1. Registration trouble? Please use the "Contact Us" link at the bottom right corner of the page and your issue will be resolved.
    Dismiss Notice

Dual Tank Systems

Discussion in 'Early CJ5 and CJ6 Tech' started by Steve's 70-5, Jan 4, 2021.

  1. Jan 6, 2021
    45es

    45es Active Member 2024 Sponsor 2023 Sponsor 2022 Sponsor

    Naches, WA
    Joined:
    Mar 31, 2007
    Messages:
    1,103
    With this comment, I would consider Fireball's recommendation.
    But since you made this comment, I would just use the original mechanical pump. It is capable of pulling fuel through a tank selector valve and suppling the Force Fuel system.
     
  2. Jan 8, 2021
    BadGoat

    BadGoat How High Can You Climb?

    Northern Virginia
    Joined:
    Apr 27, 2017
    Messages:
    326
    If you are set on using the Force Fuel system, then I'd suggest using a mechanical low pressure pump on the engine to feed it and running a transfer switch to select the tank you want to use. This keeps it to a simple system for trouble shooting.

    If you are on the fence about the Force System, I'd suggest a high pressure in-tank pump in the in-frame tank instead as the main and a transfer pump from the aux tank to the main.

    I ran 3/8" steel fuel line along the inside of my frame rail from the rear pressure regulator to the firewall, then high pressure flex line to connect the ends.

    Either way I'd suggest using steel line where feasible and modern high pressure flex line everywhere else along with 100 micron fuel filters before it gets to the TBI. It doesn't cost any more and fuel line degradation is a major cause of issues with EFI conversions. The modern fuels clean the gunk out of the lines and old tanks and clog up the injectors.


    Mike
     
  3. Jan 8, 2021
    timgr

    timgr We stand on the shoulders of giants. 2022 Sponsor

    Medford Mass USA
    Joined:
    Aug 10, 2003
    Messages:
    23,596
    Question - with a transfer pump from the aux to the main, how do you keep from filling the main all the way to the gas cap? Modern unvented tanks will hold enough vapor pressure to force the vapor out through the charcoal canister, but they don't seal well enough to keep liquid from coming out around the gas cap. Also, you need some air space above the gasoline for the vapor recovery to work. Gasoline coming out the bottom of the charcoal canister seems like a real possibility.
     
  4. Jan 8, 2021
    BadGoat

    BadGoat How High Can You Climb?

    Northern Virginia
    Joined:
    Apr 27, 2017
    Messages:
    326
    For me the aux is smaller then the main so as long as the main is empty, it's not an issue. I have a 15 gallon in frame tank and the under seat 10 gallon tank is the aux. I don't have it all plumbed up yet, but this is one of the projects for this year.

    You also need to develop a fill up routine to keep the aux tank fuel from getting stale. Something simple like whenever the main tank is empty, fill from the aux tank, then top off both tanks from the fuel station.


    Mike
     
  5. Jan 8, 2021
    timgr

    timgr We stand on the shoulders of giants. 2022 Sponsor

    Medford Mass USA
    Joined:
    Aug 10, 2003
    Messages:
    23,596
    So you wait until the main is empty, then manually transfer the fuel.

    When I had my '73 back in '73, I had one of those brass selector valves on the floor in front of the driver's seat. I would run until one tank was empty, and switch to the other.

    You can build an equivalent to the Force Fuel if you want. Use the mechanical pump to switch between tanks, and put a surge tank under the hood. Then take the feed from the surge tank to feed an electric high-pressure pump that goes directly to the EFI. Not a new idea - lots of examples of this setup online. This fits pretty easily with the existing plumbing for a carburetor, and requires the same number of pumps as any of these other schemes.

    [​IMG]
     
    Last edited: Jan 8, 2021
    BadGoat and dozerjim like this.
  6. Jan 8, 2021
    scott milliner

    scott milliner Master Fabricator

    Seattle Wa.
    Joined:
    Dec 17, 2002
    Messages:
    2,361
    Finding a way to vent the tanks will also be an issue. It took me a long time to figure out how to vent my under seat tank.
    I found my under seat tank is only vented inward, not out. With my pump in the tank and a return line. My tank wanted to
    build up pressure to the point of splitting the seams. And I did split one. I ended up putting in a roll over vent valve in the tank nozzle and running
    it through a Carbon Canister.
     
  7. Jan 8, 2021
    Jw60

    Jw60 Sitting up n buckled down. 2024 Sponsor 2023 Sponsor 2022 Sponsor

    Sedalia MO.
    Joined:
    Apr 8, 2008
    Messages:
    4,716
    I'm concerned about issues with the returns being shared in this layout. There is higher risk of back pressure on the regulator especially at idle.
     
  8. Jan 8, 2021
    Fireball

    Fireball Well-Known Member 2024 Sponsor 2023 Sponsor 2022 Sponsor

    Pullman, WA
    Joined:
    Feb 9, 2018
    Messages:
    5,622
    The FiTech document for the Force Fuel reservoir shows a returnless connection from the reservoir to the throttle body, so that's not an issue in this case.
     
    Jw60 likes this.
  9. Jan 9, 2021
    BadGoat

    BadGoat How High Can You Climb?

    Northern Virginia
    Joined:
    Apr 27, 2017
    Messages:
    326
    A secondary issue is if you are feeding the engine from both tanks then you need a way to monitor the level in both tanks. You can do this with a switch on a single gauge as long as both of your sending units work in the same ohm range. If they use different ohm ranges then you will likely need two fuel gauges. If you are using a main and an aux tank all you really care about is if the aux tank is full or not, so a single fuel gauge showing the level in the main tank is sufficient. Of course all this assumes you can keep one fuel gauge working in a CJ5, let alone two of them.

    Mike
     
  10. Jan 9, 2021
    timgr

    timgr We stand on the shoulders of giants. 2022 Sponsor

    Medford Mass USA
    Joined:
    Aug 10, 2003
    Messages:
    23,596
    First I wonder why you worry about back pressure on the regulator. Is there some evidence that return back pressure is ever a problem with any of these systems?

    Intuitively I expect back pressure is not a concern, as long as the return path to the tank is not obstructed. The return path pressure can never rise above the pressure differential needed to return the full output from the low pressure pump. As an aside, if you use the mechanical pump as the low pressure pump, the output will significantly decrease at idle. The addition of the high pressure pump can't affect the return pressure, since its output change is always limited by the peak output of the low pressure pump, and any sudden demand changes drain or fill the accumulator rather than create a lot of back pressure on the regulator. The demand change can't be more than what fuel the engine consumes. Pressure in the accumulator can never rise above what's needed to drain the full output of the low pressure pump.

    I suspect you could build a system like this that would have such issues, but I think you'd have to design something dumb to make it happen. Maybe you could put a restriction in the return path, or use a hugely oversize electric low or high pressure pump. Gasoline has viscosity of about 60% of water, and Poiseuille's law says the flow rate goes like the inverse of viscosity, ie all other factors equal, 70% more flow than water. With sensible construction, I don't see how back pressure could be an issue.

    Been a long time since I studied fluid mechanics, but I recall usually you want to determine flow from differential pressure, not the reverse. Typically a flow sensor uses an orifice and measures the pressure change across it to determine flow. If there is no orifice, the differential pressure is so small that it may be impossible to measure. You should be able to calculate peak laminar flow in the return pipe and determine a pressure difference across the length of pipe/tubing. My guess is the pressure change needed for whatever volume your low pressure pump can deliver will be tiny.

    Let's go on -

    Poiseuille's law says that the flow rate Q = (deltaP r^4)/(8 v l) where deltaP is the pressure change, r is the radius of the pipe, v is the viscosity and l is the length of the pipe. This tells us a few things. First, the pipe radius has a huge effect on pressure. For example, changing from 1/4" return to 3/8" reduces the pressure drop by five times (deltaP' = 0.2 deltaP). Also, you should be able to calculate deltaP for the system. A wise colleague of mine always said "never calculate something you can measure." But here, I think this calculation would be useful, assuming the pressure regulator were sufficiently characterized so you could recognize a problem. If it's not, that likely means that it's not a parameter that's routinely released to the public, or possibly, it's such an irrelevant quantity that nobody pays attention to it.

    Sorry if this runs on - I thought about this question for a while, and mostly wrote this out as train-of-thought. It could have been much shorter and said most of the same.
     
    Last edited: Jan 9, 2021
New Posts