Thread: Fuel Pump, Long Range Tank install questions

The TLDR;
keep the in-tank setup, clean it out with good fuel a couple times. Use the fuel as a cleaning agent for filthy parts etc. dispose of thoughtfully.

I edited my reponse, because it got a bit tangential.

why?
Cost.

If you don't care and you have about 1500 bucks to spare, then yes, external pumps, surge tank, filters (pre/post) and lots of JIS/AN fititngs and appropriate size lines and clamps and pressure regulator etc.... goes on and on.

Not sure it is worth it, unless its a forever car.

the long winded diatribe: apologies in advance.

If you are going to run an external pump like an 044, then you will need to fabricate a surge tank and larger lines.
The #1 reason bosch inline pumps fail is because they have been run dry or without a proper filter. An 044 needs a big supply line. I don't know if you are aware of that or not.
The rrc tank pickup connections are also less-than-ideal for a pickup point setup without an internal swirl pot, making the requirement for a surge tank more of an obligation, than an option.

an 044 is a bit of overkill as well. Especially given the rrc's supply and return lines are so arbitrarily small.

If you want to run external pump, May I also suggest you observe the need for a good filter BEFORE the pump(s). That is, if you want them to last at all. Especially with a 4x4.

Speaking from experience, when you have to fork out for replacement bosch pumps, finding actual genuine bosch pumps is harder than you may think - and the sticker shock is nothing short of horriffic. And when you have for example...as I have - multiple vehicles of the same marque, and they each run twin inline pumps.... then the cost of replacing the pumps and filters runs to about 900bucks per car. Not including the flexible lines from tank to pump and return line etc - i.e. the stuff that you replace when the pumps fail because they are as old as the pumps...

This is why the standard rrc in-tank pump is the cost-effective and reliable solution.
Also... weigh up what is worse. cost of the external pumps, brackets, filter, hoses, AN fittings, adaptors, electrical upgrade to handle the current for the bosch 044 unit, Surge tank fabrication and fitting....

There are some very good reasons to stay with the in-tank pump. It's also not the end of the world if you have to drop the fuel tank to replace a pump - but given that they last a couple of decades, I am struggling to work out why you really want to put yourself through this - unless....

Unless you want to swap out that rover v8 for an LSA. Then the 044 makes sense to me.

BMW, Porsche & Mercedes-Benz have excellent fuel pump designs, well packaged and quite versatile, mostly external on the older vehicles. if you were close to a wrecking yard, you could go examine their setups for inspiration, or maybe adaptation. When they went in-tank, like the majority of japanese cars, the cost of the pumps went up, as did the servicing complexity.

The tank itself is really what dictates what type of pump you end up using, or you make the decision to modify. If the tank has a swirl pot internal, with a tank strainer and a bottom-exiting feed line, then inline external pumps are a doddle. All you need is the inline filter and pump / hose / mounts and correct electrical supply.
If the pump is internal, the pump itself likely has the swirl pot as part of it's design, and you need to replicate that externally (surge tank), which complicates things. It IS mandatory, because running an inline bosch pump without any fuel for as little as 15 seconds will literally destroy the roller bearings (which run in and use the fuel as a lubricant/coolant) and the pump will be f-ed in no time flat. Pressure regulator and return will need to be appropriately sized for the volume of fuel that the pump is capable of delivering. Any restriction to flow will create unnecessary pressure, which reduces the lifespan of the pump.

The pumps have a working pressure range, and the fuel rail pressure needs to sit literally in the middle of this range, with the regulator being able to pass sufficient volume into the return line without being oversized or undersized.

for an 044 pump, delivery line to the rail is usually -8AN and return is -6AN. In other speak, 1/2" supply, 3/8" return, unrestricted. where many surge tanks run dual pumps (for redundancy as well as surge prevention) the intake to the pump can be -10 (5/8") in some cases.

There are many ways to skin a cat. I think in this case however, your understandable desire to mitigate the need for dropping the tank, or facilitating access through the floor for a once-in-a-lifetime-of-ownership job, is possibly a little too cost-prohibitive.

But worthwhile if done properly. Not much (protected) space back there, and if serious offroading is on the agenda, then literally everything just increased in difficulty fivefold, because you also need to factor in protection, accessibility, serviceability (filter or pumps or electrical) and all of that is pretty much eliminated by keeping the pump in the tank.

It's a bit of a conundrum. Yes, I would do external pumps myself, but find that since I replaced the internal one a few weeks after buying the vehicle, it's been perfectly adequate - and I haven't got a long range tank or sill tanks to complicate the deal either.

But since you were asking about filters and pumps.... a Mahle KL38 filter is about the right size for the job. Will also last a good long time as well. has metric fittings to match bosch pumps and one-way valves, all you will need to do is fabricate the lines - which absolutely should be a proper JIS or An dash type fititng.

Bosch and Mahle are metric, so keep that in mind.

Here is a twin pump setup from a K-jet Merc, which is 5.6L v8. pressure accumulator not applicable to your scenario, but pump and filter most definitely.
L1030004.jpgL1030005.jpgL1030006.jpg

MB's have swirl pots in the tank, with an internal fine mesh tank strainer and a bottom-exit fuel supply. This means that the bosch pumps always have fuel pressure (head) available to the pump inlet. In the case of a CIS injection vehicle, the accumulator (little unit with hardline exiting from left) also maintains circuit pressure at about 2/3 operating pressure when engine off, which means that the pumps are not loaded with pressure on one side when the engine is not running. Pressure is equalized, but maintained at an elevated level to prevent possibility of fuel boiling. also means priming is quick, so starting is quick.

Similar principle to older D-jetronic in BMW, MB, Volvo et al, but only Porsche & MB persisted with CIS, for it's superior atomizing, power and fuel economy. (higher pressure system).

These days rising rate fuel pressure regulators and all the artificial induction stuff going on makes this design somewhat archaic, but I can tell you that a current biturbo direct injected AMG engine is only a 15% improvement in economy for 40 years of technological advancement in engineering, then it's easy to understand that modern fuel injection systems efficiency come from the control of spark and fuel metering with additional load and speed parameters, throttle position sensors, O2 sensors, and the manifold pressure sensor as well as airflow and temperature.

It takes all that and a bag o' chips to make things improve, and it takes 40 years to do it.

If serviceability and reliability are what you seek, you could do worse than copy 45 year old german engineering. Twin pumps inline (series, not parallel), one pre-filter and one post-filter. Change your filters every 12 months or 20,000km and you will never need to worry about the lifespan of the bosch pump... They make one hell of a precursory whine and do so for quite a while before they fail - so you have plenty of warning, not to mention a fuse to protect from over-current.

I just think it's a lot of effort to go to to do it properly, that's all. plus the $$$$. If this is a keeper, a 'forever car' then absolutely do it, but do it right, spare no expense and never have to touch the system again, except for routine filter servicing)

PS. that KL38 mahle filter on most Mercedes of the era get's changed by owners (the tightarsed ones) NEVER during ownership. So there are plenty of vehicles out there running filters that are 30 years old and probably half as many again with filters that have only been changed once... and the cars are fine... They eventually do clog up, but sage advice, just as with the RRC's mahle filter, is to replace every 12 months, or 2 weeks before any big trip / holiday (time to allow for any servicing problems to manifest before you leave, rather than on the trip).

Filter not expensive. also conservatively can replace every 5-10 years depending on vehicle use (seems to be the norm from my experience with classic car owners - i.e. not as frequent as it should be, but minimal use overriding factor for extended fuel system service interval)

9/10 issues I encounter are result of stale fuel sitting for years, causing moisture and rust in tank, and the obligatory vehicle with abysmal service history. rust either clogs in-tank screen or becomes so fine that it circulates into the pumps and destroys them in no time flat. 20 minutes of checking draining tank and cleaning can prevent 1200 bucks in replacement parts and 500 bucks in labour, fixing what could have been prevented if the owner exercised their brain at the outset.

Which is still why I reckon the in tank RRC pump is still cheapest practical solution.

Mercguy. Thanks so much for the detailed response, it really clarifies a lot that I was trying to sort and understand before the tank change, glad I asked. So much more involved with potential issues than with a simple external lift pump replacement when I had the TDI. Yes its a keeper but not really interested in going down yet another rabbit hole to replace something which is working ok. With this potential hassle, I'd rather prepare an easy access through the rear fitout for potential future replacement.
That said, if I maintain the in-pump option which is now my inclination - should I be shoving back in a 30 year old pump (350k km, it was resting for a while) which works fine, or sourcing a new replacement which might last another many years - or not - depending on where its from?
Thanks. Simon

IIRC people were using a Commodore pump in tank with a couple of easy mods. Pretty sure there were posts on here.

DL

Update on this project.

- Editing this post to say, while retaining the original pump is sound advice, I decided to pursue the external pump setup, it is a keeper after all - also after looking at a friends unit that was done by Graeme Cooper many years ago. Though I couldn't see inside the tank, it used the original pickup location with an external pump and has been running like this for a couple of decades. And, I really didn't want to drop the tank in the event of a pump failure. From what I've read the standard access hole under the carpet is no good for a Long Ranger tank as its offset slightly. -

This fix finally became very necessary when the original tank started leaking way too much, with a fill parked on a camber, fuel poured freely from the top on the tank somewhere. Too may fumes, not safe and wasteful. During lockdown 8 (Geelong) seemed a good time to get it done.

I'd been collecting some components in advance - went with a Bosch 070 (delivers 3 bar), 30 micron pre-filter with cleanable/replaceable filter, hose and fittings, new seal for pump housing, 50mm fill hose, breather hose, filler tube grommet, etc. Got the pump and filter online from EFI solutions, a ricer performance company based out of QLD - reasonably priced but not ebay cheap - with the advantage you can talk to someone on the phone if needed.

Taking out the old tank requires bumper removal, if ARB at least, plus the shearing or cutting of some 30 year old rusty bolts. For hose access I also took off the mudflap, mudflap bracket to body.
Before I dropped the tank, I drained it with the help of a facet pump attached to some 8mm hose fed through the tank filler hole after the rubber hose was removed. it was on reserve (orange light) and still pumped out 25 litres into clean cans. With the tank removed, found that two of the four seals in the top had failed, (photo) as well as the 50mm filler hose which was basically porous.


The surroundings of course were filthy, with some light surface rust. All that was cleaned and vacuumed out, any of the surface rust treated and left overnight to dry - the advantage of time.


I modified the old pump housing. Carefully removed the pump, and replaced it with an 8mm pickup tube (scavenged from the TDI in-tank unit). The bottom of the old pump had an OD plastic base of 19mm, which held the strainer and parts in place at the base of the swirl pot. Fabricated a piece of stainless to hold the 8mm tube and to fit inside the strainer. Removed the power wires for the pump. Everything else remained in its original place. (Photo).
The return line to the tank was untouched. Feed line - I cut off the rigid metal/rubber hose and clamps leaving the compression ends and barbs to connect to the in-tank unit and filter - used an 8mm hose to a brass fitting which stepped up to the 12mm inlet to the pre-filter - connected to the 12mm pump inlet. Pump outlet is 8mm, same as the exposed barb on the Mahle filter, which remains in its original spot on the chassis rail.
The new pump and filter are mounted in-line on the inner chassis rail, between the fuel tank and opposite side of the rail from the filter. (photo). Pump is mounted on a 52mm bracket which has an isolating rubber liner, also from EFI. Pump is powered from the same source as the original, but with new soldered and insulated wires. I will be installing a dirt cover over the pump, as its directly in line with the uni joint and is sure to get spattered after a service.


Long Ranger Tank install. I had already cleaned and painted the tank after its years on the D1, so it was ready to go. Its a weighty lump and I was working alone. I lowered the hoist and strapped the tank underneath in the approximate place, then when high enough transferred the load to the transmission jack, for which I had made a new flat top just for this. With that secure I could work slowly and carefully to get things connected and in place. (photo, sideways sorry, can't rotate it). The tank is a good fit despite the couple of slight body differences from the D1 - its very tight, but no rubbing anywhere. Its a well engineered use of space. I did have to modify/trim the front legs of the ARB bar or they would have dug into the sides of the tank. I don’t run the rear stabiliser in favour of airbags, so no issues with that component which apparently has to be reversed somehow if retained.


With everything checked and connected except the feed line to the original filter, I replaced the 25 litres I drained from the old tank. Connected a vacuum pump (a compressor run tool which is so handy, a worthy ebay tool) and pulled fuel through the line to make sure the new pump was primed. Re connected the feed line to filter. With the ignition key on accessory position, I could hear the new pump run and click off with the relay once pressure was reached. After that the engine fired right up and ran without any coughing or issues.


Breathing. I connected the new tank directly to the original breather line - within a few km’s I checked everything out to find the tank was forming a vacuum - the filler sucked in air when undone. Ran it again and checked the breather line, sucked in air when removed. At least the tank and cap are air tight. From there I pored through the old books to find the diagram on emissions, with the supplemental for 1991, which is so different to previous models, and to some extent the Discovery 1, which uses a different tank again, plastic in that case. Anyway, removed the metal breather unit, located in the right wing but accessible from the inside in the load area. 4 x PH3 screws and a lot of careful pulling and clip disconnections underneath it was out. I found that the valve in the atmosphere (open) tube was completely blocked in both directions. Probably the reason for the fail in the original tank. Also oddly, though operational, the valve connecting the breather can to the charcoal canister seemed to be installed backwards from the factory - it would only allow vapour towards the breather can, not towards the charcoal canister. I installed an 8mm inline fuel filter at the end of the atmosphere line to keep things clean inside, no replacement valve yet but at least the tank can breathe.


With those issues resolved, I have been running the car with progressively more fuel in the tank - to check for leaks as well as seeing where 50 extra litres takes the gauge to. Its been running very well for a couple of weeks now, starts better than before. Looking forward to testing the extra range soon.

A further update on this work, for those that that may view this for a later install.

After a couple of weeks of using the new tank and external pump setup, I noticed the fuel pump was getting noisy as the tank approached between 1/4 and 1/3 remaining. This noise promptly went away with the addition of fuel.

So, one of two things assuming it was trying to pump air when there was heaps of fuel in the tank, either the pickup tube had fractured or the inlet of the swirl pot was blocked or disfunctional.

So, after purchasing a new in-tank 250mm x 8mm convoluted tube from EFI solutions, I pulled the tank again. The pickup tube looked OK (intact but brown) but changed it anyway - then did a test in water with the swirl pot to see how quickly liquid actually came into it from the tiny valve below the return hose inlet, it was dead slow, filled at only a trickle. Might have been enough for the old pump, but not for the newer slightly quicker pump (3 bar as opposed to 2.3).
To fix the slow ingress of fuel - I'd really have liked to have made or incorporated a big ball bearing valve like they have in aftermarket swirl pots, but all I had time for is to drill a series of smallish holes in the front (vehicle front) of the pot, right at the bottom - two 5mm and two 4mm - didn't feel right to make one big massive hole. Then did the immersion test again and liquid came in much quicker, kind of real time as it was immersed into the bowl as opposed to before when it stayed emptyish for a time. I drilled into the front as I figured the pot would still retain some fuel under acceleration (and we are talking about the last of the fuel, not a full tank). Also, the Long Ranger tank has really good baffles incorporated to prevent a lot of fuel slosh. Anyway, after a month or so, and have run it down to less than 1/4, there has been no noise from the pump and haven't had any kind of fuel starvation issues.

Yes I did purchase a replacement pump in case I screwed his one up, but the original is still OK.

I assume you must have an efi motor.

I had another 2 door with a long range tank that just used a cheapo external cube pump to fuel a carbie 4.4.

The pump failed so I just bought another one for $50 which worked fine.

DL

Originally Posted by 350RRC

I assume you must have an efi motor.

I had another 2 door with a long range tank that just used a cheapo external cube pump to fuel a carbie 4.4.

The pump failed so I just bought another one for $50 which worked fine.

DL

Yes, 3.9. Requirements are quite different.
A cube/facet pump is what I used previously when this tank was installed in a D1 TDI. As you say when the pump is external its easy and quick to replace, hence my desire to explore this rabbit hole of fuel requirements.