Quote:
Originally Posted by teagueAMX
I agree.
I speculated a couple of months ago (looking for thread now) that the HPFP is failing as a result of loss of pressure from the in-tank low pressure fuel pump (LPFP).
I believe BMWs split fuel tank contributes somewhat to this problem. When a car has a full tank of gas and goes around turns, goes up and down hills, stops and starts it’s not much of a problem. But as the tank level gets low the fuel “sloshes” from front to back and side to side – away from the in-tank pump fuel pickup. Getting down to a quarter tank or less really makes this worse. The engineers came up with a solution and that is to put the in-tank pump in a slosh cylinder or canister that’s open on the bottom to receive fuel, but as the tank tilts with movement some of the fuel is captured within the cylinder to prevent the pump from going dry.
This is a good design, but BMW has defeated this somewhat but using spit fuel tanks. The advantage is a more compact car design, lower center of gravity, etc. However, a low fuel problem is compounded somewhat by splitting the supply between the two sections of tank. So when you get down to a quarter tank of gas, you’re really only have one-eighth in each tank. They sort of solved it by using a remote pickup, but. . .
A low fuel supply is the primary reason in-tank electric fuel pumps fail and that’s due to a lack of fuel for lubrication, damaging the internal bearings and they overheat. You can imagine that over the lifetime of a car, in-tank full pumps experience hundreds if not thousands of low fuel situations but they are built tough to withstand it for 70- 80 thousands of miles before they fail.
Now let’s fast forward and talk about the Z4’s HPFP. We know the LPFP sends a pressurized flow of fuel to the HPFP, which in turn further boosts that pressure (as HerrK discovered) to a monstrous 2900 PSI for the direct injection system. To say the HPFP is entirely dependent on the LPFP is an understatement. Not only does it impact the fuel supply to the injectors, but IMO and more importantly, has a major impact to the life of the HPFP.
Finally seeing the HPFP guts exposed shows IMO that it’s severely overdesigned, and I agree with HerrK that is could be very “sensitive”. IMO a low fuel situation, results in LPFP cavitation and surging within the fuel line prior to the HPFP. This results in shock waves in the fuel supply and starves the HPFP of much needed lubrication and the pump overheats. The little piston cylinder walls are scored, seals are damaged and the unit no longer meet spec to enable it to produce 2900 PSI. Whalla – long cranks and limp mode.
|
This sounds like the most likely explination. I have seen standard in tank pumps fail very quickly (matter of seconds) when run dry. They rely 100% on the fuel for both cooling and lubrication. Inside the pump are radially mount blades/fins that spin within a cylinder. It is capped on each end by plates that also come in contact with the fins. Any scores of groves in those plates result in the loss of pressure as well as volume. In racing applications we worked around this by creating a sump and the bottom of the fuel cell that contained the pump. The sump and a trap door that allowed fuel but it could not splash out during cornering, accel/braking/etc. Watching a race when you see cars low on fuel weaving back and forth they are attempting to get the last bit of fuel into that sump.
To prove the theroy with the BMW HPFP we would have to monitor the fuel pressure from the tank to the HPFP. Observe if that pressure changes as the tank is drawn down. My bet is below 1/4 tank the pressure will begin to fluctuate as the fuel sloshes in the tank. This may aslo show as a lean condition in the cylinders as the HPFP begins to fail.
If the oil temp was truly the issue we would also see a large number of turbo failures and possibly engine failures.