and main tank to make sure you can flow roughly 2x engine burn rate to the main tank. 



I suggest that having more than one power source for running all this stuff is wise as you are becoming at least partial electrically dependent just to run the engine. I also suggest that you may want to have more than one independent way to power your fuel selectors. 



One of the fundamental teachings is that removing a failure mode completely is often way better than counter measures. 



Another fundamental teaching is the huge value in selective redundancy. More than one way to do something is really smart if you NEED that something done...



Perhaps a manual valve would be way smarter?



In my system, I considered jet pumps and all sorts of plumbing options. Instead I have two simple check valve equipped transfer pumps in parallel and a duplex manual valve to select tanks. The main tank vents to the source tank through the manual valve for no overboard fuel flow and no need for switching pumps on and off automatically. At max cruise, my main tank is over 100 miles of fuel, at best range I have more like 160 miles to get on the ground. That may not be good for a trip to Hawaii, but is way comfortable in the lower 48. 



I do have high and low switches in my main tank. They only run alarms, and I selected solid state industrial types - no float switches. When fuel drops below the high limit, I am warned to switch pumps OR switch tanks OR both. When the low level level light comes on, I am warned that I am about to land on something close by. 



By the time I got done doing my FMEA (several designs, and then updates meant more passes) I found that good switchology with independent wiring/powering and redundant elements was the way to keeping powered things powered. So, I have two always hot buses (each circuit fused) and two loads and two double pole double throw switches for each major function. S700-2-10 from B&C switches. Where I need power to get a function, I have four different ways to power each needed function. 



Since you need the no-power function to also be available, you would be wise to noodle on that. Perhaps a pullable breaker for each valve? 



All of this avoids logic circuits, diodes, AND/NAND gates, and Arduinos. The only extra wiring and connectors are the redundant circuits. 



Yeah, this long experienced ME is leaning away from complicated electricals and towards Bob Nuckolls' fundamentals of making it require very little in the way of intervention on any single failure while aloft. Fix it on the ground. 



Billski

user9253

Joined: 28 Mar 2008
Posts: 1961
Location: Riley TWP Michigan

Posted: Thu Feb 26, 2026 9:09 am Post subject: Re: Fuel System Manager

Billski, Thanks for your input.

I like your idea of venting the main tank back to the wing tanks.

And I agree that Failure Modes and Effects Analysis is a good idea.

It is my understanding that the original designer of the Viper Jet also 

designed the fuel system, not the current builder, Pete.  Originally the main

fuel tank level was controlled by a computer.  Since that computer is no 

longer available, Pete asked for my assistance.  I substituted a relay for the 

computer.  If that relay fails, no big deal, the main tank level can be 

controlled manually with the toggle switches.  As for float switches, there are 

none.  I drew float switches on my schematic to make it easy to understand.  

But the level switches are actually industrial solid state switches.  Pete 

posted the switch manufacturer's description in a previous post.

  I agree that the operational risk in this bird is higher than I am comfortable 

with.  However, Pete flies Jets for a living.

  Each of the 3 fuel tanks also has independent capacitor type fuel level 

sensors.  Assuming that those sensors are connected to an EFIS, the EFIS 

can be configured to alarm whenever a fuel tank level is either too high or too low.

  If something goes wrong with the fuel transfer system, the main tank holds 

more than 30 minutes of fuel.  Hopefully that is enough to fly to an airport 

with a long runway.

_________________
Joe Gores

911pete

Joined: 09 Aug 2018
Posts: 18

Posted: Thu Feb 26, 2026 11:07 am Post subject: Re: Fuel System Manager

Billski 



As Joe said I’m a builder, not a designer. Even though there aren’t many of these planes flying there are some and the fuel system has been proven to work with respect to the ability of the jet pump having enough capacity to keep the main tank full



I read through your thread on HBA.  As someone else noted in your thread this isn’t really a fuel system but a transfer system meant to keep the main tank full. In the event of total electrical failure all three valves open and supply the main tank. Yes there will be fuel venting overboard.  If it happens with anything in the wing tanks you will get a full main tank plus some of the wing tank fuel until they are dry. Should give you no less than 30 minutes until fuel exhaustion.  Thats 30 minutes at 300+ knots. 



You mention the 3000psi solenoid valve. The original design used a 150 psi max valve from Dukes aerospace. They are no longer in business and I’m having a hard time trying to contact the follow on company. The valve I’m contemplating is good up to 3000psi, but there will be nowhere near that in actual use



As far as pullable breakers to remove power I think the panel mounted switches do that



As Joe and Eric have designed this it seems very simple, one relay and two switches. The level sensors also have no moving parts



Pete

Ceengland

Joined: 11 Oct 2020
Posts: 397
Location: MS

Posted: Thu Feb 26, 2026 12:17 pm Post subject: Re: Fuel System Manager

Hi,

Late to the party (gmail has been sending Matronics stuff to spam), but one or two thoughts.



The plumbing diagram shows L & R plumbed in parallel to the jet pump. Do be aware that many 2-tank systems, if (when) fuel feeds unevenly and one tank empties, will start to suck air instead of fuel. If there isn't significant source pressure from the remaining fuel, the pump will find it easier to pump air than fuel. This usually isn't an issue with high wing a/c and a low-mounted join/pump, but is pretty common with low wing a/c. Many low wing a/c prohibit operating with the tanks in 'both' for this reason. Might not be an issue with that particular airframe, but I'd want to know that it's been proven by emptying a tank in flight (and the other tank, on a different flight) to verify, before committing to the design. Just because it's flown doesn't mean that the empty tank issue has been thoroughly tested.



The main tank vent issue has already been mentioned. There are certified a/c flying with systems like the one described, and some of those systems have caused crashes when the loss of fuel wasn't detected. There are venting plans that can avoid pumping fuel overboard (and, for that matter, possibly eliminate most of the switchology).



Bilsky already mentioned FMEA. A stuck solenoid valve, or a stuck switch (unless there's access to that switch's power source) could cause loss of all fuel in the affected tank.



Just a couple of FWIW thoughts...

_________________
Charlie

Eric Page

Joined: 15 Feb 2017
Posts: 269

Posted: Thu Feb 26, 2026 4:45 pm Post subject: Re: Fuel System Manager

I haven't gotten back to working on this but I wanted to check progress here.  It looks like Joe has the basic switching schematic pretty much licked.  Except...



As drawn, the HPFS is only under manual control by the switches, and the relay can't control it in AUTO mode.  The HPFS will be open any time either switch is in any position other than OFF, even if both LPFS are closed by AUTO mode.  Maybe that's not a problem.  Pete, what say you?



A relay with 3 or more normally open poles may be able to control the HPFS in parallel with each LPFS, but I haven't thought through the switchology...



I have the bones of a low fuel light delay and flasher circuit, so I'll press ahead with that part.  Pete, do you want an output to trigger an EFIS alarm or do your other fuel level senders have that covered?  If so, does your planned EFIS system accept 24V inputs for alarm triggers, or would 12V be preferred?

911pete

Joined: 09 Aug 2018
Posts: 18

Posted: Fri Feb 27, 2026 6:02 am Post subject: Re: Fuel System Manager

Eric



The HPFS can stay open in Auto. It just needs to be closed in the off position.  That is what is used during engine start.



I’m using the Garmin G3X with a GAD27.  The documentation says it’s for 12 volt systems, but the garmin engineers I spoke with say it will handle 24 volts. I can also use the GEA24 which is part of the system as well for the warning input. It is rated for 24 volts. Either of these units can be configured for the warning to be active high or low. 



Pete

Display posts from previous:

	Matronics Email Lists Forum Index -> AeroElectric-List	All times are GMT - 8 Hours Goto page Previous 1, 2, 3 Next
Page 2 of 3

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