Another fuel tank plumbing option

[nkemp]

In terms of using plastic exposed to gas... gas cans are made of high-density polyethylene (HDPE)...#2 recycle code. On chemical resistance chart HDPE did not get the top rating but it seems to be working for gas can mfr's.

[nkemp]

Here is an updated option. To simplify, a plug is secured inside the boot (which is cut like in inverted "J") and is held with a hose clamp. So now the boot hole is plugged and the seal around the tank top is as is today using the boot. Thus a sealed tank without drilling a bracket. A rain cover of some sort is needed to prevent water pooling in the boot (but if it does there are no parts to damage).

Buildup of vertical heights may be an issue.

[DMCVegas]

Just to kind of chime in here, I too have always thought about a solution to this issue as well. I've had an idea for some time now about how to do this as simply, and cheaply as possible. Here goes.

The retrofit would only be of 3 off the shelf parts, and 3 custom pieces that can easily and cheaply be molded out of plastic. These pieces would replace the Fuel Pump Boot, the Fuel Pump Cover Seal, and the Fuel Pump Support Ring. That alone is over $100 worth of parts, and I'm sure we can go cheaper.

The first piece is a new Support Bracket. It will hold both the fuel pump, as well as the metal Return line. What I'm picturing is a cylinder that has an opening on one side, that allow the Return Line & Fuel Pump to be put snugly inside of it, as well as a molded support on the bottom to keep the pump exactly at the correct height from now on. Finish securing both items with a Hose Clamp or two. The top of this Bracket has a 90° flange that runs all around the edge so that you can mount/suspend it to/from the tank.

The Second piece is a ring that goes around the existing opening in the tank where the fuel pump goes. Except that it's threaded on the outside. This piece would have to be welded to the tank somehow. Most likely with epoxy which shouldn't be a problem since they do make gasoline resistant glues.

The Third piece would be a locking nut that would attach to the now-threaded opening to hold everything in place.

The 3 other off the shelf items would be 1 Gasket to seal the assembly to the tank, 2 rubber boots to cover the submerged terminals on the fuel pump. If on the Bracket we molded in a small boxed area on the top with 4 sides, and then had two holes at the bottom where the fuel pump wires passed through, we could then just shoot a small gob of Instant Gasket in to the niche to seal it up.

Now I honestly think that this is a viable way to retrofit the existing assemblies that we currently use. And no, having wires running into the tank does not pose a fire hazard. The fuel pumps we use now internally expose gasoline to electrical leads. Fuel acts as a coolant. But to have fire you need 3 things: Heat, Fuel, and Oxygen. And there is no oxygen present in liquid fuel nor fuel vapors. Volvo has been doing this for years with the wires in the tanks and never had a problem. So this is quite doable.

However, as much as I beleive that this is viable, I honestly don't know if it, or ANY other method of retrofitting the Fuel Pump assembly for that matter is actually *practical*. Two big questions:

1. Is this a cheaper/easier/more efficient way to accomplish what we already have?
2. Would people actually buy this? Would they be too scared to modify their cars? Do they not WANT to modify their cars for the sake of originality?

Here is an updated option. To simplify, a plug is secured inside the boot (which is cut like in inverted "J") and is held with a hose clamp. So now the boot hole is plugged and the seal around the tank top is as is today using the boot. Thus a sealed tank without drilling a bracket. A rain cover of some sort is needed to prevent water pooling in the boot (but if it does there are no parts to damage).

Buildup of vertical heights may be an issue.

The problem with that design is there there is no way for the fuel on the outside of your Sump to refill it once the outer fuel level passes below that line. That's why you have to use a Baffle. You could keep it that high to try and maintain a constant submergence with the fuel from the return line, but I don't think it needs to be quite THAT high. Mostly though with the smaller Baffle we currently use, it's supposed to keep only the intake filter for the pump submerged, which is all you really need.

[nkemp]

The problem with that design is there there is no way for the fuel on the outside of your Sump to refill it once the outer fuel level passes below that line. That's why you have to use a Baffle. You could keep it that high to try and maintain a constant submergence with the fuel from the return line, but I don't think it needs to be quite THAT high. Mostly though with the smaller Baffle we currently use, it's supposed to keep only the intake filter for the pump submerged, which is all you really need.

The GM sumps (or equivalent) allow fuel to flow into the sump from the bottom and has backflow valves to prevent the fuel from getting out. With fuel from the return line the sump will be a higher level than the fuel tank (until it reaches the sump overflow). The problem with the GM units is that the pumps appear to not pump at the pressure and volume required of the Bosch CIS.

Some of the GM units pump from the sump bottom, some from the tank bottom. The latter can use more of the fuel in the tank.

You are correct relative to power lines in the fuel tank. If you are driving a car made in the last 10 or 15 years there is a good chance that the power lines are in the fuel. A solid connection eliminates the chance for a spark, gasoline is an excellent insulator and there is insufficient oxygen for combustion. The only real reason for the connector boot is in the event that something conductive got in the tank (like foil) and it got across the pump terminals, then there would be a spark opportunity. FWIW, some (maybe all) modern fuel pumps pump the fuel through the pump to cool it and in doing so expose the electrical to the fuel.

At this point, it appears that using the existing pump (they are cost effective and perform well) coupled with the GM sump and plumbing the lines through the tank provides a superior fuel pump solution while providing a cost effective solution. But to get the sump you'll need to do some junk yard work ... but that is kinda fun and entertaining.

To optimize this design we need the mfr of the existing boot to make a "flat top boot" or a domed one with no holes.

Nick

The Second piece is a ring that goes around the existing opening in the tank where the fuel pump goes. Except that it's threaded on the outside. This piece would have to be welded to the tank somehow. Most likely with epoxy which shouldn't be a problem since they do make gasoline resistant glues.

The Third piece would be a locking nut that would attach to the now-threaded opening to hold everything in place.

I've pondered adding threads some how and a piece attached to the tank may work. My concern is that one with sufficient thread and backing material would no longer fit the threaded locking ring. Another option is to insert "nubs" into the tank plastic in a thread formation that the locking ring could thread onto. But anytime you put a hole in the tank you need to think about how to seal it. So one solution causes another problem and you keep pushing the problem down the sequence.

I have not checked yet but I'd guess that the current locking rings are pretty close to our tank opening's OD.

If a GM unit (or equiv) could be found with sufficient flow & pressure, then cutting off the current boot leaving sufficient overlap to the GM unit top may be sufficient to hold it in the tank (Adding a bit of sealant (no rtv) would help).

I've poked at my failed Astro Gm unit for a couple years and have not come up with anything acceptable.

FWIW, true GM units are relatively expensive and look-a-likes don't necessarily perform as well or as long.

Nick

[Bitsyncmaster]

With fuel from the return line the sump will be a higher level than the fuel tank (until it reaches the sump overflow).Nick

Explain to me how the return line will fill a sump?

The pump pulls fuel out of the sump. The engine uses some of that fuel. The excess is returned via the return line. So wouldn't the sump get drained of fuel?

[DMCMW Dave]

Explain to me how the return line will fill a sump?

The pump pulls fuel out of the sump. The engine uses some of that fuel. The excess is returned via the return line. So wouldn't the sump get drained of fuel?

Yes, yes it would.

The GM pump has a hole in the bottom with a check valve to let fuel come in but not run out. So the level in the sump will be the same as the level in the tank. Maybe higher (due to fuel slosh) but not lower.

[nkemp]

Explain to me how the return line will fill a sump?

The pump pulls fuel out of the sump. The engine uses some of that fuel. The excess is returned via the return line. So wouldn't the sump get drained of fuel?

Eventually yes. But the combination of the fuel flowing in via the valves on the bottom and the return fuel keeps the level higher than the tank level (once it gets below the sump overflow). Also, depending on the pump arrangement, it appears that some pumps pull from the sump and others pull from the tank bottom and the sump. I would prefer the latter for a D and wouldn't care if the tank had a "pump well".

Nick

... but a remind everyone that "I'm subject to being wrong without notice

[Bitsyncmaster]

Eventually yes. But the combination of the fuel flowing in via the valves on the bottom and the return fuel keeps the level higher than the tank level (once it gets below the sump overflow). Also, depending on the pump arrangement, it appears that some pumps pull from the sump and others pull from the tank bottom and the sump. I would prefer the latter for a D and wouldn't care if the tank had a "pump well".

Nick

... but a remind everyone that "I'm subject to being wrong without notice

It will never fill above the tank level (not counting sloshing of the gas). Physically impossible since less gas is returned than the gas removed.

[nkemp]

I would argue (respectfully... of course) that the combination of fuel inflow via the valves (simple rubber diaphragms) and the pump return allows the sump level to be above the tank level.

Or think of it this way, the level in the sump is consistent with the level outside the sump because of the valves on the bottom of the sump. Now add the return flow and the level rises.

Note that in most operation relative to the D, this is not particularly important until the tank level gets low or when the fuel moves to another part of the tank such as taking a curve or going up/down a steep incline. In those cases having the sump level higher than the tank level helps prevent fuel starvation.

The original D pickup design was supposedly a sump design (not a baffle as often discussed), albeit it a shallow sump. At least that is how a previous employee at DMCH explained it to me. Supposedly the original design would pump a tank substantially drier than a baffle design can accomplish.

Nick

[Bitsyncmaster]

I would argue (respectfully... of course) that the combination of fuel inflow via the valves (simple rubber diaphragms) and the pump return allows the sump level to be above the tank level.

Or think of it this way, the level in the sump is consistent with the level outside the sump because of the valves on the bottom of the sump. Now add the return flow and the level rises.

Note that in most operation relative to the D, this is not particularly important until the tank level gets low or when the fuel moves to another part of the tank such as taking a curve or going up/down a steep incline. In those cases having the sump level higher than the tank level helps prevent fuel starvation.

The original D pickup design was supposedly a sump design (not a baffle as often discussed), albeit it a shallow sump. At least that is how a previous employee at DMCH explained it to me. Supposedly the original design would pump a tank substantially drier than a baffle design can accomplish.

Nick

I see how your thinking. The flow into baffle is more. Maybe your right.